Defined Terms and Documents

Study on interest rate restrictions in the EU

Final Report

Project No. ETD/2009/IM/H3/87

Submitted by

Prof. Dr. Udo Reifner, Sebastien Clerc-Renaud, RA Michael Knobloch

Institut für Finanzdienstleistungen e.V. (iff)

In association with

Zentrum für Europäische Wirtschaftsforschung GmbH (ZEW, Mannheim)

Institut für Finanzdienstleistungen e.V.

Rödingsmarkt 31/33

20459 Hamburg

Tel. +49 40 30 96 91 0

Fax +49 40 30 96 91 22

Project Team1

Coordination and contact person:

Sebastien Clerc-Renaud, iff

Legal Part:

Prof. Dr. Udo Reifner, iff

RA Michael Knobloch, iff

Economic Part:

Dr. Michael Schröder, ZEW

Christian D. Dick, ZEW

Lena Jaroszek, ZEW

Dr. Svitlana Voronkova, ZEW

Zwetelina Iliewa, ZEW

Cite study as:

iff/ZEW (2010): Study on interest rate restrictions in the EU, Final Report for the EU

Commission DG Internal Market and Services, Project No. ETD/2009/IM/H3/87,

Brussels/Hamburg/Mannheim.

Disclaimer: The views expressed in this study are the views of the contractors and

subcontractors, and not of the European Commission.

1 Stephanie Erk, Anna-Lena Huthmacher, Clemens Bonner and Georg Siegert have provided research

assistance for ZEW. Prof. Iain Ramsay (University of Kent), Damon Gibbons (Centre for Responsible Credit),

Dr. Achim Tiffe, Laura Flach and Kerim Al-Umaray have contributed to this report for iff.

iff/ZEW – Final Report on interest rate restrictions in the EU III

Executive summary

This study provides the European Commission with a comprehensive inventory of the

types of interest rate restrictions that exist in the EU Member States (Part 1) and gives

an assessment of the impact of these on both credit markets and people (Part 2).

Part 1 offers an inventory of interest rate restrictions in the EU Member States

and details the mechanisms and levels at which interest rate ceilings are set in

those countries with such a regulatory structure in place.

Part 2 discusses legal IRR as interventions in the market and its effects on

competition and on social and economic welfare.

Both parts relate to each other through their common definition of Interest Rate

Restrictions, (referred to as IRR throughout this study), and the purpose of IRR which is

to ensure that consumer credit markets function well and that they promote the social

welfare of people by means of appropriate and adequately priced credit products.

The report indicates that there is considerable variation in the attitude of EU Member

States towards the regulation of consumer credit prices. In addition, even where there is

a desire to regulate prices, Member States vary considerably in the extent to which

they seek to achieve this and the methods that they adopt:

In some Member States strict interest rate caps are defended because credit at a

high price may increase consumer insolvency and reflects the mal functioning

of markets especially for small amounts of credit.

In others, the absence of such regulations is justified primarily on the basis that

caps would reduce access to credit, especially for people with moderate means.

The report does not provide a one-dimensional answer to these questions.

The concept of usury is one uniform underlying theme. The patterns of existing

IRR are all derived from this historical principle. Interest derived from credit has

been morally rejected and even criminalised where it amounts to the exploitation

of personal weakness, but has got different legal forms according to the

differences in the development of consumer credit markets.

Public control of credit and the use of credit by consumers, as well as the general

attitude to consumer credit, remain diverse and do not lend themselves to simple

assumptions and solutions.

The report therefore provides information as to -
*    the regulatory choices,
*    the role of legal harmonisation in EU consumer credit law; and
*    information as to the factors which may favour one or the other solution.

In addition to written material in the form of case law, legislation and legal literature,

economic research and statistical data, both parts of the report rely on responses to

three questionnaires containing open and closed questions received from a legal

expert in each of the 27 Member States, as well as responses from 20 individual

providers, from 34 provider associations, from 47 consumer organisations, from 44

public authorities in all Member States and from 12 other institutions.

IV iff/ZEW – Final Report on interest rate restrictions in the EU

Legal part

The legal part of the report provides a consistent inventory of existing IRR in EU Member

States and shows how they are designed and function within the different legal systems.

In its opening theoretical section, the legal part of this study defines what amounts to

IRR in the context of the various regulatory systems and legal frameworks and cultures.

The term ‘Interest Rate Restrictions’ refers to all legal rules that limit the price

of credit contracts. In other words, the study is concerned with existing laws and

legal rules and their application in limiting charges imposed by lenders, directly or

indirectly, for the use of capital by borrowers.

As far as the terms used to describe credit-related phenomena are concerned, the

report uses the language of the EU Consumer Credit Directive (Directive

2008/48/EC, referred to as ‘CCD 2008’ throughout this study)

But we do not use the CCD 2008’s product-specific restrictions to define its

scope.

Three precautions have been taken in arriving at an analysis of the implications of

such law to this economically defined phenomenon

Legal rules are only one factor in determining access and pricing in the market.

The existence of legal rules as such does not reveal how effective these rules are.

The same legal concept may be used quite differently in different legal cultures

which have remained separate for centuries.

This has led to the definitions set out in Figure 1 of this report and in the

questionnaires. The types of credit have been broken down into: general-purpose

credit (instalment, revolving, small secured, micro) and mortgage credit. Interest rate

restrictions (IRR) have been distinguished into direct IRR on the rate level (contractual,

default), and indirect IRR on the methods of calculation (APRC, compounding,

variability), other cost elements (contractual charges, default charges), and on other

credit parameters (instalments, life-time, amount of credit).

While direct IRR limit the contractual interest rate or the amount of interest that can be

charged indirect IRR have a restrictive impact on the cost of credit. With respect to

indirect IRR, all Member States provide for rules which contain some restrictions on the

pricing of credit.

With direct IRR there is a big difference between the historical concept of usury and

some forms of interest rate ceilings.

All Member States subscribe to the principle of “good morals” or “fairness”,

which explicitly forbids usury, under criminal as well as private law, or implicitly

incriminates the intentional exploitation of the weakness of another person at an

individual level through extortionate pricing, especially in relation to credit.

With the exception of two Member States (Ireland and Romania), all have IRR in

relation to default interest.

14 Member States had either some form of an absolute ceiling (Greece,

Ireland, Malta) or a relative ceiling based on a reference rate (Belgium, Estonia,

iff/ZEW – Final Report on interest rate restrictions in the EU V

France, Germany, Italy, the Netherlands, Poland, Portugal, Slovakia, Spain,

Slovenia).

The concept of “usury”, “extortionate pricing” in credit or “unfair credit” is mostly

linked to the interest rate charged and to exploitation of the borrower. In some Member

States it may be used more indirectly in the context of criminal lending (Italy, Malta,

Estonia, Denmark), anatocism (Romania. Luxembourg) or it may simply be applied to

high-priced loans (Portugal, France, Belgium, Spain, Slovenia, the Czech Republic,

Slovakia, Hungary, Ireland, the UK and in German case law).

The body of law governing IRR still covers rules inherited from past centuries, which

may have been modernised or adapted to modern usage, or which may just remain

dormant. Such rules reflect the traditional ban on interest, which operated from

ancient times until the 19th century, as well as rules derived from the end of the 19th

century, when interest was regulated to protect agrarian interests against money

interest, rules on illegal lending practices, and modern market-driven rules intended to

prevent over-indebtedness and provide consumer protection.

While modern interest rate ceilings are typically imposed administratively, courts in

Germany have transformed the ancient subjective principle of good morals into a modern

objective interest rate ceiling, a process that would in principle be open to Member States

with no interest rate ceiling; some initial forms of this may also be identified in Estonia,

Spain and Sweden.

From the perspective of the contractual interest rate itself there are three

countries with an absolute ceiling in the tradition of usury, and this does not

seem to have had much impact on the economy (Greece, Ireland, and Malta).

Countries which use relative interest rate ceilings based on an average market

rate, multiplied by a quota such as that applied in France of one-third, or based

on a money market rate multiplied by four, as in Poland, have developed fairly

new systems with a high degree of effectiveness (Belgium, Estonia, France,

Germany, Italy, the Netherlands, Poland, Portugal, Slovakia, Spain, Slovenia).

The spread of interest rate ceilings is quite high.

It may range from as high as 453% pa for a small loan in Slovenia to a cap of

13.2% pa for a long-term loan there, while in France the spread between ceilings

for the different forms of credit was between 5.72% pa and 21.63% pa as at

March 2010.

Some countries provide such ceilings only exceptionally: for example in Spain

where they apply only to overdraft credit and protected housing loans; in Ireland

where they are confined to credit unions and moneylenders; in Greece to nonbanks;

in the Netherlands where mortgages are excluded; and in Malta where

further exemptions apply.

Countries which use relative interest rate ceilings have developed classes of

credits defined mainly by credit type as a basis for fixing the reference rate,

which may be derived from national markets or from the EU (Belgium).

With regard to default interest rates many countries provide statutory rates which

apply where no other rate has been agreed.

Austria, Belgium, Denmark, France, Finland, Greece, Hungary and the UK use the

contractual interest rate as a maximum.

VI iff/ZEW – Final Report on interest rate restrictions in the EU

Bulgaria, the Czech Republic, Denmark, Estonia, Finland, Germany, Italy,

Luxembourg, Poland, Portugal, Slovakia, Spain (overdrafts only) and Slovenia

provide a statutory external ceiling.

With regard to enforcement, many systems apply.

At the level of supervision, the central bank may have responsibility (Italy,

Portugal), with lending restricted to institutions licensed by the bank, by a

licensing authority (such as a Ministry as in the case of Belgium or a specialist

agency in the UK, France, Netherlands, Estonia, Germany), by a consumer

protection authority (Ireland, Bulgaria, Latvia), by a Market Inspectorate

(Slovenia), by a Consumer Ombudsman or by a Financial Authority as in Finland.

Civil law sanctions include the reduction of the interest to either the principal or

a permitted rate of interest or the nullity of the contract with the possibility of

judicial allocation of the obligations under the contract.

Other forms are criminal sanctions or the loss of a licence.

The effectiveness of direct IRR is related to whether legal rules manage to be selfexecuting

since official enforcement mechanisms are costly and only able to cover a

minimum of cases directly.

Law on paper (“in the books”) only creates law in practice (“in action”) where it

serves as a guideline for individual orientation, as a threat where breach leads

to ‘naming, blaming and shaming’ in the market or where it serves as an effective

barrier to entry into the market.

o Private law rules are closer to the consciousness and morals of people but

require private investment to enable access to justice.

o Administrative rules have no enforcement cost for consumers but

depend of the existence of sufficient and efficient administrative power.

o Criminal sanctions are often an exaggerated barrier and are difficult to

apply to the morally indifferent behaviour of profit maximisation, which

may favour a split between the mainstream and factually unregulated

shadow markets.

For IRR in consumer affairs it is important that the rules are clear, discriminate

well between right and wrong, do not depend on individualised administrative or

court decisions case by case and are easy to generalise and apply.

With regard to these criteria, the survey found significant differences in assumed

effectiveness according to the legal form of the regulation in question.

General principles of good morals and good faith in private law are close to

individual consciousness but produce cases with little effect on the general level of

interest rates in the market.

Administrative rules of prudential regulations in bank supervision and access, if

not combined with private law sanctions and the involvement of specialist

consumer agencies, effectively exclude outsiders and illegal lenders but have less

impact on mainstream providers.

Well placed seem to be strict and morally neutral interest rate ceilings set

administratively under private law. In this regard, Germany is an exception,

iff/ZEW – Final Report on interest rate restrictions in the EU VII

where such a system has been developed purely on the basis of the general

principle of good morals under private law. But such systems are particularly

exposed to circumvention through purely nominal pricing and ancillary services.

An assessment of all responses from all stakeholders showed quite strong agreement

even between providers and consumer organisations, as to the overall effectiveness of

their national IRR in regulating credit prices on the market. The following classification

reflects only a general assessment and does not claim to provide facts or representative

opinions, merely indications. It should be noted, especially where low effectiveness is

assumed, that it included responses from Member States with no significant IRR.

Direct IRR are seen as follows (average grade from 1 to 5 in parentheses).

(1) France, Belgium, Portugal: very effective (4.5)

(2) Cyprus, Denmark, Finland, Italy, Netherlands: effective (4.0)

(3) Poland, Austria, Bulgaria, Czech Republic, Estonia, Latvia, Lithuania, Slovakia,

Slovenia, Sweden, Romania: indifferent (3.0)

(4) Spain, Germany, Greece, Hungary, Luxembourg, Malta, UK: less effective (2.1)

(5) Ireland: not effective (1.0)

The results coincide roughly with the general assumption that strict interest rate ceilings

are the most effective, especially if they have been in place for a long time, whereas

general principles alone appear to give rise to concerns as to effectiveness.

IRR as a form of price regulation must use a comprehensive and effective interest

rate which effectively covers all cost elements, calculated objectively, which are imposed

on the consumer in relation to a credit agreement.

Historical IRR such as anatocism or absolute and even some relative legal interest

rates on default still refer to nominal interest rates. They do not cover

additional cost elements or disadvantageous rules on compounding and

calculation. They are open to circumvention and thus additional rules are needed

to fill these loopholes.

Modern IRR rate ceilings refer to the already harmonised price disclosure rules for

the APRC in Annex 1 of CCD 2008. This solves a number of these problems but

also import others.

o The method of calculation and the definition of compounding periods are

restricted to the use of the actuarial method.

o Fees charged separately from interest within the same contract or for

brokerage must be included.

o Rent agreements (financial leasing) or deferred payments (hire or

instalment purchase), in which time prices are applied that are not defined

as interest in the conventional sense, are also covered.

o While disclosure rules regulate the APRC with regard to competitors, rate

ceilings regard the factual burden a borrower is able to carry. This is why

the focus on voluntary inclusion of additional cost elements may be good

for price disclosure while leading to circumvention in usury law.

Some exceptions in price disclosure law to certain forms and amounts of credit are not

applicable to IRR and even undermine their effectiveness where they are most needed.

Nine Member States therefore have different regimes while four Member States still apply

disclosure exemptions to IRR and seven will adapt disclosure rules to the needs of IRR

legislation after implementation of CCD 2008.

VIII iff/ZEW – Final Report on interest rate restrictions in the EU

Small amounts of credit are expressly covered by IRR in Belgium, Estonia,

France, Germany, the Netherlands (recent inclusion).

Revolving credit is addressed specifically in Spain and covered without

exemption by IRR in all Member States.

Short-term loans are included but were allocated significantly higher interest rate

ceilings where IRR is specific to certain products, as in France. The ceiling in

Slovenia is extremely high.

As far as the inclusion of cost elements from ancillary services is concerned, the

debate about IRR resembles the debate about the APRC and whether such services

provide for a special advantage in addition to the loan and whether they have been

contracted voluntarily. These questions are the subject of controversy in all Member

States with IRR, which remains unresolved.

Payment protection insurance is sold with credit on a large scale, especially in

the UK, France, the Netherlands and Germany, and significantly increases the

level of payments due from the consumer. Experts and consumer advocates claim

that hidden kick-back provisions and the financing of the premiums through the

banks resemble charges which are already covered by the definition of the APRC.

Combined endowment credit which diverts loan repayments to an investment

product (capital life insurance, construction savings plans) with lower interest

earned than through direct repayments reach higher prices if both products are

seen as one.

Fees for cash withdrawal of small amounts of credit card credit can increase

the burden of payments significantly.

A number of Member States have indirect IRR which aim to address the additional cost

separately.

The traditional principle of anatocism is still applied, especially to default interest

but with decreasing effect.

Belgium has developed a special regulation of reference rates for variable credit.

Poland has a general IRR for fees. Special provisions which allow the reduction

of fees are in force in Belgium, Germany, the Czech Republic, Estonia and Malta.

Some Member States, such as France, Finland and Italy, allow judges to waive

default interest or even reduce the principal in case of default.

The CCD 2008 has had an impact on IRR in practice. Implementation had been

completed by only five Member States by March 2010, but in another 12 Member States

a draft was imminent.

Only a few countries (Portugal, Netherlands, and France) have used the

implementation expressly to introduce or change IRR.

In many countries, implementation had a strong indirect effect on the selection

of financial products for the application of IRR since most tend to harmonise

interest disclosure rules with rules restricting interest.

iff/ZEW – Final Report on interest rate restrictions in the EU IX

The idea of “(ir-)responsible lending” developed in the 2002 draft of the Directive and

repeated in the recitals of the final version in 2008 requires certain restrictions with

regard to over-indebtedness and thus comes close to being a form of IRR.

The UK and Finland have incorporated this principle into their rules.

Some providers and provider associations use it to describe their overall

behaviour in consumer credit.

Some Member States have used its impact on assessing the creditworthiness of

consumers, requiring or recommending certain loan-to-income ratios which have

an indirect effect on interest rates through risk-based pricing.

The common principles of fairness in competition, standard contract terms and

consumer information law, already harmonised through the respective EU Directives, do

not cover the pricing of credit. Its underlying legal concepts of transparency and fairness

are more procedural than substantive. They regard the way products are marketed and

serviced while the questions of IRR, especially usury and high prices, are left to the

fundamental principles of national regulation” where in civil law “good morals” are

applied. Art. II-7:301 of the Draft Common Frame of Reference define such principles in

future EU-contract law as “principles recognised as fundamental in the law of the Member

States”. But the common reference point, assumed as a market with functioning

competition, has led to the application of the fairness principle with regard to IRR in

some cases as well. The European Court of Justice has just ruled that a Spanish law

could also extend the EU concept of unfairness to IRR (EJC Dec. of June 3, 2010 C-

484/08).

Rules concerning the assessment of the debtor’s ability to pay are seen as part of

responsible behaviour in fair competition in the Czech Republic, Estonia and

Ireland.

The UK and Finland’s approach to responsible lending includes fundamental

principles as well as fairness principles. Estonia and Ireland explicitly describe high

cost of credit as potentially “unfair” in their legislation. Also Germany used

standard contract term law to void clauses on interest compounding.

Economic part

The economic part of the report aims at explaining the economic, social and financial

consequences of interest rate restrictions. Due to the complexity of market phenomena

and because of a lack of comparable datasets across countries, it is not possible to

identify a set of unambiguous effects. However, the report examines a number of

hypotheses concerning the impacts of IRR and presents the available data in relation to

these, as well as provides a theoretical framework for understanding the effects of IRR.

Within the testing of the hypotheses the economic part of the report refers only to direct

IRR in the form of interest rate ceilings. Furthermore, because IRR in 5 of the 14 Member

States with ceilings have to be considered insignificant mainly due to their scope of

application (Estonia Greece, Ireland, Malta, and Spain) analysis has centred on the group

of countries with significant IRR in place - a group which is therefore comprised of 9

Member States (Belgium, France, Germany, Italy, the Netherlands, Poland, Portugal,

Slovakia and Slovenia).

The theoretical discussion details the reasons for differential interest rates as well as

the effect of interest rate restrictions on capital allocation.

X iff/ZEW – Final Report on interest rate restrictions in the EU

The level of interest rates charged on consumer credit depends on the market

level of interest rates, the bank’s margin and a component which

compensates the lender for the risk of a borrower’s default, which in turn

depends on the collateral, the credit history and the income/wealth situation of

the borrower. Due to the fixed costs to each loan, small amounts of credit may be

relatively expensive. As the risk of low-income borrowers is perceived to be high,

lenders charge these customers higher interest rates.

Legal interest rate restrictions reduce the lender’s opportunity to charge riskadjusted

costs. Obviously, and depending on the level at which the restriction is

set, this decreases his willingness to lend. As a consequence, high-risk

borrowers may be denied credit access in the presence of legal interest

rate restrictions.

To discuss the implications of these supply-side mechanisms, the theoretical discussion

also focuses on consumer credit, taking into account both neoclassical and behavioural

views of consumer choices.

Acknowledging that consumers make choices according to their own preferences,

we demonstrate in a classical framework how a decision to finance consumption

by credit arises from the preference to smooth consumption over time. This

makes consumer credit different from other types of credit (eg. for investment

purposes), which are taken out for a project to earn a positive (monetary) return.

The neoclassical view postulates that, when households decide about their

consumption, saving and borrowing, they not only consider their current income

but also their expectations about their future income. High interest rates are

incorporated into the consumers’ choice of their optimal consumption behaviour.

This neoclassical view has an important implication for credit demand: no

borrower ever borrows more money than he or she can reasonably expect

to pay back in future periods. Credit demand is thus not simply the equivalent of

a shortage of funds. While it is possible to compensate a temporary negative

income shock by means of credit, the neoclassical model does not provide a

rationale for credit as a means to “make ends meet” for deprived households who

do not have a realistic perspective of an improvement of their situation.

Behavioural biases (such as wishful thinking or underestimation of exponential

growth) may lead to systematic (and predictable) deviations from rational

behaviour which lead to suboptimal consumer decisions. These biases challenge

the view that borrowing decisions are an unambiguously rational

consumption optimisation.

As a consequence of wishful thinking, consumers’ beliefs about event risk may be

distorted: they may be willing to borrow money even when it is rather unrealistic

that their future income will be sufficient to repay the debt. Consumers may also

underestimate the true cost of borrowing. This might occur because they hold

erroneous beliefs about the actual time period during which they will use the

credit. It might also stem from conceptual difficulties to understand the effect of

interest rates compounding over longer horizons (underestimation of exponential

growth). In these cases, consumers’ credit decisions are not necessarily

optimal, and may ultimately be an important driver of over-indebtedness.

If consumers’ credit decisions are optimal, it is advisable to remove barriers to

credit access. If they are not necessarily optimal, one solution would be to accept

barriers to credit access for selected groups of the population. As interest rate

restrictions theoretically reduce credit access, one can therefore either oppose

interest rate restrictions or endorse them, according to the extent of rationality

iff/ZEW – Final Report on interest rate restrictions in the EU XI

one assumes. However, there is a trade-off between reducing credit access

for irrational or uninformed consumers (which is beneficial, as these are

protected from becoming over-indebted) and excluding consumers who are

able to make appropriate credit decisions (which is negative as it reduces

their options to choose from). Whether subject to strong cognitive biases or not,

consumers with debts will nevertheless be subject to external factors that can put

pressure on their ability to service their debts to maturity.

In the light of this trade-off, it is important to evaluate whether the mechanisms

postulated by theory are observed in reality and to what extent their effects are

economically important (and not just minor compared to other influencing factors). To do

so, we first conduct a survey of existing studies on interest rate restrictions, which

reveals the following:

Numerous empirical studies have been devoted to determining the economic

impact of interest rate restrictions. However, most of these studies are on the

United States, not on European countries. This has an advantage for the

examination of the economic effect of interest rate restrictions: the studies

compare legally and economically relatively similar entities (ie. the U.S. states)

which may regulate interest rates in different ways. The observed differences

between the considered entities can then - with some caution - be identified as

the effect of interest rate restrictions. Due to the heterogeneity of EU Member

States, a comparable exact identification for the EU is not feasible.

However, it is a severe disadvantage for the purpose of this report that these

studies look at US interest rate caps, which are relatively low (eg. 12%). In

Europe, interest rate caps are frequently at higher rates. The results

obtained for low levels of interest rates cap (eg. reduced credit access) need not

necessarily be transferable to higher levels of interest rate caps.

Comparisons of different states in the United States of America with different

interest rate regulations typically suggest that tighter interest rate restrictions

lower credit access for low-income customers as well as total consumer

credit. Some studies also indicate that small amounts of credits are less often

available in the presence of IRR. However, they tend to deny a relationship

between interest rate restrictions and the interest rate level for average

(non-high-risk) customers.

Studies on payday loans, in particular in the USA, point at two further aspects:

credit is not per se beneficial (especially in the long run). Furthermore, when

thinking about effectively banning a financial product, it is crucial to take into

account evasion strategies of potential borrowers.

Unlike in the United States, any study which compares EU Member States will face

severe problems identifying the exact effects of interest rate restrictions, as

observations are also determined by a multitude of other economic and regulatory

factors. Existing policy-oriented reports on countries of the European Union

argue therefore either on theoretical grounds, or provide empirical facts

which cannot be unambiguously attributed to interest rate restrictions.

To demonstrate the heterogeneity of consumer credit markets across the Member States

of the European Union, and to provide a comprehensive picture of their importance, we

give an overview of the consumer credit markets in the 27 EU Member States.

We discuss the markets for total credit to households, housing credit, and consumer

credit (without housing) for all EU 27 countries. Where illustrative, we also provide

separate analyses for EU 25 or EU 15 countries or the New Member States.

XII iff/ZEW – Final Report on interest rate restrictions in the EU

More detailed information is also given for six countries, which serve as case studies in

this report: Germany, France, the Netherlands, Poland, Sweden and the UK. The reasons

behind this choice include diversity in terms of economic characteristics, financial

cultures, size and attributes of the consumer credit markets, as well as the heterogeneity

of the legal framework of interest rate restrictions in these countries. It is this

heterogeneity in the chosen countries which allows us to draw conclusions with regard to

the effect of interest rate regulation on consumer credit markets and over-indebtedness,

and to appreciate reasons for different outcomes in credit markets beyond interest rate

regulation. The case study countries can be summarised as follows:

The countries selected represent more than half (54%) of the population of the

EU 27. Furthermore, they accounted for nearly two-third of the volume of total

credit to households and consumer credit to households in EU 27 countries at the

end of 2008.

The six countries included in our study differ considerably in market size and

market structure: half of the selected case study countries – the UK, Germany

and France represent the countries with the largest national consumer credit

markets. The remaining case study countries included in the study - Poland,

Sweden and the Netherlands make comparatively small contributions to the

overall volume of credit to households in Europe. The former 3 countries are also

the most populous countries in Europe and are included in the survey alongside

smaller countries.

The selected countries have different consumer credit regulations and, in

particular, different levels and forms of IRR. Included in the study are Sweden

and the UK. Neither of these countries have interest rate restrictions. In contrast,

in Germany and France regulations of consumer credit interest rates have been in

effect for a long time and in Poland interest rate caps were only recently

introduced.

To guide the reader of this report through relevant issues regarding interest rate

restrictions, we consider a set of different hypotheses. For a discussion of these

hypotheses, we summarise existing evidence from the literature, and derive conclusions

from the collected data. In addition, we consider the responses of questionnaires

answered by particular stakeholders in the consumer credit market, such as public

authorities, consumer associations, providers and provider associations. The survey

enables us to learn from stakeholders’ experience with regard to the effect of interest

rate restrictions (or lack of thereof) on the credit markets in the respective countries. It

also allows a comparison of the answers from stakeholders in countries with IRR with the

answers of those from countries without IRR, and a qualitative discussion of the

differences. The inclusion of heterogeneous stakeholders (regulators, consumer

organisations, provider associations) will enable us to consider potentially divergent

perspectives about issues in question in a balanced way. Accordingly, we base our

judgement of the hypotheses on several sources: the interpretation of existing data,

previous empirical evidence, as well as the information from Stakeholder Questionnaires.

The first hypothesis (H1: IRR reduce credit access, in particular for low-income

borrowers) is generally found to be plausible:

High-risk borrowers requesting small-amount credit can only be served when a

certain threshold interest rate is exceeded. Hence, they may not be served credit

in the presence of IRR.

One needs to keep in mind that, due to the relatively high levels of interest rate

ceilings in most European countries, the scope of the interest rate restrictions is

not expected to be equivalent to the ones documented in the US a few decades

iff/ZEW – Final Report on interest rate restrictions in the EU XIII

ago. Rather, it is likely that access to mainstream credit (including overdrafts and

revolving credit) remains rather unaffected by IRR, while there may still be

missing credit options to low-income borrowers which are served in the high-cost

credit segment in some countries.

Note that the desirability of such credit access to this specific population group is

subject to political controversy.

In the light of our analysis, a second hypothesis (H2: IRR lead to a decline in the

volumes of consumer credit granted) appears unlikely to hold in an economically

significant way.

The relevant market segments of high-cost credit (which are most affected by

IRR) only constitute a relatively small fraction of the entire consumer credit

market in which they exist. A lack of these market segments is unlikely to lead to

an economically significant drop in the volume of credit markets, as the latter is

affected by a multitude of other factors, as well.

Economic activity is unlikely to be significantly supported by the presence of highcost

credit (ie. the absence of IRR).

Hypothesis H2a (Without IRR, more product types exist in the market) appears

plausible:

Countries without IRR tend to have a higher prevalence of personal loans/auto

loans than countries with IRR. It is also likely that interest rate restrictions

prevent the market entry of one or more forms of high-cost credit.

A relationship between the importance of certain credit types and interest rate

restrictions is unlikely to be purely mechanistic. Rather, lower levels of consumer

credit and the tougher regulation of credit (including IRR) may both be

consequences of a third factor - the country’s attitude towards credit.

With respect to H3 (IRR lead to credit from non-bank sources, such as paying bills late),

our results remain inconclusive.

Some argue that the existence of high-cost credit helps households to avoid

obtaining credit from (potentially expensive) non-bank sources, such as utility

providers. However, it does not appear that there are systematic differences in

lending from non-bank sources between countries with and without IRR.

A related claim in the context of interest rate restrictions is addressed in H4 (IRR lead to

a substantial illegal market in lending).

There is no convincing comprehensive data to evaluate this hypothesis. Due to

this lack of data, we evaluate the evidence on H4 to be inconclusive.

The related hypotheses H5 and H5a address the effect of interest rate restrictions on the

level of over-indebtedness and on its growth rate after an economic shock, respectively.2

We conclude that a direct influence of interest rate restrictions on the level of

over-indebtedness, as H5 suggests, is unlikely.

2 H5: The lack of IRR leads to a higher level of over-indebtedness; H5a: The lack of IRR has particularly

adverse effects on default rates/over-indebtedness in the presence of negative shocks (eg. recessions) to

the economy.

XIV iff/ZEW – Final Report on interest rate restrictions in the EU

With respect to H5a, we obtain inconclusive results: the most current figures do not

reveal a particularly pronounced increase of over-indebtedness in the aftermath of the

financial market crisis.

However, there are no official data covering the years 2009 and 2010, and the

views collected from stakeholders indicate that there might be a link as postulated

by H5a in those years. We therefore recommend re-evaluating this issue in the

next year when the relevant data are available.

With respect to the hypothesis H6 (The average consumer - or even more so: low-risk

consumer - would be granted cheaper credit in the presence of IRR), we find inconclusive

results.

Due to the lack of micro data on individual credit cost before and after the

introduction of IRR, we focus on average interest rates. Drawing on these data

and findings from the Stakeholder Questionnaire, there is no unambiguous

evidence that average rates are declining, as H6 postulates.

In contrast, it appears plausible that H7 holds (IRR lead to increased charges as

providers will try to compensate the reduced interest revenues by increased charges).

There are examples from different countries illustrating that providers react to

regulation by charging fees for which the regulation does not apply.

The evidence on H8 (IRR represent barriers to consumer credit market integration) is

inconclusive for two reasons.

Firstly, it appears that it is not the mere presence of an interest rate restriction,

but rather the variety of IRR (presence and non-presence) that are potentially an

impediment to market entry.

Secondly, the institutional setting is also diverse across countries along several

other dimensions, such that it is hard to assess how important interest rate

restrictions are in this environment.

Hypothesis H9 (IRR lead to lower levels of competition in the consumer credit industry)

is found to be unlikely to hold, as the level of competition differs (according to several

criteria) across countries regardless of the presence of IRR.

It has also been documented in other studies that there is a low level of

competition in high cost credit markets in the UK, a country without IRR in these

segments. This also deemphasises the role of interest rate restrictions in the

context of competition.

The last hypothesis H10 (IRR lead to a convergence of all consumer credit interest rates

at the level of the interest rate cap) implies that providers use an exogenously given

interest rate cap to coordinate their (non-competitive) price setting at a rate just below

the cap. Other studies have found some evidence on this issue. We demonstrate that the

results on this hypothesis are inconclusive:

Whether or not the phenomenon captured in H10 occurs crucially depends on the

market structure and the exact way in which interest rate restrictions are

implemented.

In summary, we find that there are less clear-cut implications of economic significance of

interest rate restrictions than it is sometimes argued. It is apparent that IRR do shape

the supply side of the consumer credit market in three respects:

iff/ZEW – Final Report on interest rate restrictions in the EU XV

Firstly, it is likely that the existence of interest rate restrictions excludes some

customer groups from credit access (which might or might not be an explicit

objective of the introduction of IRR).

Secondly, there might be a reduced set of credit types, as some credit types with

very high interest rates are not offered in the presence of interest rate

restrictions.

Thirdly, one needs to keep in mind that suppliers can (to some extent) structure

their credit product in a way so that the existing interest rate regulation does not

apply.

We also conclude that many observations on credit markets are not only driven by

regulatory conditions (affecting the supply side), but also by the actual behaviour of the

demand side:

As particularly explained in our discussion of H1 and H2a, it is more realistic to

assume that there are clear patterns in the attitude towards credit across

countries which may explain both the preferences for strict interest rate

regulations prevailing in some countries and, eg. reduced incidence of credit of

their consumers.

XVI iff/ZEW – Final Report on interest rate restrictions in the EU

Table of Contents

Executive summary ..........................................................................................III

Introduction ..................................................................................................... 25

1 Legal survey of interest rate restrictions...................................................... 28

1.1 Legal theoretical background ...................................................................... 28

1.1.1 Definition of interest rate restrictions (IRR) .......................................................28

1.1.2 Empirical survey ...........................................................................................35

1.1.3 General principles of IRR in national legislation ..................................................38

1.1.4 Effectiveness ................................................................................................57

1.2 Direct IRR................................................................................................ 61

1.2.1 Contractual interest rate ceilings .....................................................................61

1.2.2 Default interest rate ceilings ...........................................................................83

1.3 Other cost relevant factors ......................................................................... 91

1.3.1 Uniform definition of interest (rates) ................................................................92

1.3.2 Anatocism and compounding...........................................................................94

1.3.3 Variability of interest rates............................................................................ 100

1.3.4 Fees and charges ........................................................................................ 104

1.4 Effects of the CCD 2008 on IRR................................................................. 111

1.4.1 EU Directives and national IRR ...................................................................... 111

1.4.2 Implementation of CCD 2008 ........................................................................ 112

1.4.3 Small amounts of credit ............................................................................... 113

1.4.4 Short-term loans......................................................................................... 114

1.4.5 Responsible lending ..................................................................................... 115

1.5 Social impact of high-cost credit................................................................ 119

1.5.1 Credit products ........................................................................................... 119

1.5.2 Consumer welfare ....................................................................................... 124

1.5.3 Assumed effects on credit markets................................................................. 125

1.6 Stakeholder views on IRR ........................................................................ 125

1.6.1 Political and legal discussions........................................................................ 126

1.6.2 IRR and consumer over-indebtedness ............................................................ 128

1.2.1 Preferred characteristics of an interest rate ceiling ........................................... 132

1.6.3 Importance and adequacy of IRR regulation in the EU....................................... 134

1.6.4 IRR and other regulatory options ................................................................... 139

2 Economic Part............................................................................................. 142

2.1 Theoretical background............................................................................ 142

2.1.1 Outline ...................................................................................................... 142

2.1.2 The individual choice of credit ....................................................................... 144

2.1.3 Interest rate restrictions and capital allocation................................................. 151

2.1.4 Resulting Hypotheses................................................................................... 155

2.1.5 Credit to consumers vs. credit to businesses ................................................... 157

2.2 Existing studies on IRR ............................................................................ 158

2.2.1 Outline ...................................................................................................... 158

2.2.2 Natural experiments .................................................................................... 158

2.2.3 Comparison of different States within the USA................................................. 159

2.2.4 Payday loans: the discussion of the desirability of credit access.......................... 160

iff/ZEW – Final Report on interest rate restrictions in the EU XVII

2.2.5 Reports about IRR in EU Member States ......................................................... 161

2.2.6 Summary................................................................................................... 163

2.3 Credit market overview............................................................................ 166

2.3.1 Outline ...................................................................................................... 166

2.3.2 Data availability and definitions ..................................................................... 167

2.3.3 Overview of household credit in the EU 27 ...................................................... 168

2.3.4 Interest rates on credit to households in the EU............................................... 183

2.4 Case Studies .......................................................................................... 191

2.4.1 Introduction ............................................................................................... 191

2.4.2 Credit markets............................................................................................ 195

2.5 Discussion of the hypotheses.................................................................... 224

2.5.1 H1: IRR reduce credit access, in particular for low-income borrowers .................. 226

2.5.2 H2: IRR lead to a decline in the volumes of consumer credit granted .................. 238

2.5.3 H2a: Without IRR, more product types exist in the market ................................ 246

2.5.4 H3: IRR lead to credit from non-bank sources, such as paying bills late............... 263

2.5.5 H4: IRR lead to a substantial illegal market in lending ...................................... 269

2.5.6 H5: The lack of IRR leads to a higher level of over-indebtedness ........................ 274

2.5.7 H5a: The lack of IRR has particularly adverse effects on default rates/overindebtedness

in the presence of negative shocks (eg. recessions) to the economy 292

2.5.8 H6: The average consumer (or even more so: low-risk consumer) would be

granted cheaper credit in the presence of IRR ................................................. 302

2.5.9 H7: IRR lead to increased charges as providers will try to compensate the reduced

interest revenues by increased charges .......................................................... 313

2.5.10 H8: IRR represent barriers to consumer credit market integration ...................... 317

2.5.11 H9: IRR lead to lower levels of competition in the consumer credit industry ......... 319

2.5.12 H10: IRR lead to a convergence of all consumer credit interest rates at the level of

the interest rate cap .................................................................................... 323

2.5.13 Summary H1-H10 ....................................................................................... 325

3 What can the European Commission learn from this study? ....................... 328

Annex I: Bibliography..................................................................................... 330

Annex II: Mandate of the European Commission and objectives .................... 342

Annex III: Methodology of the research......................................................... 344

Annex IV: Stakeholders contacted.................................................................. 350

Annex V: Legal Experts................................................................................... 354

Annex VI: ECRI (2009) Dataset main definitions............................................ 355

Annex VII: Databases used in the analyses of over-indebtedness.................. 356

Annex VIII: Measures of over-indebtedness .................................................. 358

Annex IX: Incidence of short-term credit facilities ......................................... 363

Annex X: Stakeholder Questionnaire .............................................................. 364

Annex XI: Tables of stakeholder responses .................................................... 379

Annex XII: Provider Questionnaire................................................................. 398

Annex XIII: Provider Questionnaire - Methodology and feedback .................. 408

Annex XIV: Provider general feedback on IRR................................................ 410

Annex XV: Stakeholder general feedback on IRR ........................................... 412

Annex XVI: Observations on SME and microenterprise lending ...................... 418

Annex XVII: Quantification of the implications of H1 and H2a ....................... 429

XVIII iff/ZEW – Final Report on interest rate restrictions in the EU

List of Tables

Table 1: Definitions of key terms used in the study ................................................. 31

Table 2: Survey participation with the Stakeholder Questionnaire by stakeholder group 37

Table 3: Usury legislation in the EU....................................................................... 38

Table 4: Main reasons for introducing IRR.............................................................. 50

Table 5: Some reasons why IRR is still valid today .................................................. 52

Table 6: Member States by effectiveness of IRR ..................................................... 59

Table 7: Overview of interest rate ceilings in the EU................................................ 62

Table 8: Interest rate ceilings in the EU as of March 2010 ........................................ 63

Table 9: Interest rate ceilings in France................................................................. 67

Table 10: Interest rate ceilings in Belgium ............................................................. 69

Table 11: Interest rate ceilings in Slovakia............................................................. 70

Table 12: Interest rate ceilings in Italy .................................................................. 72

Table 13: Interest rate ceilings in Portugal............................................................. 74

Table 14: Interest rate ceilings in Slovenia............................................................. 76

Table 15: Statutory default interest rates in EU Member States................................. 84

Table 16: Default interest rate ceilings in EU Member States .................................... 85

Table 17: Overview of default interest rate regulation in EU Member States................ 89

Table 18: Borrowing rate and APRC ...................................................................... 94

Table 19: Overview of rules on anatocism in the EU ................................................ 96

Table 20: Restrictions on general fees and charges including PPI in the EU ................104

Table 21: Payment protection insurance in Germany, 2001–2008, EUR (size, claims,

claims ratio) .......................................................................................109

Table 22: Transposition of the CCD by March 2010 ................................................112

Table 23: Small credit and the CCD .....................................................................114

Table 24: Assumed problems with different credit forms/types/products ...................120

Table 25: Incidence of high-cost credit in EU Member States ...................................121

Table 26:Development of EU subprime lending markets..........................................122

Table 27: Legal and political discussions ...............................................................127

Table 28: Late payments per credit type in Belgium in 2009....................................130

Table 29: Specific types of credit, behaviour and providers in relation to overindebtedness

.....................................................................................130

Table 30: Member States facing a significantly worsening trend in over-indebtedness .132

Table 31: Adequacy of IRR regulation facing banks and non-banks...........................139

Table 32: Ranking of IRR as a policy measure alongside other alternatives ................140

Table 33: Overview of the hypotheses..................................................................163

Table 34: Countries with the highest volume of total credit to households, EUR billion.168

Table 35: Countries with the highest total credit to households per capita in EU 27;

1000 EUR..........................................................................................171

Table 36: Countries with the highest volume of housing loans; EUR billion.................173

Table 37: Countries with the highest housing loans per capita in EU 27; 1000 EUR .....174

Table 38: Countries with the highest volume of consumer credit; EUR billion .............176

Table 39: Countries with the highest consumer credit per capita in the EU 27; 1000

EUR..................................................................................................177

Table 40: Development of total credit to households as % of GDP in the EU 27; 2000-

2008 ................................................................................................182

Table 41: Lending to households: MFI interest rates (AAR/NDER) on new business .....184

iff/ZEW – Final Report on interest rate restrictions in the EU XIX

Table 42: Interest rates on lending for house purchase across the EU 27; new

business; December 2009; % ..............................................................185

Table 43: Interest rates consumer credit across the EU 27; new business; December

2009; % ...........................................................................................187

Table 44: Interest rates on overdrafts, credit cards and overrunning across the EU 27;

new business; December 2009; %........................................................189

Table 45: Market size and structure of the consumer credit market in selected

countries...........................................................................................192

Table 46: Indicators of demographic structure and financial behaviour in selected

countries...........................................................................................193

Table 47: Consumer credit market; Germany; 2009...............................................199

Table 48: Distribution of consumer credit; Germany, 2008......................................203

Table 49: Total debt burden of households; France ................................................206

Table 50: Distribution of consumer credit; France, 2008 .........................................207

Table 51: Cost structure of Flitskrediet .................................................................210

Table 52: Distribution of consumer credit; the Netherlands, 2008.............................210

Table 53: Distribution of consumer credit; Poland, 2008 .........................................214

Table 54: Distribution of consumer credit; Sweden, 2008........................................216

Table 55: Composition of consumer debt over time; UK..........................................219

Table 56: Distribution of consumer credit; UK, 2008...............................................223

Table 57: Evidence from the past – volume of credit (H2) .......................................241

Table 58: Prevalence of different product types across countries ..............................249

Table 59: Past experience with arrears on utility bills..............................................265

Table 60: Percentages of households in arrears (on mortgage or rent, utility bills or

hire purchase), 2008...........................................................................275

Table 61: Evidence from the past - over-indebtedness (H5) ....................................280

Table 62: Evidence from the past – average prices (H6) .........................................306

Table 63: Overview of measures determining over-indebtedness..............................359

Table 64: Projection of potential home (collected) lending market size......................430

Table 65: Projection of potential payday lending market size ...................................431

XX iff/ZEW – Final Report on interest rate restrictions in the EU

Table of Figures

Figure 1: Forms of consumer credit and Forms of IRR.............................................. 34

Figure 2: History of IRR in the EU Member States ................................................... 48

Figure 3: Preferences with regard to ceiling characteristics ......................................133

Figure 4: Stakeholder preferences for regulation (% who answered yes) ...................135

Figure 5: The benefits of borrowing and saving......................................................145

Figure 6: The idea of consumption smoothing........................................................146

Figure 7: The effect of lower interest rates............................................................146

Figure 8: Disadvantage of poor borrowers without credit access...............................147

Figure 9: Effect of IRR on high-risk (left) and low-risk (right) borrower .....................152

Figure 10: Credit markets under asymmetric information........................................154

Figure 11: Total credit to households in EU 27 countries, 2008; EUR billion ...............168

Figure 12: Total credit to households in EU 15 countries, average annual real growth

rates; in %.......................................................................................169

Figure 13: Total credit to households in New Member States, average annual real

growth rates; % ................................................................................170

Figure 14: Real growth rates of total credit to households in EU countries; 1998-2008;

%....................................................................................................170

Figure 15: Total credit to households per capita in EU 27 countries, 2008, EUR billion .171

Figure 16: Housing loans in EU 27 countries, 2008; EUR billion................................172

Figure 17: Housing loans in EU 27 countries per capita; 2008; 1000 EUR ..................173

Figure 18: Housing loans in EU15, average annual real growth rates; % ...................174

Figure 19: Total credit to households per capita in New Member States, average

annual growth rates; %......................................................................175

Figure 20: Outstanding consumer credit in EU 27 countries; EUR billion ....................175

Figure 21: Outstanding consumer credit in EU 27 countries; 2008; per capita ............176

Figure 22: Consumer credit to households per capita in EU15, average annual growth

rates; %...........................................................................................177

Figure 23: Consumer credit to households per capita in New Member States, average

annual growth rates; %......................................................................178

Figure 24: Development of total credit to households in % of GDP; EU 27 and USA;

2000-2008........................................................................................179

Figure 25: Development of total credit to households in % of GDP for different EU

country groups; 1995-2008.................................................................179

Figure 26: Total credit to households outstanding, % of GDP; 2005-2008..................180

Figure 27: Consumer credit in the EU 27 and the US as a % of Disposable Income; EU

and USA; %......................................................................................180

Figure 28: Consumer credit as a percentage of final consumption; Expenditure of

households; % ..................................................................................181

Figure 29: Distribution of total credit to households by MFIs in Euro area 16; EU 27

and US.............................................................................................181

Figure 30: Lending to households by credit type, % of GDP; EU 16; EU 27 and US .....182

Figure 31: Distribution of lending to households; 2008 ...........................................183

Figure 32: Heterogeneity among relevant economic factors .....................................194

Figure 33: Consumer credit per GDP in a cross-country comparison..........................195

Figure 34: Consumer credit outstanding; Germany, EUR billion................................196

Figure 35: Consumer Credit as % of GDP, Germany ...............................................196

Figure 36: Growth of consumer credit, Germany; % p.a. ........................................197

iff/ZEW – Final Report on interest rate restrictions in the EU XXI

Figure 37: Instalment credit; Germany; EUR billion................................................198

Figure 38: Consumer credit in Germany; Dynamics of instalment credit vs. other

credit; 2006-2009; new business; EUR billion ........................................200

Figure 39: Consumer credit: Dynamics of instalment credit; Germany; 2006-2009;

new business EUR billion.....................................................................201

Figure 40: Consumer credit: Dynamics of other credit; Germany; 2006-2009; new

business, EUR billion ..........................................................................201

Figure 41: Consumer credit outstanding; France; EUR billion ...................................204

Figure 42: Consumer credit as % of GDP; France...................................................204

Figure 43: Consumer credit growth rates; France; % .............................................205

Figure 44: Evolution of the use of consumer credit and bank overdraft; France; % of

total households ................................................................................205

Figure 45: Consumer credit by type; France; 1995-2008; EUR billion........................206

Figure 46: Consumer credit outstanding; Netherlands; EUR billion............................208

Figure 47: Consumer credit; the Netherlands; % of GDP.........................................208

Figure 48: Consumer credit growth; the Netherlands; % of GDP ..............................209

Figure 49: Consumer Credit Outstanding; Poland; EUR billion ..................................211

Figure 50: Consumer credit; Poland; % of GDP .....................................................212

Figure 51: Consumer credit growth; Poland; % p.a. ...............................................212

Figure 52: Types of consumer credit; Poland.........................................................213

Figure 53: Consumer credit outstanding; Sweden, EUR billion..................................215

Figure 54: Consumer credit; Sweden; % of GDP....................................................215

Figure 55: Consumer credit growth; Sweden; % p.a. .............................................216

Figure 56: Consumer credit outstanding; UK; EUR billion ........................................217

Figure 57: Consumer credit; UK; % of GDP...........................................................218

Figure 58: Consumer credit growth rates; UK, % change p.a...................................218

Figure 59: Consumer credit outstanding: Cards and overdrafts; UK; GBP million ........222

Figure 60: Perception of easy and unlimited credit access .......................................228

Figure 61: Households without access to short-term credit ......................................229

Figure 62: Households considering credit to be useful.............................................230

Figure 63: Desirability of Credit Access for low-income borrowers, by Stakeholder Type231

Figure 64: Desirability of Credit Access for low-income borrowers, by Type of Country

of Origin...........................................................................................231

Figure 65: Level of Credit Options to low-income borrowers ....................................232

Figure 66: Effects of IRR (2x average) on credit access, for low-income borrowers .....233

Figure 67: Effects of IRR (max. of 30%) on credit access, for low-income borrowers...234

Figure 68: Effects of IRR (2x average) on credit access, for average borrowers ..........234

Figure 69: Effectiveness of different measures to increase credit access for low-income

borrowers, views of provider associations..............................................236

Figure 70: Effectiveness of different measures to increase credit access for low-income

borrowers, views of consumer organisations ..........................................236

Figure 71: Consumer Credit in EUR billion, current prices........................................239

Figure 72: Trends in Consumer Credit, End-of-Period Stocks, 1995=100%................240

Figure 73: Consumer Credit in Poland, 1996-2008 .................................................242

Figure 74: Consumer Credit per Capita in Poland, 1996-2008, in EUR. ......................242

Figure 75: Consumer Credit in the Netherlands, 1995-2008 ....................................243

Figure 76: Consumer Credit per Capita in the Netherlands, 1995-2008; in EUR 000s ..244

XXII iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 77: Volume of Instalment Consumer Credit in Germany, different maturity

groups, in bn. Euros...........................................................................245

Figure 78: Distribution of loans in EU 27 Member States, 2008 ................................247

Figure 79: Prevalence of mortgages among EU-25 households, 2005........................247

Figure 80: Owner-occupation rates for the EU 27 Member States, 2008 ....................248

Figure 81: Market penetration for different consumer credit products, 2006...............251

Figure 82: Prevalence of personal loans across EU 25 Member States, 2005 ..............252

Figure 83: Prevalence of auto / vehicle loans across EU-25 Member States, 2005 .......252

Figure 84: Prevalence of personal loans across countries (0: not existent, 1:

uncommon - 5: widespread) ...............................................................253

Figure 85: Prevalence of auto/vehicle loans across countries (0: not existent, 1:

uncommon - 5: widespread) ...............................................................253

Figure 86: Prevalence of overdraft facilities across EU-25 Member States, 2005 .........254

Figure 87: Prevalence of overdraft facilities across countries (0: not existent, 1:

uncommon - 5: widespread) ...............................................................255

Figure 88: Prevalence of credit cards across EU-25 Member States...........................255

Figure 89: Prevalence of credit card credit across countries (0: not existent, 1:

uncommon - 5: widespread) ...............................................................256

Figure 90: Importance of alternative lending (higher cost small volume credit) as % of

total lending activities ........................................................................258

Figure 91: Impact of a relative/floating IRR (2x market average) on the product

variety available to low-income borrowers .............................................259

Figure 92: Impact of a relative/floating IRR (2x market average) on the product

variety available to an average borrower...............................................259

Figure 93: Effectiveness of different regulation measures with respect to wider product

variety (responses from provider associations).......................................261

Figure 94: Effectiveness of different regulation measures with respect to wider product

variety (responses from consumer organisations)...................................262

Figure 95: Cross-country comparison of households in arrears on utility bills (2008) ...264

Figure 96: Polish households in arrears on utility bills, development 2005-2008 .........266

Figure 97: Dutch households in arrears on utility bills, development 2005-2008 .........267

Figure 98: Importance of community/informal lending as % of total lending activities .268

Figure 99: Presence of illegal lending ...................................................................270

Figure 100: Impact of relative/floating IRR (2x market average) on illegal lending to

low-income borrowers.......................................................................271

Figure 101: Impact of relative/floating IRR (2x market average) on illegal lending to

low-income borrowers.......................................................................272

Figure 102: Impact of relative/floating IRR (2x market average) on illegal lending to

an average borrower ........................................................................273

Figure 103: Over-indebtedness in the EU in 2008 ..................................................276

Figure 104: Prevalence of arrears on hire purchases or loans, comparison of overall

population and low-income households in 2008 ....................................277

Figure 105: Households' ability to make ends meet in 2008.....................................278

Figure 106: Low-income households' ability to make ends meet in 2008 ...................279

Figure 107: Polish households in arrears, development for entire population 2005-

2008 ..............................................................................................280

Figure 108: Polish households in arrears, development for low-income households

2005-2008 ......................................................................................280

Figure 109: Development of over-indebtedness amongst Polish households 2005-

2008 ..............................................................................................281

iff/ZEW – Final Report on interest rate restrictions in the EU XXIII

Figure 110: Dutch households in arrears, development for entire population 2005-

2008 ..............................................................................................282

Figure 111: Dutch households in arrears, development for low-income households

2005-2008 ......................................................................................282

Figure 112: Development of the Dutch mortgage market and mortgage repayment

arrears 2005-2008 ...........................................................................283

Figure 113: Development of over-indebtedness amongst Dutch households 2005-2008284

Figure 114: Severity of over-indebtedness, by type of respondent............................285

Figure 115: Severity of over-indebtedness, by country of origin with respect to

interest rate regulation .....................................................................285

Figure 116: Severity of over-indebtedness, responses from consumer organisations ...286

Figure 117: Effectiveness of different measures to reduce over-indebtedness (view of

providers) .......................................................................................287

Figure 118: Effectiveness of different measures to reduce over-indebtedness (view of

consumer organisations) ...................................................................287

Figure 119: Effects of IRR (2x average) on over-indebtedness, for low-income

borrowers .......................................................................................289

Figure 120: Effects of IRR (max. of 30%) on over-indebtedness, for low-income

borrowers .......................................................................................290

Figure 121: Effects of IRR (2x average) on over-indebtedness, for average-income

borrowers .......................................................................................290

Figure 122: 6-Country comparison of households in arrears 2005-2008 ....................293

Figure 123: 6-Country comparison of low-income households in arrears 2005-2008....294

Figure 124: 6-Country comparison of subjective over-indebtedness 2005-2008..........295

Figure 125: 6-Country comparison of subjective over-indebtedness among low-income

households 2005-2008......................................................................295

Figure 126: Cross-country comparison of households falling behind on their mortgage

repayments 1994-2001.....................................................................296

Figure 127: Cross-country comparison of low-income households falling behind on

their mortgage repayments 1994-2001 ...............................................297

Figure 128: Trend in credit defaults by private households after the financial crisis .....298

Figure 129: Growth rates of PVI (Germany) 2004-2009 ..........................................300

Figure 130: Real interest rates on consumer credit with initial fixation of up to 1 year

in the six EU countries, 2003-2010, in percent. ....................................303

Figure 131: Real interest rates on consumer credit with an initial rate fixation of more

than 1 year and less then 5 years in the six EU countries, 2003-2010, in

percent...........................................................................................304

Figure 132: Real interest rates on consumer credit with an initial rate fixation of more

than 5 years in the six EU countries, 2003-2010, in percent. ..................305

Figure 133: Selected consumer credit interest rates in Poland, 2005-2010, in percent.307

Figure 134: Selected consumer credit real interest rates in the Netherlands, 2003-

2010 ..............................................................................................308

Figure 135: Instalment consumer credit interest rates in Germany, average cap and

bond yields (in percent), 1968-2003. ..................................................309

Figure 136: Instalment consumer credit interest rates in Germany, in percent, 1968-

2003. .............................................................................................310

Figure 137: Consumer credit interest rates in Germany, spreads and interest rate cap,

in percent, 1968-2003. .....................................................................310

Figure 138: Impact of a floating IRR on the cost of credit for the average borrower ....311

Figure 139: Acceptance of regulation of fees and charges .......................................316

Figure 140: IRR as a barrier to the cross-border provision of consumer credit ............318

XXIV iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 141: IRR as a barrier to the cross-border provision of consumer credit

(divergence of opinion among stakeholder categories)...........................318

Figure 142: Competition measures by the Herfindahl-Index.....................................320

Figure 143: Competition measured by the Learner Index ........................................321

Figure 144: Competition measured by the H-Statistic .............................................322

Figure 145: Consumer credit interest rates and ceiling in Poland, 2005-2010 .............324

Figure 146: Process of the surveys for the empirical work .......................................349

iff/ZEW – Final Report on interest rate restrictions in the EU 25

Introduction

In line with the dual objectives of this study, namely to identify existing interest rate

restrictions in EU consumer credit markets and to assess their effects on those markets

and society in general, this report is structured in two parts with annexed materials.

A first part of the report contains a comprehensive inventory of the types of

interest rate restrictions that exist in the EU Member States and outlines the

differences between underlying national legal, social and cultural traditions and

circumstances surrounding IRR and the protection of borrowers against exorbitant

charging by credit providers. It starts by providing the theoretical background and

framework for an understanding of the concrete situation in the Member States. In

Chapter 1.1, definitions are elaborated, terms of the study clarified, the philosophy of

regulation and concept of usury explained and the range of possible forms of IRR

presented. In these opening sections and throughout the report, the study applies a legal

methodology to assess the common roots, principles and conflicts of existing rules. A

socio-legal methodology is used to assess the cultural background for such rules and

their effectiveness in practice.

Chapter 1.2 details the mechanisms and levels at which interest rate ceilings are set in

those countries with such regulation. Direct regulation of credit markets through the use

of contractual interest rate ceilings are the most recognisable form of IRR and pages 64-

78 give details on the few Member States with absolute ceilings followed by those

countries here a relative interest rate ceiling (also known as a floating cap) is applied.

Country tables are provided showing the level of the ceilings for the different classes of

credits where distinctions are made on the basis of credit type or amount. This part of

the report also contains information on sanctions, enforcement regimes and an

explanation of the various ways in which ceilings can be calculated and set. In addition,

Section 1.2.2 gives a comprehensive inventory of rules governing interest rate levels in

default.

While direct IRR limit the contractual interest rate or the amount of interest that can be

charged, Chapter 1.3 presents the various forms of indirect IRR in Member States.

Restrictions on other cost relevant factors may have an equivalent effect to interest rate

ceilings. Details on such general restrictions on fees and charges are reported here. In

addition to providing information on the rules governing the use of compound interest

rates in EU Member States, there is also a section covering on the regulation of interest

rate variability, which is currently more related to disclosure and bilateral requirements

rather than strict restrictions on the interest rate as such.

With harmonisation of consumer-related rules in the different Member States, the legal

survey then proceeds to discuss the EU Consumer Credit Directive of 2008 (CCD 2008)

and an extension of its scope to smaller amounts of credit in some jurisdictions in

Chapter 1.4. The study also addresses certain legal aspects related to high-cost credit

and their social impact in Chapter 1.5. The first part of the report ends with Chapter 1.6

summarising and reproducing the views of Stakeholders of European consumer credit

markets on various aspects of IRR.

A second part of the report focuses on the economic, financial, and social impacts of

interest rate restrictions. Related debates about the economics of interest rate

restrictions can be at least traced back to a controversy between Adam Smith and

Jeremy Bentham in the late 1780s:3 Interestingly, Adam Smith, the proponent of the

invisible hand, argues in favour of a legal interest rate ceiling to shift capital allocation

towards more productive users, an outcome which he considers to be socially more

3 For more details of this controversy, see Persky (2007).

26 iff/ZEW – Final Report on interest rate restrictions in the EU

desirable. In contrast, Jeremy Bentham points out that high cost of credit will attract

economically weaker people who may rationally decide that the expenses for interest

payments are outweighed by the value of the loan to them.

Modern economics has advanced in its understanding of complex credit markets

(including the role of asymmetric information and imperfect competition). Chapter 2.1

details the reasons of differential interest rates as well as the effect of interest rate

restrictions on capital allocation. Economic theory has also refined its analysis of

consumer choices. In general, it acknowledges that consumers spend, save and borrow

money according to their individual preferences and their budget. The chapter also

demonstrates how a decision to finance consumption by credit arises from the preference

to smooth consumption over time. However, individual financing decisions are found to

have adverse individual as well as social effects such as over-indebtedness. Therefore,

Chapter 2.1 also shows the limitations of consumer rationality due to cognitive biases

and the inability to manage complex decisions. Based on these theories we state a series

of conceivable hypotheses on the impact of interest rate restrictions.

Note that, for three reasons, this study primarily focuses on consumer credit rather than

credit to businesses (eg. SMEs, self-employed, micro enterprises): firstly, interest rate

restrictions are frequently introduced as a means of consumer protection, since

consumers are the ones considered to be otherwise offered (excessively) high interest

rates. Secondly, as banks are reluctant to lend at high risk premia to small businesses to

avoid adverse selection, small enterprises (and micro enterprises) are frequently financed

by credit types originally designed for consumption purposes (eg. overdraft of the

(private) current bank account, credit card). However, it is not feasible to make a clear

empirical distinction between consumer finance and the financing of a (start-up)

enterprise. We therefore conjecture that the implications of the hypotheses derived for

consumers also hold for small and micro enterprises. Thirdly, a thorough discussion of

the financing of small and medium-size enterprises (SME) would require considering a

variety of aspects (eg. optimal leverage, moral hazard, the role of mezzanine capital and

equity) which are clearly beyond the scope of this study. We will nevertheless briefly

discuss similarities and differences between consumer credit and business credit

(including microfinance) in the context of interest rate restrictions in Section 2.1.5.

Due to the relevance of consumer credit markets and their regulation, it is not surprising

that economic impacts of interest rate restrictions have been the focus of empirical

studies, both with an academic as well as a policy-consulting perspective. Chapter 2.2

provides a comprehensive survey of this literature. It stresses that empirical evaluation

of interest rate restrictions can only be made when the counterfactual situation is

identifiable, eg. due to close similarity of countries unaffected by this regulation. As

Chapter 2.2 details, this requirement is more easily met in the analysis of the US

markets rather than European markets. However, the typical historical interest rate cap

in the US is lower than typical interest rate restrictions in European countries, which

makes the transferability of these results questionable. Chapter 2.2 therefore also takes

a closer look at more recent studies on interest rate restrictions in Member States of the

European Union.

Chapter 2.3 then turns to the description of the markets of credit to households. It

provides a comprehensive overview of the markets for total credit to households, housing

credit, and consumer credit (without housing) for all EU 27 countries. Where illustrative,

it also provides separate analyses for EU 25, EU 15 countries or the New Member States.

Chapter 2.4 presents a more detailed discussion of the credit markets in the case study

countries. These countries are Germany, France, Sweden, Poland, the UK and the

Netherlands. As detailed in Section 2.4.1, these countries are particularly well suited as

case studies as they exhibit typical features in terms of the interest rate restrictions as

well as economic circumstances.

iff/ZEW – Final Report on interest rate restrictions in the EU 27

Chapter 2.5 discusses the hypotheses stated in Section 2.1.4. To do so, we summarise

existing evidence from the literature, and derive conclusions from the collected data. In

addition, we consider the responses of questionnaires answered by particular

stakeholders in the consumer credit market, such as regulators, consumer associations,

providers, and provider associations. The survey enables us to learn from the experience

of stakeholders with regard to the effect of interest rate restrictions (or lack of thereof)

on the credit markets in the respective countries. It also allows a comparison of the

answers from stakeholders in countries with interest rate restrictions (IRR) with those

from countries without IRR, and a qualitative discussion of the differences. The inclusion

of heterogeneous stakeholders (regulators, consumer agencies, supplier associations) will

enable us to consider potentially divergent perspectives about issues in question in a

balanced way. Accordingly, we base our judgement about the hypotheses on several

sources: the interpretation of existing data, earlier empirical evidence, as well as the

information from stakeholder questionnaires.

28 iff/ZEW – Final Report on interest rate restrictions in the EU

1 Legal survey of interest rate restrictions

1.1 Legal theoretical background

In this section we provide the theoretical background and framework for an

understanding of the concrete situation in the Member States as described through the

reports of legal experts and the various stakeholder groups.

1.1.1 Definition of interest rate restrictions (IRR)

The Handbook of research on international consumer law4 introduces its chapter on price

controls with the following short overview of the presence of interest rate ceilings:

“Many countries in both the developed and developing world have interest rate ceilings on

consumer credit.5 These include France, Belgium, Netherlands, Poland, Slovakia, Ireland,

some Australian states, Canada, some US states, Brazil, South Africa and Japan.6 The

German Supreme Court has established a de facto ceiling through its interpretation of the

BGB. In addition, Islamic banking prohibits the taking of interest and uses a profit-sharing

model.7

The primary contemporary objective of interest rate ceilings is to protect those of modest or

low income from paying excessively high prices for credit. These ceilings often include both

consumer and business borrowers in the scope of their protection. Given the different markets

and forms of credit, countries often have a variety of ceilings. In France, interest rate ceilings

differ depending on the type and length of loan: they are established at one third above the

prevailing market rate for the particular market segment.8 South Africa has adopted a similar

approach, with loans divided into seven categories.9 In Germany, the Supreme Court has

established a very strong presumption that interest rates that are double the relevant market

rate are contrary to good morals (section 138 of the BGB).10 In contrast, the UK has not had

general interest rate ceilings on credit since 1854 when Parliament, following Jeremy

Bentham's strictures against usury, abolished the usury laws.11

This reveals that there are a number of definitions of interest rate restrictions, stemming

from the fact that they have often been introduced to achieve distinct policy objectives.

We therefore begin by defining what we mean by interest rate restrictions and

considering some of the complexities that arise for the conduct of the study. A set of key

terms for use in the study is set out as a result.

4 Ramsay, I. in Howells, G., Ramsay, I. , Wilhelmsson Th. (2010) pp 397 ff.

5 There is a large literature from differing academic perspectives on interest rate ceilings. For a recent

overview, see Howell, N. (2005) and references cited. Elizabeth Warren and Amelia Warren Tyagi have

argued for the reintroduction of interest rate ceilings in US consumer credit. See Warren, E., Tyagi, A.

(2003), pp. 144-52. See references in Kraft et al. (2008), pp. 409-12. For economic explanations, see

Glaeser, E., Sheinkman, J. (1998), Posner, E. (1995), p. 283 et seq.

6 See discussion of Japan in Kozuka, S.; Nottage L. (“The Myth of the Cautious Consumer: Law, Culture,

Economics and Politics in the Rise and Partial Fall of Unsecured Lending in Japan” in J Niemi, I Ramsay and

W Whitford (eds) Consumer Credit, Debt & Bankruptcy: Comparative and International Perspectives

(Oxford, Hart 2009).

7 See Aidit bin Hazi Ghazali (1994), p. 443 et seq.

8 See Code de la Consommation L.313-3.

9 See 42(1) NCA.

10 For a brief account of the German law of usury, see Markesinis, B. et al. (2006), pp. 250-53. The German

courts may use two tests: double the average and a standard of 12 per Cent above relevant rates. The

latter was used during periods of high Inflation and consequently high interest rates.

11 Although the Moneylenders Acts 1900-1927 (abolished in 1974) did contain a presumption that interest

rates over 48 per cent were unconscionable.

iff/ZEW – Final Report on interest rate restrictions in the EU 29

We then proceed to consider the concept of usury, detailing the provisions which exist in

this regard in each of the Member States, and historical fluctuations in its application.

We then categorise existing forms of IRR, which we then use in the remainder of the

Chapter when considering responses to the survey of experts, and we conclude with

consideration of enforcement issues.

1.1.1.1 Legal and economic definitions

The term, ‘Interest Rate Restrictions’ in consumer credit markets refers to all rules that

limit the price of credit contracts. The study is concerned with existing law and its

application. It is not a legal study in its strictest sense, which would need to consider the

broader discussions that have taken place since the middle ages concerning how

concepts of usury became part of modern law12 and the specific forms of interest rate

restrictions that arose as a result. Instead, this study is concerned with interest rate

restrictions that currently exist as legally or factually binding rules that limit the amount

of money a lender can charge, directly or indirectly, for the use of capital by the

borrower. Interest Rate Restrictions will be abbreviated and referred to as ‘IRR’ for

simplicity and ease of reading throughout this report.

It should be noted from the outset that there are three complicating factors of relevance

to the study:

Firstly, there is no direct way of assessing the effects of the enormous amount of credit

legislation with regard to the pricing of credit, since too many variables may intervene.

From a supplier’s perspective, all legal rules require them to take special precautions or

put procedures in place, and these ultimately contribute to the overall cost of providing

credit. The study cannot take all of these variables into account and is therefore only

concerned with identifying the legal forms of IRR that exist and comparing whether, and

how far, we can attribute observed impacts on prices to them.

Secondly, the definition and measurement of the price of the credit itself is of critical

importance, and rules affecting the calculation and disclosure of interest have effects on

the operation and forms of IRR. For example, the interest rate defined in Article 3 of the

CCD 200813 is used both to disclose the total price of a credit contract in a comparable

way (“Annual Percentage Rate of Charge”) and as a means of calculating the amount of

interest due (“Borrowing Rate”).

Thirdly, some legal concepts are either not present in all Member States or are used

differently amongst them. In order to facilitate the studying of the economic and social

effects of IRR, we have adopted a broad definition as follows:

“Legal Interest Rate Restrictions are all legal rules, from whatever source they

come, which intentionally restrict the price of consumer credit.”

As far as self-regulatory rules exist, they are either directly or indirectly part of legal

rules because private law gives contractual consensus the status of state sanctioned law

(“freedom of contract”) or incorporates such rules as “commercial habits”, “good morals”

or “good faith” into the body of law.

12 See for Germany: Liebner, K. (2010); Dilcher, J. (2002); USA and Great Britain: Temin P., Voth, J.

(2008), pp 743-758; Bodenhorn, H. (2007), pp.179-202; France: Pikulska-Robaszkiewicz, A. (2000), pp.

715-734; Italy: Olivi (1990); Spina, R. (2008); Spain: Vitoria, F. (2006).

13 Directive 2008/48/EC of the European parliament and of the Council of 23 April 2008, on credit agreements

for consumers repealing Council Directive 87/102/EEC. Referred to as the “Consumer Credit Directive”

or “CCD 2008”. Available at: http://eurlex.

europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2008:133:0066:01:EN:HTML.

30 iff/ZEW – Final Report on interest rate restrictions in the EU

As far as the enormous amount of historically developed moral, religious and ethical rules

are concerned these rules have mostly led to legal rules which have thus gained

acknowledgement and effectiveness. In so far those moral rules which are broadly

shared in a country are usually incorporated into the law. Those exemptions where in a

rather small and homogeneous community like for example the Scandinavian countries

moral restrictions are so widely accepted that even without legalisation they have a high

degree of effectiveness we have integrated them into our observations.

In economic terms, this is therefore a study of the impact of price controls in consumer

credit markets.

For the economic part of this study the word “consumer credit” and “interest rate” are

common, well-defined and explored. This is even true for EU law. Article 3 (c) Directive

2008/48/EU defines a “credit contract” legally by reference to the economic concept of

“consumer credit” and sets out a number of its legal forms including ‘deferred payment’

and ‘loan’, but it also adds a saving clause by introducing the term “similar financial

accommodation”.

„‘credit agreement’ means an agreement whereby a creditor grants or promises to grant to a

consumer credit in the form of a deferred payment, loan or other similar financial

accommodation, except for agreements for the provision on a continuing basis of services or

for the supply of goods of the same kind, where the consumer pays for such services or goods

for the duration of their provision by means of instalments;”

It thus combines the Common Law tradition - which uses economic denominations in its

credit legislation - with the civil law approach, which in the Roman tradition of formalised

legal language (“Begriffsjurisprudenz”14) keeps more strictly to forms.15

Assessing the national rules on “credit” is therefore easy in those countries that use this

word as a core legal concept.

But in the civil law tradition16 such purpose-driven denominations are seen as detrimental

to the requirements of the rule of law. For example, France and Germany both prefer

concepts that refer to the legal forms in which the economic activity is exercised. In

private law they prefer legal forms like a “loan” (Darlehen, prêt) or an instalment

purchase (“Stundung” deferred payment).

To cover the same rule in the UK as in Germany the words credit and interest rate have

to be translated into the different legal forms used in their respective legislation. For this,

EU law provides assistance. The 1987 EU Directive on consumer credit (CCD 1987), the

original CCD, aimed at bringing about a degree of approximation of the laws, regulations

and administrative provisions of the Member States concerning consumer credit17,

14 Especially for money terms see Ott, K.-O. (1998).

15 See for example Part II of the UK Consumer Credit Act 1974 (1974 c. 39) which is similar to the EU

Directive defining credit contracts by “credit”.

16 Section 488 (1) of the German Civil Code uses the term “loan”: “(1) The loan contract obliges the lender to

make available to the borrower a sum of money in the agreed amount. The borrower is obliged to pay

interest owed and, at the due date, to repay the loan made available to him.” Similar the French definition

in Article 1892 Civil Code: “A loan for consumption is a contract by which one of the parties delivers to the

other a certain quantity of things which are consumed by use, on condition that the latter shall return as

much to him in the same kind and quality.”

17 Council Directive 87/102/EEC of 22 December 1986 for the approximation of the laws, regulations and

administrative provisions of the Member States concerning consumer credit. This Directive was

subsequently revised twice by Council Directive 90/88/EEC of 22 February 1990 amending Directive

87/102/EEC (setting out a single mathematical formula for calculating the APRC throughout the Community

and for determining credit cost items to be used in the calculation), itself amended by Directive 98/7/EC

(focusing on the calculation of the APRC). Directive 87/102/EEC was later repealed by Directive 2008/48/EC

of 23 April 2008 on credit agreements for consumers (CCD 2008).

iff/ZEW – Final Report on interest rate restrictions in the EU 31

introduced a purpose-driven approach into the national systems of countries governed by

codified civil law. Germany incorporated the use of the word “Credit” (Kredit), formerly

unknown in Private Law in its 1971 Consumer Credit Code. It returned to the formalised

legal language (“Loan”) in 2002 when integrating credit law into its Civil Code.

The maximum harmonisation approach of CCD 2008 Article 3 is now the key to the

different legal languages. Each Member State will have to define which legal forms and

rules of their own legal order qualify for what the Directive defines as “credit” for

“consumers”.

As the Directive aims to cover all legal forms in all Member States which serve the

purpose of credit it leaves some scope for interpretation. Whilst it sets out a number of

forms of credit which must be included, such as loans, deferred payments, overdrafts,

credit cards, it remains open to other and new forms of what would be recognised as

credit or an interest rate in economic terms.

We share this open approach and seek to cover interest rate restrictions for all legal

forms which objectively function as credit, even if they have a different legal

designation18.

1.1.1.2 Terms used in the Study

This is why the study is focussed on all legal forms covered by the definition of the

CCD 2008 without regard to its exemptions in Article 3. We therefore cover all

regulations with reference to “credit” or its forms like “loans”, “leasing” “hire purchase”

etc where we expected that different forms of IRR might be feasible.

Definitions of the key terms used in the study are set out in the table below.

Table 1: Definitions of key terms used in the study

Credit “whereby a creditor grants or promises to grant to a consumer credit in

the form of a deferred payment, loan or other similar financial

accommodation” (from CCD 2008).

Instalment

credit

The borrower is provided with a fixed amount to be repaid over a given

period by a fixed number of repayments called instalments (usually

constant over time) Examples: Personal loans, car loans, and hirepurchase

agreements.

Revolving

credit

A permanent reserve of credit whose limit is authorised by the creditor;

the consumer repays the sum used according to the allowances stated

in the credit contract and the reserve reconstitutes itself as repayments

progress. Mechanism of repayments can take different forms, including

the regular repayment of a percentage of the outstanding balance, with

or without a minimum amount, the periodic payment of a fixed amount,

or the payment of interest charges regularly and the repayment of the

credit at the end of the agreement Examples: credit cards, revolving

credit accounts, and overdraft facilities. Examples: line of credit,

running account credit, overdraft, credit card credit.

18 Historically this was especially true for certain contracts which are presented in the form of a hire

agreement. Article 3 of the Directive defines as “credit” (d) hiring or leasing agreements where an obligation

to purchase the object of the agreement is (not) laid down either by the agreement itself or by any separate

agreement”

32 iff/ZEW – Final Report on interest rate restrictions in the EU

Secured

credit

A credit in which the borrower pledges some asset as collateral for the

credit. Creditor recovery of the surety can be limited to the collateral

(non-recourse) or not. Examples: mortgage loans and home equity

loans (a form of equity release).

Interest

Rate

Restrictions

Absolute or relative rate ceilings (fixed administratively, by statute or

court rulings); Laws designed to prevent exploitation and unfair

competition with effects on credit cost; Capped default interest rates

and early repayment fees; Restrictions on the compounding of interest

and the use of variable rates; Other forms of restrictions to the level or

rate of interest including moral consensus; Anti-Trust regulation.

Borrowing

rate

“(j) ‘borrowing rate’ means the interest rate expressed as a fixed or

variable percentage applied on an annual basis to the amount of credit

drawn down” (from CCD 2008).

Annual

Percentage

Rate of

Charge

(APRC)

“(i) ‘annual percentage rate of charge’ means the total cost of the credit

to the consumer, expressed as an annual percentage of the total

amount of credit, where applicable including the costs referred to in

Article 19(2); (g) ‘total cost of the credit to the consumer’ means all the

costs, including interest, commissions, taxes and any other kind of fees

which the consumer is required to pay in connection with the credit

agreement and which are known to the creditor, except for notarial

costs; costs in respect of ancillary services relating to the credit

agreement, in particular insurance premiums, are also included if, in

addition, the conclusion of a service contract is compulsory in order to

obtain the credit or to obtain it on the terms and conditions marketed”

(from CCD 2008).

Default

interest rate

The interest rate with which the amount of money concerning capital

due after default is calculated (ie. charges for non-compliance, late

payment or interest charged on overdue payment).

Legal

interest rate

The interest rate provided by statute to be used when no contract or

agreement exists between the parties. Sometimes referred to as the

statutory interest rate (eg. a rate used by default because none has

been specified, or rate used for statutory payments such as taxes).

Though the term is sometimes used to refer to the IRR ceiling, this is

misleading and inaccurate. It is distinct from the “lawful interest rate”

(rate that falls within the limits allowed by the usury laws) and from the

contractually-agreed rate of interest (conventional rate).

Cost In addition to interest the cost of the credit is made up of fees and

charges:

Finance

charges

The charges for interest. These charges depend on the borrowing rate,

the amount and the duration of the credit. (The borrowing rate can be

fixed or variable and its level depends on the characteristics of the

credit, the creditor and the borrower. Calculation of interest charges

could be straightforward or not when the credit includes difference

balance segments with different borrowing rates, limits, and

introductory rates and charges).

Non-finance Administrative fees (set-up costs, maintenance costs), servicing fees

iff/ZEW – Final Report on interest rate restrictions in the EU 33

charges linked to payment transactions and drawdown, fees and charges for

sureties and ancillary services (bank accounts, credit insurance and

payment protection insurance), but also early repayment fees and fees

for failures to comply (late payment charge, exceeding credit limit).

Providers Banks (commercial banks, mutual and cooperative banks, savings

banks); Finance companies/mortgage specialists; Doorstep tallymen;

Moneylenders; Brokers; Insurance companies; Vendors; Mail order

companies; Service providers; Associations/coops/unions (if not

banks); State agencies (housing, welfare etc); Private persons (family,

friends etc); Pawn brokers; Other.

Legal

sources

Statutory law; General Civil Law ; Special Credit legislation;

Competition Law; Penal Law; Procedural Law; Administrative (Public)

Law; Court decisions; Self Regulatory rules (codes of conduct, issued by

arbitration bodies, professional rules) ; General Conviction, religious,

moral, voluntary ethical rules etc.

Enforcement

agencies

Civil courts; Consumer courts; Consumer Ombudsmen (official or

private); Police; Central bank; Banking Authorities (supervisory); Fair

Trading Offices; Cartel Offices; Attorney General etc.

Source: iff.

34 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 1: Forms of consumer credit and Forms of IRR

Types of Credit (listed by form)

I. General-purpose credit

A. Instalment credit

Instalment loan (auto loan, home

appliances, energy saving projects)

Variable rate credit (variable interest rate,

variable repayment. Not credit card)

Fixed repayment credit for general

purpose (single repayment)

Financial leasing

Hire purchase agreement (vendors,

service providers etc)

Point-of-sale financing (vendors)

Deferred payment in sales contracts

(instalment sales)

Home equity loan (second mortgage loan

for financing consumption)

B. Revolving credit

Overdraft (including credit/deferred payment

from debit cards also called “credit cards”)

Overrunning (exceeding current account

overdraft)

Revolving credit account (with or without

credit card support)

True credit card credit (if repayable in

instalments with own credit facilities)

Deferred debit card credit (incl. if charge

card pay-later facility is less than 3 months of

credit, also from stores)

C. Small secured credit

Pawn broking

Payday loan (up to 6 months)

Other: please indicate

D. Micro credit

II. Mortgage Loans

Mortgage loan

State subsidised mortgage loan

Savings and loan schemes

Endowment loan (linked to capital life

insurance, savings or investment contracts)

Other: please indicate

Interest Rate Restrictions (IRR)

I. IRR on the rate level

Contractual interest rates

Rate caps (absolute or relative) – objective

control

Usury (general ceilings, case by case, in

connection with other items, administrative or

civil law, penal) – subjective control

Default charges

Default interest rate caps

II. IRR on methods of

calculation of the rate

Anatocism (restrictions on interest on

interest)

Mathematical rules (beyond the CCD 2008

restricting interest)

Interest compounding periods

(compounding interest to capital)

Variability of interest rate (interest rate

caps to floating or progressive rates)

III. Restrictions on other cost

elements

Contractual charges

Insurance fees (amount, purpose, financing,

commissions etc)

Broker fees (amount, purpose, financing)

Account holding fees

Maintenance fees

Default charges

Penalties

Amortisation (allocation of payments to

reduce outstanding principal, priority in reducing

interest, cost or capital)

IV. Restrictions on other credit

parameters

Instalments (size, number, period)

Lifetime of the credit (duration)

Total amount of credit

Net amount of a credit

iff/ZEW – Final Report on interest rate restrictions in the EU 35

1.1.2 Empirical survey

The main source drawn on by this report is questionnaires and discussions with experts

and stakeholders. The questionnaires and discussion materials have been developed on

the basis of a number of studies iff has already conducted with regard to interest rates

and over-indebtedness in Europe. However, existing literature and research on the

specific question of IRR is still rather limited. A number of economic studies in the U.S.

are discussed in the economic part19 and there is a vast literature discussing the legal

dogmatics of usury, especially from an historical perspective.20 But there are only very

few empirical studies which assess the existing modern forms IRR.

Only two studies directly using empirical data have recently been published – one in the

UK in 2004, the other in France in 2009. Other studies published in Poland, the

Netherlands and Slovakia mainly use English data, while a forthcoming study by the

Danish government has yet to be made available to the public. According to the

information from national experts, there are additional reports on interest rate

restrictions in another six Member States.21

The English study, widely distributed in Europe22 and elsewhere, was conducted by TNS

Global Market Research for Policis, a consultancy in specialist research, which conducted

this project at the request of the UK Department of Trade and Industry.23 Its main finding

was that usury ceilings in Germany and France had led to a high degree of exclusion

from small loans for poor people in these countries and created a market for illegal

lending. These findings have been referenced on several occasions but have also been

criticised.24 The French study was conducted in 2009 by the French government25. It

concludes that the problems of usury today are concentrated in consumer credit, and

especially in certain forms of revolving loans and small loans, which play an important

role among insolvent households in France, as opposed to mortgage loans. The report

19 See at pp 158 ff.

20 See above FN 5.

21 From the national discussion concerning IRR in a wider sense a number of publications, papers,

parliamentary motiongs and studies have been named by the experts which mostly do not provide empirical

evidence beyond those known from the main studies. Belgium : Révision du mode de fixation des taux

annuels effectifs globaux en matière de crédit à la consommation (2006) at

http://statbel.fgov.be/fr/binaries/354_tcm326-41864.pdf; Czech Republic: Ministry of the Interior of the

Czech Republic - Prevention of indebtedness study. Liberalni Institut (Schwartz, 2007): Denmark Ministerial

working group to analyze the effect of APR-ceilings in other countries (report not yet published) . France:

Rapport sur les modalités de fixation du taux de l'usure", General Inspectorate of Finance & General

Inspectorate of Social Affairs, Februray 2009. See http://www.responsiblecredit.

net/index.php?id=1980&viewid=45728; Lefebvre, M. F. (2008), Rapport d’Information sur les

emprunts immobiliers à taux variable, Paris: Assemblé Nationale Treizième Légistlature N° 744 (26 mars

2008) p 53 (II A 2 c); Germany: Report by the Bundesbank concerning the change in statistics and the

Report on its implications by Hartmann-Wendels/Spörk in 2006. See:http://www.responsiblecredit.

net/index.php?id=1980&viewid=45755; Netherlands: study on SMS-credit and pawn-brokers was

done in 2009. See Report from "Research voor beleid'. http://www.responsiblecredit.

net/index.php?id=1980&viewid=45753.; Poland: KPF commissioned a study evaluating the impact

IRR have had on the market. The report was published in March 2009 (English translation avaliable).;

Slovakia: A report by the Hayek Foundation see http://www.responsiblecredit.

net/index.php?id=1980&viewid=45753; UK: Financial Inclusion Taskforce on Access to affordable

credit (2010); OFT study on High Cost Consumer Credit (2010).

22 See Eurofinans (2010), pp 6 ff; Project Associates, Briefing Note on Interest Rate Ceilings, 2009 (made for

Provident in Poland); Information on the use of these data in the Japanese discussion was made available

by Saya Ojama at the iff-Hamburg Conference on Financial Services, July 3, 2010.

23 For a more critical evaluation see now Office of Fair Trading (2010).

24 Policis (2006b); Policis (2006a), pp 47-48; Policis (2004a); for a critical review of its methodology, sample

and legal understanding see Reifner, U., Knobloch, M. (2009). The respective part for France has not been

mentioned in the French survey on these issues in 2009.

25 Inspection générale des finances, Inspection générale des affaires sociales (2009).

36 iff/ZEW – Final Report on interest rate restrictions in the EU

cites Belgium, the Netherlands and Italy as other countries which have comparable IRR

to France and sets the focus beyond access and onto two opposing views - the positive

impact of rate ceilings on the risk of insolvency, which high cost credit entails on one

hand, and its negative impact on access to small loans for people with low credit score

values on the other.26 Both studies are discussed later in the report.27

In our empirical survey, a variety of different data sources were used, firstly to

comprehend the idiosyncrasies of each Member State’s credit market and secondly to

assess the possible impact interest rate restrictions may have had and the potential

effects these restrictions could have when applied to the markets in other Member

States. We have restricted the selection of data to reliable industry and public sources

that allow for comparison across the Member States and have also incorporated some

national sources for specific phenomena which are measurable.

Telephone interviews with certain stakeholders were conducted and a number of face-toface

meetings were organised with certain specific stakeholders. Though stakeholders

vary as to their usefulness and knowledge of the subject matter they have been able to

contribute to the research. Additionally to the bilateral communication with stakeholders,

part of the research team has presented the study to the Forum of users experts in the

area of financial services (FIN-USE) set up by the European Commission.

The following section gives an overview of the stakeholders contacted and having

contributed in some way to our research, however, the details of our methodology have

been included in Annex III: Methodology of the research.

The Stakeholder Questionnaire (Annex X: Stakeholder Questionnaire) has been sent to

over 330 agencies. Due to the fact that the three groups of stakeholders (public

authorities, provider associations and consumer organisations) in countries with no actual

discussion on IRR found that responding was not necessary for various reasons, iff had to

use additional time and sources to solicit stakeholder involvement. To improve the

response rates to the questionnaires, the participation of the provider associations at the

EU level in the dissemination of both the Stakeholder Questionnaire and the Provider

Questionnaire to their members and their members’ members was respectively secured.

By securing the involvement and support of Eurofinas and the European Mortgage

Federation in this way, we hoped to encourage responses from the national level.

Of the 333 agencies contacted and invited to participate in our research:

96 have completed a Stakeholder Questionnaire (though not all questions were

answered in each of these);

12 have declined to answer because they did not consider their institution as

sufficiently knowledgeable or with the competency about the issues covered by

the survey;

41 have replied by sending material or a significant answer but without

completing a questionnaire.

26 “L'augmentation des taux d'usure, en élargissant les possibilités de pret aux catégories les plus risquees, a

pour effet d'accroitre le niveau des difficultés d'endettement des ménages, Selon Georges Gloukoviezoff, le

flux de clientèle supplémentaire permis par la hausse des taux d'intérêt pratiques correspond a l'admission

d'emprunteurs plus risques, donc davantage sujets a des défaillances de paiement.” op. cit. Annex 5 p 1.

27 See 2.2.5 at pp 161 ff.

iff/ZEW – Final Report on interest rate restrictions in the EU 37

In addition, the economic team of this research conducted a survey of individual

providers. Further details on the provider survey can be found in the Annex XII: Provider

Questionnaire. All findings are reflected in the analysis of the hypotheses (Chapter 2.5).

Table 2 gives an overview of the numbers involved in the stakeholder survey and the

respondents from the Member States that participated.

Table 2: Survey participation with the Stakeholder Questionnaire by stakeholder group

Provider

associations

Consumer

associations

Public

Authorities

Countries Others

Invited Replied Invited Replied Invited Replied Invited Replied

Austria 3 0 2 0+(2) 3 1 0 0

Belgium 3 1 5 2+(1) 3 1+(1) 0 0

Bulgaria 2 0 6 1 2 1 0 0

Cyprus 2 0 2 0 5 3 0 0

Czech Rep. 3 3 4 1+(1) 3 1 2 2

Denmark 4 1 2 1 3 1 1 1

Estonia 1 0 2 0 5 2+(2) 1 1

Finland 1 1 4 2 3 2 1 1

France 3 2+(1) 13 5+(2) 6 1 6 1

Germany 10 2+(2) 3 1 4 1+(2) 5 0

Greece 1 0 5 1+(1) 1 1 0 0

Hungary 3 0 5 1+(1) 3 1 0 0

Ireland 4 0+(1) 4 0+(1) 2 1 1 0

Italy 6 2 8 2+(1) 2 2 1 0

Latvia 2 0 3 1 3 1 0 0

Lithuania 3 0 3 0 3 1+(1) 0 0

Luxembourg 1 0+(1) 1 0+(1) 2 1 1 1

Malta 2 0 2 0 3 1 0 0

Netherlands 4 1+(1) 2 1+(1) 3 2 2 1

Poland 5 1 2 1+(1) 3 3 0 1

Portugal 3 0 5 2 1 0+(1) 2 1

Romania 2 1 1 1 1 0+(1) 1 0

Slovakia 1 1 2 0 2 1 0 0

Slovenia 1 1 1 1 3 2 0 0

Spain 5 0+(3) 5 2+(1) 1 1 2 0

Sweden 4 1 2 1 5 1+(1) 1 0

UK 11 5+(1) 10 1+(2) 8 2+(3) 17 0+(2)

EU Institutions 8 0+(1) 2 0 0 0 2 0

Total 98 23+(11) 106 28+(16) 83 35+(12) 46 10+(2)

Source: Stakeholder survey. Note: The Stakeholder group Public Authorities (comprised of Regulators and

Government) and Other (comprised of debt advisors, scientists and other experts) have been merged together

for the purposes of analysis of the effects of interest rate restrictions in the later Chapters evaluating the

impact on stakeholder groups. The number in brackets refer to additional responses that were not sent in the

form of a completed questionnaire, thus are not part of the quantitative analysis in later Chapters of this report.

Some authorities have responded to one questionnaire together ie. have submitted one response to which the

different bodies have contributed. Furthermore, individual provider responses to the provider questionnaire are

not included in this table (see: Annex XIII: Provider Questionnaire - Methodology and feedback).

38 iff/ZEW – Final Report on interest rate restrictions in the EU

iff contacted all national bankers’ associations, bank supervision authorities and central

banks, as well as consumer organisations in each Member State, using various channels.

The level of non-response may in some cases indicate relative low levels of interest in

this area in the country concerned (eg. some German providers), but a lower than

expected response rate is primarily down to limited resources by the participants invited

to contribute their views (eg. especially the consumer associations). The response rate

from regulators was also more difficult to obtain than expected. Feedback would tend to

suggest that other priorities have meant that a prompt and dedicated response to our

survey was not always possible, though a public authority from every Member State was

contacted, called and a statement of their views collected.

The legal findings have been integrated with those from the economic part28. The

inclusion of heterogeneous stakeholders (regulators, consumer agencies, provider

associations) enabled the economic team to consider potentially divergent perspectives

about issues in question in a balanced way. These responses to the Stakeholder

Questionnaire were taken into account when describing the hypotheses used to analyse

the impact, which interest rate restrictions can have. Alongside these written

contributions, some oral communication helped clarify certain details and helped assess

the strength of certain responses received from those respondents that completed a

questionnaire. Chapter 2.5 reports on the quantitative evaluation of the responses from

the questionnaire in light of the hypotheses. In addition Chapter 1.6 Stakeholder views

on IRR contains further material and a selection of qualitative answers are provided in

Annex XI: Tables of stakeholder responses and Annex XV: Stakeholder general feedback

on IRR.

To favour transparency of the research content and methodology, the findings in this

report were subject to verification ahead of its final version. This primarily took the form

of forwarding our understanding of the legal situation, but also the economic or market

information we had, to the public authorities to ensure that the content of the answers

collected, primarily from our legal experts for the legal details, is accurate.

1.1.3 General principles of IRR in national legislation

There is one term that seems to summon up all forms of existing IRR: usury. It is used

to describe a fixed rate cap in France (taux d’usure); governs the German Austrian model

of “Wucher” (usury); and is the term underpinning those systems where the exploitation

of market failure by the stronger party, which results in excessive prices, is seen as a

contravention of good morals. The following table summarises sum indicators which are

described more in depth in the following text.

Table 3: Usury legislation in the EU

Country Denomination “exploitation” “ceiling” Other

Austria Wucher Art. 879 alinea 2 4th

sentence und alinea

3 Civil Code (ABGB);

Articles 154 and 155

Criminal Code Article

1 Law on Usury

Belgium Woeker/usure Article 494 of the

penal code

Article 1907ter

BW

Article 87 No 1

WCK (ceiling)

28 See 2.1.4 Resulting Hypotheses on page 155 for the hypotheses that were tested by the survey responses.

iff/ZEW – Final Report on interest rate restrictions in the EU 39

Bulgaria лихварство Article 10 (3) of the

Obligations and

Contracts Act

(default interest)

Art. 294

Commerce Act

(anatocism)

Cyprus Τοκογλυφία/

Tokoglifia

Czech

Republic

Lichva/dobré

mravy

Section 3 of the

Czech Civil Code

(Act. No. 40/1964)

also in the criminal

code (Act No.

40/2009 Coll)

“Four times the

average” (court

rule)

Denmark åger Yes

Estonia liigkasuvõtmine “good morals” (Civil

Code)

Finland Kiskominen/

Koronkiskonta

Penal Code

France Usure L313-3 of

consumer code

Germany Wucher Art. 138 al. 2 Civil

Code; Art. 291

Criminal Code

Art. 138 al. 1 Civil

Code (“good

morals”, double of

the average)

Greece τοκογλυφία Art. 404 Criminal

Code

Hungary Uzsorakamat Yes

Ireland Excessive “excessive”

Consumer Credit Act

1995 section 45

Italy usura Art. 644 Criminal

Code

Latvia Augļošana Criminal Law Section

201 “deprivation of

liberty”

Lithuania lupikavimas

Luxembourg usure Art. 1907-1 of the

Civil Code Article 494

penal code; article

1154 Civil code

anatocism

Judge can lower

interest.

40 iff/ZEW – Final Report on interest rate restrictions in the EU

Malta usury Illegal but with many

exemptions for

professionals

Netherlands Woekeraar/

wettelijke rente

No legal term. Usury

is deemed a criminal

offense

Fixed ‘legal

interest rate’

Poland Lichwa Usury for all

transactions

Portugal Usura cf. article 282 of the

Civil Code (“Código

Civil”) and article

226 of the Criminal

Code (“Código

Penal”)

cf. article 1146 of

the Civil Code –

non banking - and

article 28 of

Decree-Law nr.

133/2009

Romania camatarie Criminal Code of 28

June 2004, published

in the Official

Gazette, Part I no,

575 of 29 June 2004

“usury” in art.

450 Criminal Code

unauthorized

lending if interest

is higher than

allowed or

anatocism or for

less than one

year.

Slovakia úžera,

úžerníctvo,

úrok

Criminal offence

according to art. 235

of The Act No.

300/2005 Coll. - The

Criminal Code.

>30% pa (Decree

Ministry of

Justice)

Slovenia oderuštvo Article 119 of Code

of Obligations;

Obligacijski zakonik,

OZ-UPB1, OJ 97/07.

Article 214 of Penal

Code; Kazenski

zakonik, KZ-1, OJ

55/08)

Spain Usura Abusively high

interest rates Law on

Usury of 23rd July

1908 «Ley

Azcárate».

Art. 19 Consumer

Credit Law (2,5

times “legal

interest rate” for

overdrafts’

(Banks/Financial

Inst.)

Sweden Ockret Contracts act

(Avtalslagen 1915:

218) section 31

paragraph 2.

Swedish Penal

Code/Brottsbalken

chapter 9 s 5 p 2:

credit in a

business activity

or habitually or on

a large scale, and

extortionate

iff/ZEW – Final Report on interest rate restrictions in the EU 41

United

Kingdom

Usury/unfair Lending at excessive

or exorbitant interest

(not defined in

legislation

For Credit Unions

Source: Stakeholder and Expert survey.

In Austria, usury (“Wucher”) is regulated in civil law (Article 879 alinea 2 4th sentence

und alinea 3 Civil Code (ABGB). It means exploiting a disadvantage (such as

inexperience, carelessness, or financial distress) to exact a larger than usual benefit. This

exploitation is necessary for a situation to qualify as usury. The same definition can also

be found in Article 1 Law on Usury (WucherG); similar to Articles 154 and 155 Criminal

Code (StGB).

In Belgium, usury (“woeker”) was regulated in 1935 in the civil code (BW). Violation

could lead to a reduction of the contractual interest rate to the statutory interest rate. A

usurer can also be penally sanctioned, according to article 494 of the penal code (Sw).

The usury ceilings are laid down in article 1907ter BW. For its application an objective

element (ie. an abnormal high interest rate) as well as a subjective element (ie. the

abuse of one’s financial subordination) should be present. This applies to all types of

credit. The Belgian legislator has also adopted specific rules on usury in the Belgian

consumer credit regulation (“WCK”). According to article 87, 1° WCK the interest is not

payable by the consumer if the APR exceeds the legally determined APR.

According to Bulgarian legislation and court practice usury is prohibited by Article 10 (3)

of the Obligations and Contracts Act in so far as it is charged on default interest

determined in accordance with Bulgarian National Bank regulations. Anatocism is allowed

between commercial entities but has to be stipulated according to Art. 294 Commerce

Act.

In Cyprus, usury, “tokoglifia” transliterated is not legally defined and there have been no

interest rate ceilings for over a decade. Though lending transactions between private

individuals have never been regulated, a usury bill called ‘the Penal Code (Amendment)

Law of 2010’ is currently being proposed which will contain a definition of what usury

actually is, including a reference rate which is likely to be stipulated.

In the Czech Republic the term “usury” (in Czech: “lichva”) concerns an obligation which

is inappropriate to the profit obtained through it. In most cases it is being understood as

a credit with high interest. The duty not to apply usury interest arises from Section 3 of

the Czech Civil Code (Act. No. 40/1964 Coll. as amended) which governs “good morals”

(in Czech: “dobré mravy”). A court ruling of the Highest Court of the Czech Republic No.

22 Cdo 1993/2001 from 08. April 2003 defines usury contracts (in Czech: “Lichevní

smlouvy”) as contracts which involve abusing the inexperience, intellectual weakness or

distress of a party to the contract, and whereby the contract arranges for the other

party, or others, to be provided or promised to be provided with a performance which is

in gross disproportion to the mutual performance. Usury contract under civil law is a

contract where the party knew or should have known from the circumstances of the case

that the other party is affected by the circumstances mentioned above, and utilized this

fact. It is not required for this conduct to have been simultaneously identified as an

offence in criminal proceedings. Usury contracts are null and void. A court ruling of the

Highest Court of the Czech Republic from 15. December 2004, No. 21 Cdo 1484/2004,

defines when an interest rate is considered to violate good morals and thus a credit

contract as null and void (in Czech: “neplatný”). This is when the interest exceeds

quadruple of an ordinary interest provided by commercial banks for the given credit type

(a violation of good morals results in principle in nullity of legal action).

42 iff/ZEW – Final Report on interest rate restrictions in the EU

This matter is also covered by the current criminal law (the Czech Penal Code, Act No.

40/2009 Coll., as amended): Anyone who arranges for themselves or others to be

provided or promised to be provided with a performance which is in gross disproportion

to the mutual performance, or if he enforces or transfers the receivables arising with the

intent to enforce them, commits a criminal act. This applies to cases when someone

takes advantage of another’s weakness, distress, inexperience, carelessness or

disturbance only. Higher penalties are awarded to those who acquire for themselves or

others considerable profit or a profit of significant amount, or who commit this crime as

members of an organised group, or who by committing this crime cause a state of severe

need to others, or if this crime is committed during the state of emergency or state of

war or during a natural disaster or other circumstances seriously threatening the life or

health of people, public order or property.

The Danish word for usury is: åger (aager). Usury is illegal in Denmark and therefore

the term is used to describe the criminal action. During the financial turmoil there has

been a tendency to use the term more generally for high interest rates.

In Estonia “Usury” (liigkasuvõtmine) is not a legal term. In public discussions it refers to

the current activities of many moneylenders, which offer easy ways of borrowing money

(either through mobile phone, internet) but which charge unreasonably high interest

rate. Legally it would fall under good morals a principle of the General Part of the Civil

Code Act (Tsiviilseadustiku üldosa seadus).

In Finland usury means generally to take advantage of a weaker person’s situation.

Legally it is in the penal code and refers to abusively high interest rates of a loan. Usury

means either charging very high interest because of the borrower’s weak situation or

charging in general such high interest that it is disproportionate compared to the lender's

risk and costs.

In France usury describes excessive interest rates. Any contractual loan granted at an

annual percentage rate which, at the time of its granting, is more than one third higher

than the average percentage rate applied by the credit institutions during the previous

quarter for loans of the same type presenting a similar risk factor, constitutes a usurious

loan (L313-3 of consumer code).

The German word for usury is “Wucher”. It is used in the German civil code, section 138

(2) as a special example for a breach of “good morals”: (1) A legal transaction which is

contrary to public policy (public morals) is void. (2) In particular, a legal transaction is

void by which a person, by exploiting the predicament, inexperience, lack of sound

judgement or considerable weakness of will of another, causes himself or a third party, in

exchange for an act of performance, to be promised or granted pecuniary advantages

which are clearly disproportionate to the performance.” In the German Criminal Code,

section 291 states: “Usury”: “(1) Whosoever exploits the predicament, lack of

experience, lack of judgment or substantial weakness of will of another by allowing

material benefits to be promised or granted to himself or a third person 1. for the rent of

living space or additional services connected therewith; 2. for the granting of credit; 3.

for any other service; or 4. for the procurement of one of the previously indicated

services, which are in striking disproportion to the value of the service or its

procurement, shall be liable to imprisonment of not more than three years or a fine.

The Greek definition of usury is in the Criminal Code (art.404) which since Greece has

taken the German BGB as the basis for its legislation is identical with the above cited

Article 291 of the German Criminal Code.

In Hungary the concept of usury is based on a Calvinist interpretation as profit from a

neighbour's loss, rather than profit from a loan. This is why usury concerns a party’s

excessive benefits through exploitation of another party’s situation.

iff/ZEW – Final Report on interest rate restrictions in the EU 43

The Irish Consumer Credit Act 1995 section 45 makes reference to where credit charges

are “excessive”. However, this provision remains largely untested in the courts. In any

event, it does not apply to “credit institutions” as defined in section 2 of this act.

The translation of usury in Italian is usura. Usura is the activity of the one who lends

money and ask for an excessive interest. There is no legal definition of usury in civil laws

but it is covered by the definition of usurious interests. The criminal code describes the

crime of usura at art. 644, as to obtain, or to obtain the promise of interests or other

profits that are usurious under the law. It is specified that only law can determine when

interests are usurious, except the case of the so called usura impropria: in fact even the

judge can establish that interests (or other forms of benefits) are usurious if they are

disproportional to the situation considered, or if the debtor is in need or in economic

difficulty. When this is committed by a bank or another financial intermediary it is always

usury, and the sanctions are more severe.

In Latvia usury is called “Augļošana”. It applies to the situation were a person makes

loans, in whatever form, and knowingly takes advantage of the grave economic situation

of the borrower, to impose terms and conditions that are excessively burdensome for the

borrower. The applicable sentence is deprivation of liberty for a term not exceeding five

years, or custodial arrest, or community service, or a fine not exceeding one hundred

times the minimum monthly wage (Criminal Law Section 201).

In Lithuania usury “lupikavimas” is not legally defined and there are no restrictions on

interest.

In Luxembourg Article 1907-1 of the Civil Code states that in case the credit provider

abuses from the inexperience of the borrower in order to obtain from his/her the promise

to pay an excessive interest rate or any other obviously excessive benefits according to

the risk coverage, the Judge can lower the interest up to the reimbursement of the

principal plus the legal interest rate. Article 494 of the Luxembourg penal code states

that whoever had provided another person with any kind of values, to an amount rate

exceeding the legal interest rate, by abusing the weakness or the passion of the

borrower can be sentence to imprisonment (from1 month to 1 year) or to pay a fine from

€500 to €25,000). Apart from that article 1154 forbids anatocism applying for a period of

less than one year (exception to this rule: when this anatocism is performed on a current

account between a bank and its consumer).

In Malta usury is illegal but for professionals, banks and certain products many

exemptions apply from its rate ceiling. It therefore concerns mainly non-professionals

who prey on the desperation of other individuals who have gambling, drug or other

substance abuse problems.

In the Netherlands Woeker (usury) is not a legal term. Professional lenders need a

license and they risk legal sanctions if they charge more interest than the legal

maximum. In the public discussion about the benefits of selling linked products like

insurances which are not transparent to the borrower are labelled as usurious. Also

short-term credit with high costs are deemed to be usurious.

Poland prohibits usury for all transactions between persons, not only contracts between

professionals and consumers.

The Portuguese word for usury is “usura”. In social terms, usury is considered to be the

conduct of someone who charges extremely high interest rates. Legally, usury is slightly

more complex than that. On one hand, interest rates above the maximum allowed by law

44 iff/ZEW – Final Report on interest rate restrictions in the EU

(where there is any legal limitation) are considered usurious interest.29 On the other

hand, businesses where someone, with intent to achieve a pecuniary benefit for himself

or for another person, exploiting a personal situation of need, mental illness, incapacity,

incompetence, inexperience or weakness, causes the debtor to undertake to grant or

promise to grant, in any way, benefits which are manifestly disproportionate, excessive

or unjustified, may be voidable and give rise to criminal liability30.

In Romania, generally usury known as "camatarie" is the practice of charging much

higher interest rates than the market rate. In the Romanian Criminal Code31 the offence

of “usury” was provided in art. 450. It was defined as : a) the operations of lending

money or bonds as a profession by unauthorised persons, directly of through camouflage

papers, if the interest is higher than the interest established by the law; b) the

operations of lending money or bonds, carried-out by unauthorised persons directly of

through camouflage papers, if they establish an interest capitalization for interests owed

for a period of time shorter than a year. Usury is also defined by practices like interest on

interest (anatocism) and is also mentioned in connection with a maximum interest rate

or legal interest.

In Slovakia, usury is a criminal offence according to art. 235 of The Act No. 300/2005

Coll. The Criminal Code says usury is committed by the one who, abusing anyone's

pressure, inexperience or rational weakness or any distress, receives (himself or for

anyone else) a performance or a promise, and its value is in gross disproportion to the

value of mutual fulfilment. Furthermore, usury is committed by the one who transfers

such a claim with the purpose to exercise it or who exercises such a claim. To pay more

than 30 percent for a loan, including fees for the year, was seen as usurious, according

to a decree of the Ministry of Justice of Slovakia.

In Slovenia, usury (“oderuštvo”) describes a practice of taking advantage of someone’s

position by charging too high interest rates or by setting too high price. A usurious

contract (“oderuška pogodba”) in private law is a contract, where one party takes

advantage of the other party’s emergency, difficult material position, his lack of

experience, frivolity or dependence and assures to himself or to someone else a benefit

that is in obvious disproportion to his own obligation (Article 119 of Code of Obligations;

Obligacijski zakonik, OZ-UPB1, OJ 97/07). In criminal law usury (“oderuštvo”) means the

taking or assuring of a disproportionate benefit by taking advantage of weak party’s

difficult material position, lack of housing, need, lack of experience or frivolity (Article

214 of Penal Code; Kazenski zakonik, KZ-1, OJ 55/08).

In Spain usura describes abusively high interest rates. Interest rates to consumer loans

are – generally speaking - free, subset to the agreement among the parties. However,

Case Law and some specific Acts – namely the Usury Act of 1908 - establish that, the

freedom to agree upon a certain level of interest rates should be analysed in the light of

a number of issues like the relation between rates set in contracts and legal cost of

money (legal interest rate), market practices, whether contractual or default interest.

The contractual clause can be qualified as abusive clause in the sense of the Abusive

Clauses Directive if the contract is concluded in relation with urgent necessities of the

consumer. Failure to pass the test of the above criteria, may lead to the contract been

declared void.

Usury in Sweden is regulated in penal law as well as in private law and both comprise all

sorts of claims. The Swedish Penal Code (Brottsbalken) chapter nine section 5 paragraph

29 Cf. article 1146 of the Civil Code – only for non banking operations - and article 28 of Decree-Law nr.

133/2009.

30 Cf. article 282 of the Civil Code (“Código Civil”) and article 226 of the Criminal Code (“Código Penal”).

31 Criminal Code of 28 June 2004, published in the Official Gazzette, Part I no, 575 of 29 June 2004.

iff/ZEW – Final Report on interest rate restrictions in the EU 45

2 states: “… A person shall also be sentenced for usury who, in connection with the

granting of credit in a business activity or other activity that is conducted habitually or

otherwise on a large scale, procures interest or other financial benefits which is

manifestly disproportionate to the counter-obligation. If the crime is gross, imprisonment

for at least six months and at most four years shall be imposed.” In private law, usury is

regulated in Contracts act (Avtalslagen 1915: 218) section 31 paragraph 2. Usury arises

when a party unjustly exploits a contract situation, for example where the other party is

in desperate straits, is in a dependent relationship or more vulnerable due to a lesser

understanding. In addition an obvious imbalance must exist between the performances

under the contract, for example, that the services rendered are not worth the amount

charged. Usury can be seen in contrast to duress and fraud, where the wrongful party

created the situation. In the usurious context, the wrongful party exploits an already

existing situation.

In the United Kingdom usury is commonly taken to mean lending at excessive or

exorbitant interest, but it is not a term defined in legislation and there are no legal

sanctions for the behaviour provided the lender is licensed by the regulator. However, a

ceiling on the permitted level of interest that can be charged by Credit Unions in the UK

does exist, and courts have the power to intervene in consumer credit contracts where

the relationship between borrower and lender is considered to be ‘unfair’.

1.1.3.1 “Prohibition of Usury”

Usury exists as a legal concept in the criminal and/or the civil codes of twenty-one

Member States:

Denmark, Finland, Latvia, Romania, Slovenia, and Malta have incorporated usury

within their criminal codes.

Estonia, France, Hungary, Bulgaria and Spain have incorporated usury within their

civil codes.

The Czech Republic, Belgium, Austria, Germany, Greece, Portugal, Slovakia,

Poland, Sweden, and Italy have incorporated usury within both their criminal and

civil codes.

As a legal concept, usury commonly refers to the exploitation of another person’s need,

inexperience or weakness, for personal gain that is disproportionate, excessive or

unjustified, although the precise position varies between states. For example, although

no specific mention is made of the term usury in Luxembourg law, both its civil and

criminal codes do contain provisions that prevent the charging of excessive interest as a

result of the weakness of the borrower. In contrast, the law in Spain prohibits usury but

does not define it. In Lithuania it is also possible that protection similar to usury law is

afforded by a general prohibition of abuse of a person’s rights contained in its civil code.

Penalties for the offence of usury within criminal codes vary. For example, Austria

provides for the imprisonment of usurers for up to three years, whilst Latvia provides for

terms of imprisonment of up to five years, community service, or a fine not exceeding

one hundred times the minimum monthly wage. In practice, however, some of the usury

legislation is not considered to be effective. For example, case law in Denmark has

indicated that the courts are reluctant to intervene and that only an extremely high rate

of interest would qualify as usury.

There are also variations in approach in respect of usury laws contained in civil codes.

Some countries provide for usurious contracts to be voided in their entirety (eg.

Hungary), while others provide only for a reduction in the interest rate payable (eg.

Luxembourg).

46 iff/ZEW – Final Report on interest rate restrictions in the EU

1.1.3.2 Other legal terms

Aside from the incorporation of usury into legal codes, few Member States have adopted

legal definitions for other terms such as ‘sub-prime’, ‘last resort loans’, ‘fringe lending’

etc., even though these are often in common use in their countries more generally.

The exceptions to this are:

Belgium, which defines ‘sub-prime’ lending in respect of secured credits as “a

credit agreement given to persons, whose income is not sufficient to repay the

credit, but who receive the credit due to the supposed increase of the asset value

of their immovable property”.

Romania, which defines mainstream credit (‘credit de larg consum’) as a loan

granted for purchasing mainstream goods; and which defines moneylenders

(‘camatar’) as people offering loans with very high rates of interest compared to

the interest rates established by law.

Legal definitions of moneylenders are also present in Bulgaria, Ireland, Poland,

and Spain.

In the UK, the terms sub-prime and non-status lending have been used by regulators

when conducting market investigations and setting out guidance for use by lenders in

those market segments, and it is also possible to find case law which sub-divides the

mortgage market into ‘primary lenders’, ‘secondary lenders’, and ‘tertiary or non-status’

lenders.32

1.1.3.3 Historical impact

All these forms seem to reflect one single historical objective: the restriction of interest

rates on credit as outlined in the historical review which follows below. However, as this

overview reveals, these forms have quite distinct historical roots and serve different

purposes even today.

Exploitation: Most countries have a legal term known as “usury” in both criminal

law and civil law. The wording of Article 138(2) of the German Civil Code (BGB) is

typical. It identifies usury as the intentional exploitation of a weak person or

situation to make excessive profits. Such models can be found in Germany,

Austria, the Baltic as well as the Scandinavian states, in Greece, the Netherlands

and Slovakia.

High interest: A second form of IRR can be found in Romance countries such as

Portugal, France, Belgium, Spain but also in Slovenia, Czech Republic, Slovakia,

Hungary, Ireland, the UK and in German case law. Instead of exploitation this

form of usury falls under civil law and relates only to the excessive level of the

interest rate charged, in comparison with the average market rate.

Anatocism: Romania and Luxembourg use the word usury in a more traditional

sense, describing conduct in which ancient rules governing the calculation of

interest calculation, and specifically anatocism, have been disregarded.

Criminal lending: In Italy, Malta, Estonia and Denmark the meaning of usury

seems to be reversed. All illegal credit activities are styled usurious, including

32 Lord Justice Dyson in Broadwick Financial Services Ltd. v. Spencer [2002] 1 AER (Comm) 446.

iff/ZEW – Final Report on interest rate restrictions in the EU 47

illegal money lending, loans from unauthorised persons, and dubious lending

procedures.

West's Encyclopaedia of American Law summarises the historical development of usury

regulation as follows: “The charging of excessive interest in exchange for a monetary

loan has been considered reprehensible from the earliest times. Chinese and Hindu law

prohibited it, while the Athenians scorned persons who charged more than a moderate

rate of interest for a loan. The Romans at one time abolished the practice of charging

interest. Although they later revived it, the rates were strictly regulated.

During the Middle Ages in western Europe, the Catholic Church censured usurers, and

when they died, the Crown confiscated their lands and property. In England, until the

thirteenth century charging any interest was defined as usury. As commerce and trade

increased, however, the demand for credit grew, and usury was redefined to mean

exorbitant interest rates. In 1545 the English Parliament set a legal maximum interest

rate. Charging higher interest constituted usury…. Organized Crime has traditionally

relied on loan sharking as a source of income.” 33

The following chart has been constituted from literature and the answers of the experts

consulted, and shows the fluctuations in the use of IRR over the course of the 20th

century.

33 See West's Encyclopedia of American Law on “Usury” http://legal-dictionary.thefreedictionary.com/usurer.

48 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 2: History of IRR in the EU Member States

19th

Century 1900-39 1940-69 1970-89 1990's 2000-04 2005-10

Today

mid2010

Future

post-

2010

Austria -

Belgium I ceiling ch

Bulgaria -

Cyprus -

Czech Rep - I

Denmark - I

Estonia I ceiling

Finland -

France I ch ceiling ch

Germany I R I ch ceiling

Greece I ch ceiling

Hungary I R -

Ireland I ch ceiling

Italy I ch ch ceiling

Latvia -

Lithuania I R -

Luxembourg -

Malta I ch ceiling

Netherlands I ch ceiling

Poland I ch ceiling

Portugal I R I ceiling

Romania I ch ch R -

Slovakia I ceiling R

Slovenia I ch ceiling

Spain I Overdrafts

Sweden -

UK I R I Credit Unions

Source: Expert survey and various literature. Note: I=introduction, R=removal, ch=change in IRR method.

Answers in the “post-2010” column refer to those Member States where discussions are ongoing and where a

potential change could take place. No noteworthy history or current discussion on ceilings in: AT, BG, CY, FI,

LV, LU, SE.

From this we can identify the following five historical waves of IRR:

Ban on credit: The oldest form of IRR is a ban on money interest on arrears.

Examples are mentioned in the Bible, the Koran, and the doctrine of Buddha and

other religious figures. The charging of interest as such is criminalised. These

restrictions appear to have reflected the concerns of agrarian and pre-commercial

economies in which trade and money was seen as a threat to the agricultural

basis of society. Aristotle still thought that taking interest for lending money was

theft and should be punished. It should come as no surprise that interest on

interest (anatocism) was regarded as the worst form of unearned benefits,

especially where this was charged to people already in default.

Regulated interest: The historic development from agriculture to trade, industry

and finally to the money (credit) society has gradually abolished these ancient

attitudes, although they were still upheld by Canonical law until the end of the

19th century. But relics are still discernible in the rules on anatocism in contracts

(ie. Article 248 BGB) and default (ie. Article 289 BGB), as well as in a number of

legally determined interest rates (ie. Article 246 BGB (4% p.a.)).

Supervised Moneylenders: Removal of these barriers to interest in order to

facilitate commerce and banking resulted in a side effect in the form of the

iff/ZEW – Final Report on interest rate restrictions in the EU 49

emergence of usurious moneylenders, as described in the works of Dostoyevsky.

While some countries reacted with a general bank monopoly on credit (ie. France,

Germany, Netherlands, Italy), other states developed supervision and admission

rules which legalised private moneylending while often subjecting it to specific

restrictions in terms of the level of interest rates (eg. the UK Moneylenders Act,

1927).

Productive Credit Policies: In the thirty years that followed the 1929 crash and

the economic depression, the Keynesian idea that consumers should save money

that industry could then borrow to support growth became a cornerstone of

economic activity. Consumer credit, on the other hand, was seen as a threat to

overall economic productivity for which savings were needed as a source of

investment in production. Credit for consumption was seen as destructive. Stricter

supervision also resulted in a number of IRR which were gradually removed when

the money supply was liberated and consumption was identified as a motor of

economic development from the late 1950’s onwards (when, for example there

was a significant and progressive relaxation of restrictions in the UK, and personal

loans for consumption purposes were introduced at this time) and particularly

following the Keynesian crisis of the 1970’s.

Consumer Protection/Prevention of Over-indebtedness: From 1970

onwards, the explosion of consumer credit and the corresponding phenomena of

life-long debt and over-indebtedness stimulated a public debate about credit and

debt. Catholic countries in particular argued that less credit for the poor would be

better while Protestant cultures favoured easy access to what they considered an

essential service. Meanwhile, the principle of “responsible lending” and restricting

access to credit in some circumstances has been established. Rate caps are seen

as part of this.

To adapt these findings to the different legal systems we have to take into account that,

in civil law in particular, the same rules may persist but will have changed their purpose

over time, which also in turn changes the way in which they are applied. Many of the

ancient rules which tried to ban or at least restrict interest have been revitalised for

consumer protection.34 This is why, for example, rules which once served to ban interestbearing

credit altogether have been rediscovered as rules to limit over-indebtedness and

poverty, especially in Romance countries. Furthermore, ancient rules in relation to

exploitation are being revitalised as a means to cap interest rates, particularly with

regard to consumer credit. As such old ideologies still persist and sometimes are upheld

as a moral background, the true functions of this legislation may sometimes be obscured

and excluded from the debate. In any event, the revitalisation of IRR within the last 30

years in Europe and more recently in America shows that not only do they have an

economic impact, but their moral and cultural significance should also be taken into

account.

Based on the questionnaire responses received from stakeholders and details from the

legal survey, Table 4 shows some of the main reasons identified for which Member States

have decided to introduce interest rate restrictions.

34 As an example for the principle of anatocism see Reifner, U (1992), pp 227-343.

50 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 4: Main reasons for introducing IRR

MS Main reasons for introducing IRR

With IRR

ceilings

Belgium Consumer protection to prevent excessive rates and to prevent

excessive volatility of variable rates in order to protect customers.

Estonia Consumer protection, the need to control over-indebtedness and mainly

to prevent SMS loan providers from collecting unreasonably high

interest rates.

France Other than the interest rate ceilings in the 19th century that were

abolished in civil transactions in 1918, France has had interest rate

ceilings from 1935. However they were modernised in 1966 in order to

prevent abusive practices due to the market power of French banks.

Ceilings were used to prevent the development of predatory lending to

households in financial difficulty and were motivated by the problem of

over-indebtedness and unfair practices of the banks that do not give all

information needed to the consumers when they grant credit. They were

also introduced to balance the relations between consumer and lender to

prevent le lender from imposing conditions on the borrower at any price.

Germany When in 1976 interest rates in instalment loans dropped significantly the

press revealed that some specialist instalment banks, which were mostly

subsidiaries of well-known commercial banks or savings banks, still sold

instalment credit at the old high prices. This led notably to a court

decision by the Stuttgart Court of Appeals which actively publicised its

decisions based on the Roman principle of the laesio enormis. When the

Supreme Court was finally seized of the question in 1981, public

pressure and support for the Stuttgart court was so high that the

Supreme Court sought a compromise rejecting the price control

approach of the Stuttgart court but introducing an unrebuttable

presumption that a bank abuses its power when selling overpriced credit

to inexperienced consumers.

Greece The Monetary Commission has issued a large number of decisions as far

as the essential banking activities, deposits and provision of credit were

concerned defining the interest rates, sometime providing a fixed

ceiling, sometimes providing for a minimum or a maximum rate.

Gradually, within the framework of the overall effort to liberate the

function of the financial market, banking interest rates are also freely

negotiable among banks and their clients either on the basis of decision

of the Bank of Greece either on the basis of the lack of any relevant

provisions. Such freedom led to abuses. This was the reason why the

Bank of Greece intervened again, introducing a restriction regarding the

default interest rate in the Act of the Governor of the Bank of Greece

2393/15.7.1996.

Ireland The interest rate ceiling that exists for moneylending in Ireland,

currently just below 190% APR, is a de facto ceiling in practice as it is

the rate at which the highest charging moneylender is licensed to trade

by the Financial Regulator (with annual applications for renewal of

licences). With regards to the ceiling for credit unions (1% p.a.) this is

largely to ensure the creation of sources of credit for the mutual benefit

of its members at a fair and reasonable rate of interest.

Italy The usury threshold has been introduced to prohibit credit where

providers can apply excessive interest rate to people with poor credit

histories and consequently to counter the crime of usury, which was

very widespread in the past.

iff/ZEW – Final Report on interest rate restrictions in the EU 51

Malta The maximum rate of interest has always been fixed by the Civil Code

since its promulgation in 1868.

Netherlands Controlling illegal financial activities, protecting consumers by

preventing the charging of excessive interest rates and decreasing risktaking

behaviour on the part of credit providers.

Poland To protect borrowers from excessive interest charges.

Portugal The main purpose of IRR in Portugal is consumer protection which arises

from two different types of reasons: i) ethical reasons; and ii) protection

of the weakest party. The recent APR usury ceilings on consumer credit

agreements (introduced following discussions concerning consumers’

rights in 2009) are related to both types of reasons and were

particularly established to avoid the charge of especially high interest on

agreements entered by consumers who are not entitled to obtain credit

in traditional banks. Likewise, the main goal of restrictions on unilateral

changes of interest rates and restrictions on rounding interest rates is to

protect the weakest party which is, in the overwhelming majority of

cases, the consumer. The main reason was the economic situation/crisis

and a concern to ensure that financial service providers did not distort

the market with absurd interest rates from some institutions at

unacceptable cost. The interest rates charged by credit institutions in

consumer credit exceeded 30% because they used predatory lending in

credit to expand the circle of debtors in order to compensate for rising

numbers of bankruptcies and competition in the credit market; this

reveals the 'reverse Robin Hood effect’ of lending to the poor to

maintain low interest rates for the wealthy. Excessive rates give rise to

further breaches, thus maintaining people perpetually hostage to

adverse credit scoring.

Slovakia Excessively high interest rates on consumer credit and the popular

perception that there are many people whose property is taken into

possession by loan sharks as a result of high credit prices, which they

can not repay.

Slovenia While rules on usurious interest rates are relatively old, the rules on

maximum APR for non-bank providers were introduced because of

usurious practices of these providers in the past. Adoption of the Euro

was also seen as a reason for revising the IRR regime.

Spain Consumer protection was considered in the past as necessary for

current account holders, who are in some cases unaware of the high

cost of occasional overdrafts (ceiling on overdraft only). The abuses of

financial institutions in a fully liberalised market, the ineffectiveness of

the supervisory bodies and the systematic violation of the rights of

consumers.

Without IRR

ceilings

Austria The in-built economic imbalance of suppliers and consumers (for rules

on default interest only).

Cyprus The objective of the existing “usury bill” is to extend the basic penal

code to criminalize the lending of money between private individuals at

an unreasonably high interest rate (usury executed by loan sharks). In

1999, the interest rate ceiling was abolished in order to allow full

liberalization of the interest rates in Cyprus in an effort to enhance

competition in the banking sector.

Czech

Republic

To bring more clients into the system, more stable deposits, less risky

loans and cheaper resources for granting loans (for the state subsidy of

building savings only).

52 iff/ZEW – Final Report on interest rate restrictions in the EU

Latvia To protect the average consumer and to reach a balance in the financial

and capital markets between industry and consumers.

Romania The Gov. Ordinance 60/2009 established a governmental fund to

support young people to obtain a loan to buy a home. The State

guarantees the credit offered by the bank and imposes a maximum limit

on the interest rate (though access to such a fund is difficult). In respect

of civil agreements, the limitation on legal interest was due to public

concern, but no economic modelling was produced, therefore it was

deemed a populist measure.

United

Kingdom

Default/risk-based re-pricing - these restrictions were introduced mainly

in connection with customers in financial difficulty. The Government

decided that restrictions would take into account circumstances where

defaults occurred. These provisions were agreed following consultation

with the lending industry and do not seek to restrict pricing beyond

default cases.

Source: Stakeholder survey.

In almost all Member States, the reasons mentioned above are still seen by stakeholders

in those countries as valid. Comments from the minority of stakeholders tending to

disagree with the original reasons for ceilings included “too restrictive regulation has

counterproductive effects such as closure of the market, less innovative products”

(Belgium) or that the credit market is much more competitive now than it was when the

legal usury rate was introduced (France). This same respondent nevertheless went on to

say “still, the financial crisis led the French government to consider that IRR are an

effective protection against excessive exposure to risk by lenders and borrowers. Indeed,

sub-primes and near-primes do not exist in France. The interest rate of revolving credit

in France is relatively lower than in other countries, though it approaches the usury

ceiling with an average APR of around 16-17%”. Similarly, another French respondent

mentioned that the current level of competition and state of the property market did not

make the debate on IRR particularly topical with regards to mortgage credit but

confirmed that, in the area of revolving credit (especially that combined with and

accessed through a bank card), the need for ceilings was much more pronounced

because of the particularly high cost of funds to lenders.

Table 5 gives some views on whether those reasons are still valid today:

Table 5: Some reasons why IRR is still valid today

MS Stakeholder

type

Reason

IT Provider

Association

We don't think that these reasons are still valid. In fact it has to

be underlined that the limitation on the cost of credit has been

ineffective in combating the illegal use of credit and has resulted

in an instrument that has prevented the development of the

credit market.

NL Other They are still valid; coverage will be extended to loans with

duration of less than 3 months, which have been exempt in the

past. The extension of coverage has been decided because

misconduct was apparent in relation to short-term loans.

SI Consumer

Organisation

They are very valid and call for further improvements. In the

new act on consumer credit, the above-mentioned rule will also

be valid for credit below the limits on maturity/amount of the

CCD, while membership fees" will now be part of the APR"

iff/ZEW – Final Report on interest rate restrictions in the EU 53

PL Other Positives of IRR regulations were discovered, and now it is well

established as pro-consumer regulation.

PT Other Leaving the markets unregulated means that the situation is still

very fragile.

RO Provider

Association

There is no evidence as to their effectiveness. Research shows

they are not effective; they drive illegal lending and restrict

access to regulated sources of credit. Furthermore, they

encourage loan sharks and credit into non-taxable

environments. Personal security issues also arise.

SK Provider

Association

We have never been provided with hard data from state

institutions or regulators as to the reality. We believe that there

were only tens of people suffering from a loss of property who

complained to state institutions. So there were and are no real

reasons for IRR regulations. What was really needed was the

regulation of illegal lending.

ES Consumer

Organisation

Some view ceilings as more valid than ever in the context of the

current crisis. Others consider that the maximum ceiling on

overdrafts does not vary and is not suited to the reality of credit

or to the reality of the Spanish family economy. This existing

ceiling will presumably be suppressed once the transposition of

the CCD 2008 is completed in Spain.

UK Provider

Association

The reasoning was reviewed extensively in 2009 after the credit

crisis but remained unchanged. In the current uncertain

economic climate there is no impetus to change the provisions in

place for default fees and when risk-based re-pricing can be

undertaken.

Source: Stakeholder survey.

1.1.3.4 Philosophy of regulation

The underlying philosophy of IRR may be reduced to five legal notions which roughly

correspond to three distinct goals:

(1) the ethical and religious concept aimed at preventing the exploitation of need and

weakness (“Do not exploit the poor because they are poor” Spr 22,22); (2) the market

concept, aimed at regulating prices where competition either does not suffice or where it

produces unwanted impacts on more vulnerable parts of society (“The poor pay more”

(Caplovitz); (3) unwanted credit products seen as detrimental for the national economy

(“Unearned bread” Eucken).

These aspects apply to all forms of IRR since the threat of circumvention of purely

interest rate-related regulations leads to a number of additional rules concerning

variability, compounding methods, annexed products, penalties and fees.

(1) Exploitation

o Individual usury proscriptions, which view high interest as evidence of

the exploitation of a weaker party. These exist in both private and criminal

law. It presupposes ill intention, knowledge of the situation of the debtor

and its active exploitation through exorbitant interest rates. These

regulations are, however, not applied to commercial consumer credit

providers, which target high risk groups with predefined interest rates and

which thus escape the definition of “exploitation”. It is a broad concept,

however, in that it encompasses all charges or methods of calculation, and

not interest charges alone.

54 iff/ZEW – Final Report on interest rate restrictions in the EU

(2) Market regulation

o Good Morals: at one time, individual usurious relationships were merely

restricted in some countries on the grounds of morality. This has now

developed into a principle of market evaluation. It provides objective rate

ceilings through court rulings in which, instead of individual exploitation,

lack of market strength on the part of certain groups in society leads to

pricing which operates as cartel practice. Like special regulation in other

countries, the average market rate is used as a competitive “as if price” ,

which can be compared to the contractual interest rate. A limit is set at a

margin above this rate. In Germany the Roman laesio enormis of double

the average market rate”, as is still cited in the Austrian Civil Code, has

been cited as a justification.

o Administrative rate ceilings: fixed by the Central Bank (France), a

ministry (Italy), or a royal decree (Belgium), such ceilings no longer have

any moral connotations. They simply assume that for certain types of

consumer credit prices have to be kept within a certain acceptable limit.

(3) Product regulation

o Specific rate ceilings: with regard to certain unwanted forms of credit

such as the limits on overdraft credit in Spain, or in order to make certain

forms of credit more affordable, for example pawnshop credit in Germany,

small business loans in France or loans from Credit Unions (12%) in

Ireland and the UK.

o Consumer credit prevention: administrative rules in force in the 1950s

were fixed by the central banks in order to curtail credit extension to

consumers. These were all abolished by 1990 and are no longer in force.

(Greece, Luxembourg, Germany, Italy).

1.1.3.5 Fairness and Good Morals – towards a general principle for IRR in

European contract law?

Since the reception of the Roman ius commune in all Europe, all countries have a certain

set of common legal foundations which also provide restrictions on the freedom to

contract. As far as these common foundations are concerned, it may be expected that all

EU Member States have some form of principles in private consumer or commercial law,

as well as in administrative and penal law restricting levels of interest.

The similarity of the way in which Roman law treated interest on money (“specialis”) to

the way it treated rent for things (“res”) and labour income (“operarum”), all of which

were attributed to one unified rent contract, the locatio conductio (specialis, operarum or

rei), suggests that existing principles governing wage and rent restrictions in all Member

States should resonate in consumer credit law.35 This is not, however, so. Tenancy and

labour law have been kept separate. Credit law is treated as an integral part of contract

law and, once an individual loan contract36 has been issued providing for interest, it is

governed only by restrictions common to all prices, and especially those in relation to

sale contracts.

35 See Reifner, U., Nogler, L. (2010), pp 365 - 407; Reifner, U., Nogler, L. (2009) pp 437–455.

36 In Roman law the loan contract was still split into two separate agreements one for the “free” loan and an

additional agreement (stipulation) for the interest.

iff/ZEW – Final Report on interest rate restrictions in the EU 55

Within the general principles of contract law, based on sales law principles, only few

exemptions to the principle of the freedom to fix of prices exist. These apply to goods of

first necessity, such as water, electricity and telephone. The setting of prices is seen as

the heart of a market economy and it should come as no surprise that EU Directives

governing general consumer and contract law refrain from making general price

restrictions which could also affect interest rates in consumer credit agreements.37

The notion of “unfairness” could be misunderstood as a basic principle which covers both

commercial practices and pricing. This is, however, only the case if the notion of

unfairness is used as in continental European law, where the notion of “good morals”

(gute Sitten, bonnes moeurs) or “good faith” (Treu und Glauben, bona fide) apply.

Fairness is a procedural category which does not affect the substantive elements of the

contractual exchange.38 This is clear in codified civil law and seems also to be the case in

common law, which the legal expert39 for the UK explained as follows:

“The Unfair Commercial Practices Directive 2005/29/EC (UPCD) is not used to control

costs or charges. This is dealt with through the prohibition on Unfair Relationship

provisions of the Consumer Credit Act. We are not aware of any attempt to use the UK's

implementing law in this way, but direct control of credit costs - as the question seems to

envisage - is not what the Unfair Commercial Practices Directive set out to achieve.

Instead this is a law aimed at outlawing practices that distort normal market function,

by, for instance, misleading the consumer. UK version of these regulations have been

used to impose some control on credit card default charges but not bank charges (which

were held not to be default charges and part of the 'price'.

The Unfair Terms in Consumer contracts Directive 93/13/EEC (UCTD) is implemented in

the UK by the Unfair Terms in Consumer Contracts Regulations 1999 (UTCCRs). The

UTCCRs protect consumers against unfair standard terms in contracts they make with

businesses. The FSA is one of several named bodies who can take legal action under the

UTCCRs to prevent the use of such terms. Our ability to control the cost of credit under

the UTCCRs is limited due to the exemption from the assessment for fairness of the

adequacy of the price in exchange for goods or services (Regulation 6(2)(b)). Following

the Supreme Court decision in the recent bank charges test case in the UK (OFT v Abbey

National plc and Others [2009] UKSC 6), it is clear that this exemption means that the

level of a charge cannot be assessed for fairness under the UTCCRs where that charge is

levied in exchange for goods or services; this means the UTCCRs can not generally be

used to restrain the cost of credit.

We understand that the current negotiations around the [draft] Consumer Rights

Directive could change the effect of the exemption in the UCTD (and consequently in the

UK implementing legislation), which could allow the level of a charge to be assessed for

fairness but only for charges that are not part of the essential bargain (ie. only those

charges which are not the main price). Even if this change were to happen, it is unlikely

that interest rates could be assessable for fairness, as interest rates are likely to be seen

as part of the essential bargain" of a contract, and hence not assessable for fairness.”

The question of fairness apart, prices are not totally exempted from restrictions under

contract law. The principle of good faith and good morals, which is partly referred to as a

37 See below 1.4 Effects of the CCD 2008 on IRR at p 111.

38 See for the procedural use of fairness in the Draft Common Frame of Reference III. – 1:103: Good faith and

fair dealing (1) A person has a duty to act in accordance with good faith and fair dealing in performing an

obligation, in exercising a right to performance, in pursuing or defending a remedy for non-performance, or

in exercising a right to terminate an obligation or contractual relationship.

39 See Annex V: Legal Experts for names of the country experts.

56 iff/ZEW – Final Report on interest rate restrictions in the EU

substantive fairness in Common Law40, incorporates the idea of the exploitation of

weakness which underpins the moral usury principle. “Unconscionable”, “excessive”,

“unfair” pricing41 is therefore the counterpart for what in codified civil law is seen as

contrary to “good morals” and more specifically to “usury”. With its principle of “laesio

enormis” it ultimately provides one single principle of IRR.

This principle appears to have lost its significance where individual exploitation is no

longer the problem. Instead, lenders are not reproached for systematically overpricing

products targeted at poor customers for the same reasons. Only Germany and Estonia

have used this individual principle and extended it to such systematic overpricing. Of

course individual usury remains a principle under all general contract law and it may

under extreme conditions also be applied to a specific credit contract if individual

exploitation can be alleged.

What in Germany and Lithuania has been developed from the legal doctrine of

“Individualwucher” to a legal doctrine of “Sozialwucher” can probably better be explained

by the two aspects of competition law: unfair competition and cartel law. While unfair

competition is purely procedural, cartel law is focussed on the mechanisms of price

generation in a competitive market as laid down in Articles 101 ff Lisbon Treaty. Under

Art. 101 (1) (a), “purchase or selling prices” are subject to restrictions if competition has

been excluded intentionally by consent. But the common justification for interest rate

ceilings and objective applications of IRR refer to the systematic misuse of power in

certain market segments, where vulnerable consumers pay high interest placing them at

risk of overindebtedness. This comes closer to the idea of Article 102 (ex Article 82 TEC),

according to which “any abuse by one or more undertakings of a dominant position

within the internal market or in a substantial part of it shall be prohibited as incompatible

with the internal market in so far as it may affect trade between Member States. Such

abuse may, in particular, consist in: (a) directly or indirectly imposing unfair purchase or

selling prices or other unfair trading conditions”. Similar prescriptions exists in all

national cartel law.

This principle does not, however, seem to be applied to IRR. Instead, one case was

reported from Italy in which anti-trust law was invoked to void high interest rates for

overdraft credit. Unlike typical antitrust cases, the lack of competition cannot be

attributed to a mono- or oligopolistic offer structure, nor does it systematically affect

general competition in these markets. Consumer credit markets are highly competitive.

The existence of uniformly high prices close to the ceiling in revolving small credit, as

condemned in the French government report42, is more of a problem in terms of

borrowers’ behaviour. It is the borrower who accepts disadvantageous prices or

additional costly products because he or she does not seem to be able to make proper

decisions in either -refraining from borrowing or in the choice of a more appropriate

product. These problems are typical for consumer law but not for cartel law.

With the increase of interventions into the pricing of consumer credit for low-income

consumers, a general principle is about to develop in which the ideas of “basic services”

and of “price cartels” merge into something that could be called a compensatory pricing

mechanism for fair competition.

The question has now been decided by the European Court of Justice (EJC Dec. of June 3,

2010 C-484/08) in the case Caja de Ahorros y Monte de Piedad de Madrid v Asociación

de Usuarios de Servicios Bancarios (Ausbanc) decided first by the Supreme Court of

Spain on the question whether the Spanish law on general contractual conditions can

40 Reifner, U. (1999) pp 269 ff.

41 See for example Financial Services Authority (2009).

42 See IGF/IGAS (2009).

iff/ZEW – Final Report on interest rate restrictions in the EU 57

declare prices as unfair in the sense of Directive 93/13/EEC and with regard to Articles 2

EC, 3(1)(g) EC and 4(1) EC.

Advocate General Trstenjak in his opinion on this case delivered on 29 October 2009

(62008C0484) provides the following information on the Spanish law:

“8. Article 10a(1) of General Law 26/1994 of 19 July 1984 for the protection of

consumers and users (Ley 26/1994 general para la defensa de consumidores y

usuarios), which was added by Law 7/1998 of 13 April 1998 on general

contractual conditions, provides as follows in relation to the definition of unfair

terms: ‘All those terms not individually negotiated which, contrary to the

requirement of good faith, cause a significant imbalance in the parties’ rights and

obligations arising under the contract, to the detriment of the consumer, shall be

regarded as unfair terms. In any event, the terms listed in the additional provision

of this Law shall be regarded as unfair.”

In his decision of June 3, 2010 the ECJ concluded in a way which allows the inclusion of

IRR into the fairness principle of the EU-Directives on a national level:

“1. Articles 4(2) and 8 of Council Directive 93/13/EEC of 5 April 1993 on unfair

terms in consumer contracts must be interpreted as not precluding national

legislation, such as that at issue in the main proceedings, which authorises a

judicial review as to the unfairness of contractual terms which relate to the

definition of the main subject matter of the contract or to the adequacy of the

price and remuneration, on the one hand, as against the services or goods to be

supplied in exchange, on the other hand, even in the case where those terms are

drafted in plain, intelligible language.

2. Articles 2 EC, 3(1)(g) EC and 4(1) EC do not preclude an interpretation of

Articles 4(2) and 8 of Directive 93/13 according to which Member States may

adopt national legislation which authorises a judicial review as to the unfairness of

contractual terms which relate to the definition of the main subject matter of the

contract or to the adequacy of the price and remuneration, on the one hand, as

against the services or goods to be supplied in exchange, on the other hand, even

in the case where those terms are drafted in plain, intelligible language.”

The Spanish example is not isolated. In relation to the Unfair Commercial Practices

Directive 2005/29/EC (UPCD), some experts identified indirect interest rate restrictions,

including unfair changes of interest rates (UK), unethical business practices (Hungary),

illegally high interest rates on default and in provisions preventing legal proceedings

against a debtor when he has reached more than the accrual of arrears of three

instalments.

1.1.4 Effectiveness

IRR are regulated under private or administrative law, have penal or administrative

sanctions, lead to reduced claims on interest and are enforced either by special

authorities, the attorney general or the civil courts. Our chapter on existing legal

sanctions shows the greatest diversity, and this aspect should not be under-estimated. In

sociology of law there is a distinction between the law on paper and law in action.

Theodor Geiger43 goes so far as to assume that legal rules which are not effective are not

in fact part of the law. The implementation of a legal rule therefore depends on the way

sanctions are structured and whether they have the ability to threaten the addressee

effectively. It also requires accessible institutions sufficiently equipped to prosecute

43 Geiger, T. (1964), p 44.

58 iff/ZEW – Final Report on interest rate restrictions in the EU

circumventions of the law. Because in private law consumers have to mobilise these

sanctions, the effectiveness of the law also depends on access to the existing courts, the

incentives to sue, and the expectation they have that the court will act in a reasonable

time at reasonable transaction cost.

Moreover, administrative rules often lack effectiveness due to poor staffing of the various

institutions. This is why sanctions that give consumers an incentive to sue their creditors

or oppose their enforcement activities in court can sometimes be much more effective. If

for example in Germany and Belgium a lender has higher interest rates than legally

permitted he loses all the interest, while in the Netherlands only the element of the

interest which is above the ceiling is void. But additional penal or administrative

sanctions will again change the picture of effectiveness.

Finally, only countries with high levels of banking supervision, which are subject to

ongoing duties to provide information on the loans they provide and have a legal

monopoly over consumer credit provision, are in a position to produce valid data that

could enable sound interpretation of economic effects. Even when reliable data exists,

there will nevertheless be a need for assumptions to be made. This was evidently the

case in the Policis report on Germany44, where the significant market for illegal lending is

hard to assess. The research on crime statistics45 indicates the, often insurmountable,

difficulties that exist in this respect.

Another important factor is the applicability of the IRR. If, as we have seen for Malta, the

rules have many complex exceptions, consumers will probably not be able to apply the

law to their individual case. This may also be true for ceilings that are fixed on a case-bycase

basis by the court system as in Germany and Hungary, and where the outcome of

the case is not predictable. Administratively fixed and adjusted interest rate ceilings such

as those that exist in France may also be easier to publicise than ceilings that are fixed

by law as a proportion of market rates, such as those that still currently exist in relation

to overdraft credit in Spain.

The legal experts either underlined that IRR should be made effective or indicated that

IRR had been “effectively” incorporated into the body of law. From their answers we

therefore cannot derive any conclusion on how effective these regulations are in general,

since this information is normally dealt with in economic or social science, while lawyers

understand effectiveness mostly as enforceability.

Perhaps some insight can be derived from the opinions of all those interviewed experts,

consumer organisations, providers and regulators. They were asked to assess the

effectiveness of the existing IRR on a scale between 1 (not effective at all) and 5 (very

effective) for their country.

One might have assumed that since these opinions are purely subjective, regulators

responsible for effectiveness and providers, who take a more critical view of IRR would

tend to overstate effectiveness while consumer organisations would have more doubts.

Instead the average ranking consumer organisations gave to the effectiveness of the

relevant IRR was even slightly higher than that of regulators and government officials.

Only in the Netherlands, Austria, Finland and Poland did answers differ by more than two

2 grades, while for example in Cyprus, Italy and France the different respondents

provided identical replies.

44 See: Policis (2004b)

45 Eg. Helfgott, J.B. (2008), pp 13 ff.

iff/ZEW – Final Report on interest rate restrictions in the EU 59

The following ranking should be taken with caution. Only half of the respondents gave

this rough evaluation and they are not representative by country. It gives, however, at

least an indication that countries with strict IRR are seen as having the most effective

regulations.

Table 6: Member States by effectiveness of IRR

Member State Average

grade

France, Belgium, Portugal 4.5

Cyprus, Denmark, Finland, Italy,

Netherlands

4.0

Poland, Austria, Bulgaria, Czech

Rep, Estonia, Latvia, Lithuania,

Slovakia, Slovenia, Sweden,

Romania

3.0

Spain, Germany, Greece, Hungary,

Luxembourg, Malta, UK

2.1

Ireland 1.0

Source: Stakeholder and Expert survey.

The propositions with regard to IRR give some hints as to existing cultural and economic

differences as well as to existing problems with IRR. We have refrained from sorting

them according to the different groups.

The scope of application has been recently extended to small loans with a term of less

than three months and the interest rate ceiling has been lowered recently in the

Netherlands as a response to the ineffectiveness of the previous one. Similarly a Latvian,

a Lithuanian and a French respondent argued that the system should be changed from

product-specific ceilings (revolving, instalment credit, etc.) to a ceiling related only to the

amount of credit.

French, Czech and German respondents criticised the fact that the price of the loan does

not include all possible costs. The interest rate ceiling should have a clear relationship to

the market price.

Better supervision was wanted in Hungary and Belgium. Supervision should include

regular testing. A stricter licensing system instead of interest rate ceilings could

overcome illegal lending in Italy. In France, it was suggested that a register of credit

agreements would be better than the existing rate ceilings.

Some respondents asked for a focus on factors which are not sufficiently taken into

account. In Italy, the charges made by debt collection agencies, in Ireland and Lithuania

mortgage loans and variable interest mortgage rates were mentioned. It was felt that

traders should be more closely monitored.

Some Italian and French respondents asked for a complete removal of IRR, others, for

example in Hungary, assumed that their markets are either not prepared for them or find

them unnecessary. Respondents in the Czech Republic, Ireland, Finland and Lithuania

would favour their introduction.

60 iff/ZEW – Final Report on interest rate restrictions in the EU

In conclusion, the effectiveness of IRR regulation depends largely on the existing national

culture of law enforcement and the degree of bank regulation which leaves more or less

space to high-priced credit. In general small loans, revolving credit and variable interest

rates are seen as a problem. The choices for effectiveness lie between stricter

supervision and private law approaches. Harmonised forms of IRR would probably have

to take questions of enforcement and sanctioning into account in order to achieve

comparable results with similar rules in all countries.

It also makes clear that the mere existence of IRR rules cannot be related to the

questions of access to credit without taking into account their effectiveness. For a deeper

insight, an in-depth study of two countries with opposing systems could help to clarify

what the most effective forms of regulation could be.

iff/ZEW – Final Report on interest rate restrictions in the EU 61

1.2 Direct IRR

An ordinary credit contract often has two lives: the contractual state where the parties

have set prices and terms, and the post-contractual state where a credit is in default

after cancellation. The applicable interest rates between both periods may differ: the

contractual interest rate and the default rate. While the first is at the core of the principle

of freedom of contract, for the second, most legal orders either provide a legal regime or

at least regulate these rates since the assumption that they have been freely convened is

weaker even where this has been done in advance.

A further clarification is also necessary. The CCD 2008 distinguishes between the APRC

for price disclosure and the borrowing rate for interest calculation. This distinction is due

to the fact that the APRC does not allow for a comprehensive calculation of interest (it is

in fact not an interest rate but a growth rate (Interest = C1 – C0; C1 = C0*(1+i)^t) while

on the other hand the borrowing rate which allows for a much easier form of interest

calculation (Interest = C0 * i * t) does not provide for a correct representation of price

relevant factors like different compounding periods or the inclusion of upfront fees and

charges. Both forms are therefore in use. Though they do not lead to much differences in

non-bank and short-term credit, it should nevertheless be kept in mind that historical

interest rate ceilings, especially absolute interest rate ceilings or IRR designed for nonbanks

where computers have been or are still not available, use the borrowing rate. This

is particularly the case where the “legal interest rate”, which is no IRR but a rate which is

applied in such cases where no interest rate has been set by the contracting parties, is

still regulated in the form of a borrowing rate. This is also true for all regulated default

interest rates (for which the CCD 2008 also fails to prescribe the form of an APRC). On

the other hand, modern IRR in the form of a relative ceiling for contractual interest rates

in loans extended by professional money lenders, use the form of the APRC which is less

prone to circumvention. In the following text we have tried to clarify this by using the

distinctions of the CCD 2008.

1.2.1 Contractual interest rate ceilings

1.2.1.1 Types of ceilings – Which Member States and at what levels?

As shown in Table 7, in thirteen member states there are no ceilings in place that limit

the amount of contractual interest that can be charged on typical credit agreements46.

However, the remaining fourteen states do have some form of ceiling for this purpose.

Of these, three Member States use absolute ceilings (fixed nominal rate caps) and eleven

Member States have relative ceilings (ie. the ceilings are calculated in relation to a

variable such as the average market rate or base rate), although Estonia and Germany,

have only de facto ceilings.

There are four different institutions which can be involved, either alone or in combination

with another institution, in fixing the maximum interest rates: the Central Bank,

Government Administration, Legislator or Courts. While in Italy, Portugal and Malta the

legislator fixes rate ceilings, in France, Belgium, Estonia and Poland (Lombard rate) it is

the central banks that fulfil this task. Likewise, in Greece and Spain, the central bank is

the core institution in so far as it fixes the legal interest rate upon which the IRR is

based. This is in contrast to Bulgaria or Ireland where the government and a special

46 In the UK, there is a limit on contractual interest rates that can be charged by Credit Unions, but these

constitute less than 3 per Cent of the total consumer credit market, and there are no ceilings in place in

respect of other types of lenders, so for the purposes of the classification, we have treated the UK as having

no ceilings.

62 iff/ZEW – Final Report on interest rate restrictions in the EU

administration is respectively in charge. Finally, in Germany, the courts fix such general

rate ceilings while in Hungary and the Czech Republic courts may fix individual limits.

Ceilings are set for a variety of periods. Poland and Germany set ceilings each month

whilst over a third of Member States do not determine the period at all and do not review

their ceilings at regular frequencies. Ireland reviews its ceiling on moneylender rates

annually, but Belgian ceilings have only undergone four revisions since 1992. However,

quarterly adjustments are the most common.

Differences also exist in respect of the number of decimal places that the APR ceiling

contains (for example, in Belgium this ceiling is rounded up to the nearest half a percent

whereas in Slovakia, two decimal places are shown).

Interest rate ceilings are, of course, only one form of IRR. In the remainder of this

report, we refer to interest rate ceilings irrespective of whether or not the ceilings are

absolute or relative. Table 8 on the following page thus sets out further details of interest

rate ceiling levels in operation across the EU as at March 2010 and shows that for

example high priced doorstep credit extended in the UK would face barriers to enter the

market in those countries where IRR exist (see Section 2.5.3 discussing hypotheses

H2a).

Table 7: Overview of interest rate ceilings in the EU

MS with contractual

IRR (absolute)

MS with contractual IRR

(relative)

No IRR

Greece*; Ireland;

Malta*

Belgium*;France; Germany;

Estonia*; Italy; Netherlands;

Poland; Portugal; Slovakia;

Slovenia; Spain*

Austria; Bulgaria; Cyprus;

Czech Rep; Denmark;

Finland; Hungary; Latvia;

Lithuania; Luxembourg;

Romania; Sweden; UK*

Source: Expert survey. *Notes: This table does not report on other IRR forms (such as default interest ceilings,

anatocism or specific limits affecting specific forms of credit such as state housing loans etc.). BE: Calculation

method for the ceiling specifies the adjustment mechanism based on a reference index (Euribor), but the initial

interest rate ceilings that served as a base were set by the government when the ceilings were first introduced

in 1992; GR: Ceiling is for non-banks only; MT: Subject to exceptions (enacted by Legal Notice 142 of 2009);

ES: Only concerns overdraft in current account; EE: Ceiling rates are published but no strict limits exist. We

have included Estonia in this category because we consider the ceiling as a defacto ceiling as it is based on a

similar court based system as that used in Germany); UK: Credit unions do face a ceiling but in view of their

very small share of lending, we have classified the UK as having no IRR.

iff/ZEW – Final Report on interest rate restrictions in the EU 63

Table 8: Interest rate ceilings in the EU as of March 2010

Member

State*

Maximum

APR in

Consumer

Credit

Average

Interest

Rate

Scope** Comments

Belgium 10-19.5% All Maximum APR vary according to the amount and the type

of credit (12 categories). From 10% (Revolving Credit

Account with no credit card for more than €1250) to 19.5%

(for instalment loan agreements less than €1250). Credit

term is not a discriminatory factor anymore.

Estonia 63.9% All Maximum APR equals 300% of average APR computed by

central bank (21.3%*3 =63.9%). Reset Monthly.

France 5.72%-

21.63%

8.12%-

16.6%

All Maximum APR equals a relative 133% of average APR

computed by the Central Bank every 3 months depending

on credit type (pending reform) and amount: from

5.72% (variable rate credit for home purchase) 21.63 %

(Consumption credit for < 1524 €). Separate ceiling

category for mortgage credit.

Germany 8.18%-

16.4%

4.09%-

8.2%

All Maximum APR equals 200% of average APR computed by

the Central Bank depending on credit type and term:

from 8.18% (new business/housing loans with an initial

rate fixation of >5 years and <10 years) to 16.4% (new

business/consumer credit with an initial rate fixation of >5

years). A further ceiling condition alongside double average

is a maximum of 12 percentage points over average rates.

Greece Partly

6.75%

On non bank

credit only

Greece abandoned the system of setting limits by

administrative provisions in 1989. Hence, bank interest

rates are free. But their non-bank credit rates are limited to

absolute rate of 6.75% per annum.

Ireland Partly

187%

Moneylenders

only

Moneylenders are assigned maximum APR when getting

their annual license. Credit Unions cannot charge more than

1% per month (12.68% APR).

Italy 4.38%-

27.20%

4.38%-

16.97%

All Maximum APR equals a relative 150% of average APR

computed by Ministry of Economy and Finance every 3

month depending on credit type and amount : from

4.38% (variable rate mortgage) to 27.20% (other family

loans). There are 23 different ceilings.

Malta Partly

8%

Many

exemptions

All type of credits. Exemptions exist for banks.

Netherlands 15% Not mortgages.

Credit <€200

(to be included)

Maximum APR ceiling is equal to 12%pts + the legal

interest rate (non contractual interest in case of default. As

of march 2010: 3%). Legal IR reset every 6 months.

Poland 20% All Maximum borrowing rate equals to 400% of the Central

Bank Lombard Rate (5%, monthly reset). No distinction

made by category or size of credit.

Portugal 6.7%-

31.6%

5.03%-

23.75

Maximum APR equals to 133% of the average APR

computed every 3 months by Central Bank depending on

credit type only. From 6.7% (Instruction, health and

renewable energies credit) to 31.6% (Credit cards, credit

lines, bank current accounts and overdraft).

Slovakia 11.52%-

79.8%

Maximum APR equals 200% of the average APR for the

type of consumer credit in question (30 ceilings by type,

amount, term), and at the same time it cannot not

exceed the amount equal to 400% of the weighted average

APR for all types of consumer credits.

Spain 10% Overdraft (could

change), social

housing loans

The APR cannot exceed 2.5 times the legal interest rate.

For social housing loans interest is set by reference to

indexes based on the legal interest rate.

Slovenia 13.2%-

453%

6.6%-

226%

On non bank

credit only

Maximum APR equals 200% of average APR computed by

Central Bank every 6 months depending on credit term and

amount: from 13.2% (120 month 20K€) to 453% (2

month 200€).

Source: Expert Survey and national websites. Notes: *For greater detail on the usury ceilings in place, please

refer to the respective country section in the following chapter. **The usury ceilings apply to all consumer

credits (mortgage and non-mortgage) unless specified.

64 iff/ZEW – Final Report on interest rate restrictions in the EU

1.2.1.2 Absolute interest rate ceilings

All three Member States with absolute ceilings limit the application of these to specific

types of credit provider or place significant limitations on their scope. In Greece, the

ceilings do not apply to banks, and Malta also provides banks with significant

exemptions, whilst Ireland limits the interest rates that can be charged by moneylenders

and credit unions only.

1.2.1.2.1 Greece

There is a maximum contractual borrowing rate in Greece for non-bank credit only. The

ceiling is currently 6.75% per annum. Bank credit is exempted in its entirety. The

maximum contractual interest rate is regulated by administrative law and the

government sets the interest rate, more specifically by Acts of the Council of Ministers on

the basis of proposals from the Monetary Committee of the Bank of Greece.

1.2.1.2.2 Ireland

Generally speaking there are no interest rate caps on loans in Ireland. However, there is

a maximum borrowing rate and this controls the level of interest rate that Credit Unions

can charge (1% per month of the outstanding balance, so 12.68% APR). This figure is

incorporated in legislation under the Credit Union Act.47 Furthermore, the Financial

Regulator is able to specify maximum interest rates that can be charged by

moneylenders in Ireland as part of the licensing process. Moneylender licenses are

reviewed annually and the current maximum level of interest that the regulator allows

them to charge is 187% APR.48 Furthermore, finance houses that are not credit

institutions and who are mostly involved in car finance and hire purchase agreements are

only prescribed for the purposes of the consumer credit legislation to charge an APR of

up to 23%.

We will however, consider Ireland as a country without IRR for the purposes of our

analytical work as there are no restrictions on interest rate pricing effectively in Ireland

at present. Though there is no focus on restrictions on interest rate pricing at the current

time, there are nevertheless certain prescribed scenarios (eg. where a fixed interest rate

is set there are provisions that the fixed rate must be correctly applied etc.. with

breakage applicable on early settlement of the agreement by the customer).

Furthermore, under the Consumer Protection Code for Licensed Moneylenders of January

2009, prior to entering the agreement, the moneylender must

“prominently indicate the high-cost nature of the loan on all loan documentation where the

APR is 23% or higher. This disclosure must take the following form: ‘Warning: This is a highcost

loan’.”

1.2.1.2.3 Malta

The maximum borrowing rate in Malta is set by law and has always been fixed by the

Civil Code since its promulgation in 1868. According to Article 1852(1) of the Civil Code,

the rate of interest on loans cannot exceed the interest rate of 8% per annum. Between

1868 and 1974 it was 6 % per annum. This rule however is subject to exemptions which

47 Section 38 of the Credit Union Act 1997.

48 See following link for the procedure: www.financialregulator.ie/industry-sectors/moneylenders/

Pages/authorisation.aspx. Moneylenders apply for renewal of their licences annually from the

Financial Regulator and this can be denied in case of infringement. Up until the Consumer Credit Act 1995,

Moneylenders were licensed under the Moneylenders Acts 1900 and 1933. They were not allowed to charge

more than 39% interest per annum (although this ceiling was often exceeded in practice).

iff/ZEW – Final Report on interest rate restrictions in the EU 65

results in banks not being subject to this ceiling.49 In its last paragraph the law states

that the misuse of such exemptions by “artificial, fictitious or simulated” “contract,

arrangement, scheme, transaction” will void such arrangements. The exemptions can be

summarised as follows:

financial transactions where one of the parties is a designated entity: Provided

that no party to the financial transaction may be a natural person;

secured by a mortgage registered or recognised under the Merchant Shipping Act

or the Civil Aviation Act;

governed by a foreign country and the agreed rate of interest and, or

compounding of interests are in accordance with international market conditions

and the payer of interest is not a natural person;

security (including guarantees, suretyships, indemnities and other similar

undertakings, pledges, privileges, hypothecs, mortgages and any other collateral

arrangements, whether by way of title transfer or otherwise) which is governed by

Maltese law and which is entered for the purpose of, or in connection with, any

transaction specified in paragraph (c);

financial instruments of more than two million Euro;

debt arrangement inside linked companies.

1.2.1.3 Relative interest rate ceilings

Although eleven Member States have relative interest rate ceilings which besides Spain

are all provided in the form of an APRC there are considerable differences between them

in respect of the way that these operate. For example, some implement different ceilings

according to the amount of credit that is being extended, some impose different ceilings

according to credit type, and some distinguish levels of ceilings on the basis of the

duration of the credit agreement (see Table 7 on p63 for the overview). Portugal stands

out with its new 2009 ceilings as it regulates them based on the purpose of the credit (ie.

with separate categories of ceilings for education loans or auto loans). Four of the eleven

Member States (Netherlands, Spain, Poland, and Estonia) have one unique ceiling level.

It should also be noted that in Slovenia ceilings only apply to non-banks as it was judged

that the banks face sufficient competition to not warrant such regulation. However, in all

of the other Member States there is no distinction made between bank and non bank

credit as regards the scope of ceilings (although sanctions may differ as for example in

Portugal where the rules of procedure and amounts of fines will be different depending

on the creditor’s nature ie. financial institution or not). Whereas all systems based on an

absolute ceiling calculate their ceiling from an exogenous source, of the eleven that have

a relative ceiling that can fluctuate over time, only Belgium, Poland, and the Netherlands

do not rely on prevailing market rates alone.

1.2.1.3.1 Germany

German court based jurisprudence limits lenders' flexibility by requiring them to charge

no more than double the average market rate. It also imposes a second condition limiting

the ceiling to a fixed pre-determined maximum margin set at 12 percentage points over

49 The Interest Rate (Exemption) Regulations 2009 enacted by Legal Notice 142 of 2009 issued in exercise of

the powers conferred on the Minister of Justice by article 1855A of the Civil Code further defines the special

law: According to Article 3(1) thereof.

66 iff/ZEW – Final Report on interest rate restrictions in the EU

the average interest rate. This means that when the average market rates move above

12.1% APR, the ceiling level applied will no longer be twice this (ie. 24.2%) but instead

be limited by the second condition to 24.1%. To illustrate the potential significance that

the extra moderating effect of the second condition contains, imagine average market

rates reach as high as 30%: this would lead to an authorised ceiling level set at 42%

instead of 60% if only the first rule existed.

The average interest rates are related to real market developments. The banks had to

report their interest rates and their spread to the German Central Bank, which then

calculated the average. These average market rates were published by the German

central bank (formerly referred to as the “Schwerpunkt Zinssatz”) for overdrafts and

instalment credits separately. In June/July 2003, there was a change because the

“Schwerpunktzinssatz” were replaced by the average interest rates reported by the

European Central Bank. Therefore, the ceilings lowered in July 2003 by approximately 7

percent. Interest rates exceeding the limits cause contracts to be held as usurious and

declared void by the court.

1.2.1.3.2 Estonia

There are no fixed interest rate ceilings regulated in the law, however the latest court

practice has shown that the court will intervene if the level of interest payments in

comparison to the principal is unreasonably high. In a similar approach to the German

system this has been found to be the case where the interest charged is three times

more than the market average.

The Estonian central bank publishes the average APRs for all consumer loans granted by

credit institutions to individuals on their website every month. There is no official

publication of the ceilings but the current average is available on the central bank

website50 and as of March 2010 is 21.3% which would imply a ceiling of 63.9%. However,

compared to the German system it is as yet uncertain as to how relevant and how much

of a credible deterrent this ceiling is in Estonia.

1.2.1.3.3 France

France has a long history of using interest rate ceilings. Other than the interest rate

ceilings in the 19th century that were abolished in civil transactions in 1918, the interest

rate ceilings introduced in 1935 were modernised in 1966 and its system is once again

undergoing change. It currently specifies a relative maximum APR of 133% of the

average of rates51 found for different types and amounts of credit, for example by

providing separate ceilings for revolving and instalment credit and for small and large

sum credits. This led to a system of twelve separate ceilings, including six ceilings

applicable to consumer credit. The rates as of the end of the first quarter of 2010 are

shown in the table below.52

50 The fixed rate ceiling (concerning the annual percentage rate of charge) is published on the website of Bank

of Estonia. These values are publicly available on a historical basis at

http://www.eestipank.ee/dynamic/itp/itp_report.jsp?reference=152&className=EPSTAT&lang=en.

51 As opposed to the average interest rate statistics calculated by the ECB, the French central bank does not

assign a weighting based on the size of the loan when calculating its average for the market segments.

52 See: http://www.banque-france.fr/fr/statistiques/taux/usure.htm.

iff/ZEW – Final Report on interest rate restrictions in the EU 67

Table 9: Interest rate ceilings in France

Credit category Average APR Interest rate ceiling

at 01.04.10 (133% of

average APR)

Mortgage loans to individuals (loans falling within the scope of Articles L.312-1 to

L312-36 of the Consumer Code)

Fixed rate loans 4.72% 6.29%

Variable rate loans 4.29% 5.72%

Bridge loans 4.69% 6.25%

Consumer loans to individuals (not falling within the scope of Articles L312-1 to L312-

36 of the Consumer Code (called “cash loans”, crédits de trésorerie)

Loans of less than or equal to € 1524 (1) (2) 16.22% 21.63%

Overdrafts, loans and permanent financing of

purchases or instalment sales over €1,524

mortgage and loan Life (1) (2)

14.59% 19.45%

Personal loans and other loans of a greater

amount to 1524 €

6.64% 8.85%

Loans to legal entities with no industrial, commercial, craft, agricultural or professional

non-commercial53

Loans for purchases or instalment sales 7.15% 9.53%

Loans with an original maturity exceeding 2

years, variable rate (4)

3.99% 5.32%

Loans with a maturity over 2 years, fixed rate 4.31% 5.75%

Overdrafts (3) 10.00% 13.33%

Other loans with original maturity up to 2yrs 4.30% 5.73%

Loans to individuals acting on their business needs and legal persons having an

industrial, commercial, craft, agricultural or professional non-commercial.

Overdrafts (3) 10.00% 13.33%

Source: Bank of France. Notes: (1) Expressed as an APR as stipulated by the Consumer code; (2) Only the

amount of credit actually used is considered when assessing the APR of an overdraft or a permanent loan; (3)

Rates do not include the commissions on the highest overdraft amount reached during the month - the average

rate of such fees actually collected in January 2010 amounted to 0.05%; (4) Average APRs of corporate loans

with an original maturity greater than two years, variable rate, and an amount of less than €152,449 (rate is

used by the tax authorities for calculating the maximum deductible interest on linked current accounts and was

3.99% for this category of loans in Q1 2010). Also not that the French ceilings do not apply to hire purchase

agreements (unlike the Belgian regime).

53 The concern with usury ceilings (as they are referred to in France) mainly concerns consumers. This is due

to Article 32 of Law No. 2003-721 of 1 August 2003 for the economic initiative, which eliminated the crime

of usury for lending to commercial, industrial or financial entities. An analysis is available on

http://www.mediateur-republique.fr/fic_bdd/pdf_fr_fichier/1237289023_Modification_du_taux_d_usure.pdf.

68 iff/ZEW – Final Report on interest rate restrictions in the EU

The reform law on usury ceilings in May 2010 brought the following changes:

Article 1 (amended Articles L. 313-3 of the Consumer Code) on usury rates states that

the maximum interest rates (‘les taux de l'usure’) will no longer be classified by credit

type. Before the changes, specific usury ceilings were allocated to specific loan categories

(cash/treasury loans, overdrafts, term loans, hire purchase financing, lifetime mortgages

(equity release scheme), personal loans and mortgages). These usury ceilings are now

defined only according to the amounts borrowed (with the exception of mortgage loans

(home loans), loans to local authorities and business loans). A new decree will specify

the new usury ceilings. Transitional measures will be introduced (over a maximum

transition period of 8 consecutive quarters) by the Government to promote the

implementation of the reform. In addition, to measure the impact of this reform

(potentially only the first on usury ceilings) the Government has established a committee

chaired by the Governor of the Bank of France in charge of 1) monitoring and analysing

the effects of the changes made to the method of determining the rates (‘mode de

fixation’) applicable to the level and the evolution in interest rates on consumer credit; 2)

examining the methods and ways of financing (modalités de financement) available and

being used by credit institutions doing the lending; and 3) analysing the level,

developments and components of provider lending margins. The Committee, made up of

the Governor of the Bank of France, an elected deputy, a senator and the Director

General of the French Treasury, will meet at the initiative of its president at least once

per quarter for two years. It will produce an annual report, which will be submitted to

Parliament and Government.

1.2.1.3.4 Belgium

For consumer credit there are provisions on the maximum APR. The maximum APR is

determined by Royal Decree (art. 20, § 1 WCK - Consumer Credit Act). But the APR is

only the representation of the total cost of credit. It is nowhere used to calculate these

cost for which other parameters especially the borrowing rate is used. In so far the

government can determine the maximum total costs of credit, it can also set the

maximum borrowing rate and, in the occurring case, the maximum recurring costs and

the maximum non-recurring costs vis-à-vis a revolving credit account (art. 20, §2 WCK).

Consumer credits with a credit term of more than 5 years can be sold with variable

interest rates. In this case additional rules apply the change of the APR according to

article 9 WHK (Mortgage Credit Act) stipulating the rules on variable rate mortgage loans

- art. 30, §2 WCK. For revolving credit accounts there is a specific rule providing for an

absolute maximum interest rate. Art. 4, §4 of the Royal Decree of 4 August 1992 on the

costs, percentages, the duration and the terms of repayment of the consumer credit

stipulates that “if the revolving credit account foresees various borrowing rates

depending on the drawdowns or on the instalments, none of these borrowing rates may

be higher than the maximum APR determined in function of the amount of credit”.

Overdrafts on bank accounts, which fall outside the scope of the WCK, are regulated by

the law of 14 May 2001. This law applies to every bank account opened by a consumer at

a bank or at the Postal Office and on which a debt balance occurs to which the WCK does

not apply (art. 2 of the aforementioned law). The annual borrowing rate is capped to the

maximum APR applicable pursuant to the WCK on open-ended revolving credit accounts

where the total amount of credit does not exceed EUR 1,250. The costs linked to the

credit cards do not need to be included in the total cost of the credit (art. 3).

The maximum APR for consumer credit in Belgium is based on a hybrid of mechanisms.

While initially set as absolute rate ceilings, through a rather sophisticated ceiling setting

mechanism, the ceilings are now relative ceilings since the setting of the ceiling level is

dependent on changes to determined reference rates. The reference indices, which

determine changes made to the ceiling, and the calculation method for mortgage loans

iff/ZEW – Final Report on interest rate restrictions in the EU 69

are set by the King (by Royal Decree) after consulting the Banking, Finance and

Insurance Commission (“CBFA”) (art. 9, §1, 3° WHK). Using monthly computed reference

indices for variable rate mortgage loans on the basis of a constant-maturity yield curve,

published by the Securities Regulation Fund (“Rentenfonds”), if significant changes have

been registered, the maximum APR is then adjusted by an administrative procedure (by

Royal Decree). Article 21 WCK further detailed by the Royal Decree of 4 August 1992 the

relevant costs, percentages, duration and terms of repayment are taken into account.

The ceiling depends on the credit type and the credit amount. For all consumer credit

types 12 different maximum APR’s are determined. Instalment loans, deferred payments

in sales contracts, financial leasing, revolving credit card accounts and revolving credit

accounts without cards are distinguished. A simplification in 2006 abolished a further

distinction according to the duration of the credit.

The maximum APR is calculated on the basis of a reference rate. For all consumer credit

agreements, with the exception of revolving credit accounts, the reference rate is based

upon treasury certificates for 12 months (for credit amounts up to €1,250), linear bonds

on 2 years (for credit amounts between €1,250 and €5,000) and linear bonds on 3 years

(for credit amounts above €5,000). The reference index for revolving credit accounts is

linked to the monthly average of the 3 month Euribor. The reference rates are calculated

by Belgostat. The applicable maximum APR corresponds to the respective rounded

reference rates.

Recently the maximum APR, applicable to all consumer credits, was adjusted for the 4th

time since 1992. The maximum APR is published in the Official Journal. It is analysed

every 6 months, to see if reference rates (Euribor term rates) have changed beyond 75

basis points in which case reference indexes and then APR ceilings will be adjusted

accordingly, with a rounding to the nearest half a percentage point for the ceiling.

For consumer credit Table 10 shows the APR limits currently in place:

Table 10: Interest rate ceilings in Belgium

Credit Amount Index Instalment

Loans/deferre

d payment

Revolving

credit account

with card*

Revolving credit

account without

card*

Financial

Leasing

€1,250 Index A 19.5% 15.0% 11.0% 13.5%

> €1,250 up

to €5,000

Index B 15.0% 13.0% 10.0% 11.0%

> €5,000 Index C 12.5% 12.0% 10.0% 10.5%

Source: Official Journal. Note: Changes in the indices, when they are greater than 0.75 points, determine the

adjustment mechanism of these rates. *These indices do not apply for revolving credit, the changes to the

ceilings of which are based on the changes in the 3-month Euribor.

An opinion from the consumer council on changes to the fixing of maximum APRC was

issued in May 2006 and led to some significant changes.54 The objectives of the council

were to: 1) simplify the existing IRR system by reducing the number of different ceilings

(which stood at 28 before the change) and bringing the product categories in line with

prevalence in the market; 2) introduce an automatic system for adapting the ceilings

going forward (moving from the previous dependency on the King to make a decree to a

method of objective adjustments based on evolution of a reference index from the

54 Conseil de la Consommation, Belgique (2006).

70 iff/ZEW – Final Report on interest rate restrictions in the EU

financial markets); and 3) to reach a reasonable balanced solution acceptable to all

parties involved (consumer associations and providers).

Finally, the maximum borrowing rate for overdrafts on bank accounts, which are

regulated by the law of 14 May 2001 on overdrafts on bank accounts, is currently 11%.

The maximum APR’s are also published in the Official Journal. For variable rate mortgage

loans, the reference indices are published monthly in the Belgian Official Journal. The

current reference indices are also publicly available on the websites of the Securities

Regulation Fund and of the Banking, Finance and Insurance Commission.

1.2.1.3.5 Slovakia

Slovakia introduced interest rate legislation only very recently (in its Consumer Credit Act

2008). Government Regulation No. 238/2008 Coll. sets the maximum APR to serve for

the purpose of a usury ceiling). It is currently 79.08% (until 15th of May 2010). The

Ministry of Finance of the Slovak Republic in cooperation with Národná banka Slovenska

set these reference rates. The source data to inform this decision is provided by creditors

who are obliged by art. 7a sec. 1 of the Act on to submit data on their interest rates to

the Ministry and to the national central bank.

According to art. 1 sec. 1 of Regulation No. 238/2008 Coll., the maximum interest rate

allowed is twice the average value of APR for the type of consumer credit being lent, and

at the same time it cannot exceed four times the value of the weighted average APR cost

and average interest rates for all types of consumer credits valid at the date of

concluding the contract (Table 11).

The fixed rate ceiling is adjusted quarterly and the level is published on the webpage of

Ministry of Finance of the Slovak Republic. The values of interest rate APR ceilings are

publicly available at page of The Ministry of Finance of the Slovak Republic.55 Slovakia

also has rules concerning default interest regulation, which are set out later in this report

(Chapter 1.2.2.3).

However, there is currently a discussion taking place and a new Proposed Act on

consumer credit. All the interest rate caps related legislation could be repealed by this

with alternatives suggested by increasing price transparency and improving the financial

literacy of consumers. The interest rate ceilings do not apply to mortgage credit

contracts. These are provided by mortgage banks according to general terms and

conditions stipulated in art. 75 sec. 1 letter e) of the Banking Act. General terms and

conditions of the mortgage credit and municipal loans may include some IRR.

Table 11: Interest rate ceilings in Slovakia

Type of consumer credit Contractual maturity

Value 3 to 6

months

6 to 12

months

1 to 5

years

5 to 10

years

Over 10

years

Credit cards - the average

interest rate (not APR)

39.00 x x x x x

Consumer loans with

security or lease of up to

€1,500 inclusively

x 76.00 45.24 48.58 35.06 21.38

55 See http://www.finance.gov.sk/Default.aspx?CatID=6865.

iff/ZEW – Final Report on interest rate restrictions in the EU 71

Consumer loans with

security or lease ranging

from €1,500 to €6,500

inclusively

x 68.60 44.54 37.60 28.56 12.86

Consumer loans with

security or lease of more

than €6,500

x 47.74 34.50 22.78 20.48 11.52

Other consumer loans (not

included above) amounting

up to €1,500 inclusively

x 76.00 76.00 76.00 48.28 0.00

Other consumer loans (not

included above) ranging

from €1,500 to €6,500

inclusively

x 51.58 49.08 41.44 37.18 32.86

Other consumer loans (not

included above) of more

than €6,500

x 24.22 20.50 27.54 30.50 13.32

Four times the weighted

average of the average

values of the APRC and the

average interest rate for

all types of consumer

loans

76.00 x x x x x

Source: Ministry of Finance of the Slovak Republic. Note: The ceilings are those of the first quarter 2010 and

are valid for credit agreements concluded from 16 May 2010 to 15 August 2010. The maximum ceiling level in

% is set at an accuracy of two decimal places.

1.2.1.3.6 Italy

Usury is a criminal offence in Italy and it provides a detailed system of usury ceilings

based on 50% above calculations of the average charges in the market (APR or ‘TEGM’)

for different types of credit and different credit amounts. While Italy, along with all other

countries with relative IRR with exception of Poland, use the APRC of the CCD 2008, a

big debate is taking place concerning the treatment of insurance costs which this

Directive did not include if concluded “voluntarily”.

The ceilings are effective for every kind of transaction or financial/credit operation, and

for every kind of subject. Civil/contractual remedies, which include voiding the contract

and/or substituting new interest rates into the contract can also be combined with

criminal sanctions. The different types of credit, amounts, and ceilings put in place by

Decree of 24th December 2009 are set out in Table 12.

72 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 12: Interest rate ceilings in Italy

Type of credit < €5,000 > €5,000

Savings and bank account (secured

or unsecured)

12.85% + 50% 9.59% + 50%

Anticipations and corporate loans 10.26%+50% 5.43%+50%

Consumer Credit (instalment loans) 12.53%+50%

Revolving credit (credit card) 16,97%+50% 12.79%+50%

Special purpose credit (finalizzato) 14.18%+50% 12.17%+50%

Mortgage loans 5.36%+50% (if fixed rate loans) 2.92%+50%

(if variable rate loans)

Salary and pension loans 15.53%+50% 12.46%+50%

All other family loans, finalized or

not finalized including pawnbroking

18.13%+50%

All other family loans if provided by

non-banks companies

14.41%+50%

Other amounts

Auto leasing (and aeronautic

leasing)

Ceiling for credit under

€25,000:

14.31%+50%

Ceiling for credit over

€25,000:

12.67%+50%

Equipment leasing 10.34%+50% (under

€25,000)

6.56%+50% (over

€25,000)

Factoring Ceiling for credits

under €50,000:

6.14%+50%

Ceiling for credits over

€50,000: 4.05%+50%

Overrunning on bank account Ceiling for credit under

€1,500: 19.96%+50%

Ceiling for credit above

€1,500: 13.12%+50%

Real estate leasing 4.39%+50%

Source: Bank of Italy.

The types of credit are decided every year by Banca d’Italia, which collects data from all

the credit providers. The rate ceilings are adjusted every three months by the Ministry of

Economy and Finance, who approves the rates by decree and are published in the Italian

Official Journal (Gazzetta Ufficiale). Furthermore, Italy has now explicitly harmonised the

calculation of the APRC in usury with the APRC in price disclosure according to CCD 2008.

A 2009 Decree of the Ministry of Economy and Finance involved changes to the statistics

iff/ZEW – Final Report on interest rate restrictions in the EU 73

behind the interest rate series on which the ceilings are based clarifying the classification

for the applicable average interest rates for the usury test.56 According to the press

release of Banca d'Italia the new regime is especially important with regard to the

calculation of the contractual interest rate. The comprehensive review of costs to be

included in the calculation of the TEGM (APRC) and revision in the calculation of the

interest rate series was part of the efforts by the authorities and legislators to harmonise

the basis of calculating the APRC. Alongside changes in the statutory calculation and

reporting scheme, the law has led to the inclusion of certain costs previously excluded

from the Italian interest rate series used to determine the ceilings, such as the

introduction of overrunning fees (commissione di massimo scoperto, CMS), of all kind of

brokerage fees (when borne by the consumer) along with insurance cost.57 The Bank of

Italy collaborated with the Ministry of Economy and Finance in order for the new series of

rates (and thus average rates) to be reported by providers starting in the third quarter of

2009 with changes to the ceilings based on the new methodology starting in 1 January

2010. The new APR and interest rate ceilings for Italy are now more comparable with the

APR rates stipulated by the CCD 2008.

1.2.1.3.7 The Netherlands

The Netherlands stipulates two fixed ‘legal interest rate’ (wettelijke rente), one for

business transactions and one for transactions with consumers. The consumer rate is

currently 3% per year and this is the maximum that can be charged on default. An

administrative ceiling is then set which governs the maximum APR that can be charged

on contractual agreements as x% in addition to the legal rate. This is currently 12%, so

the ceiling is 12% plus 3% = 15% for consumer credit agreements.

The level of the ceiling is set by the Minister of Justice and is published in the offical

Staats courant (an official bulletin of the State).

This 12% extra interest was lowered in 2006 from 17% to make it harder to charge high

interest rates. However, some types of credit are not regulated, including credits that last

for less than 3 months. Over the past 2 years, there has been a growth in payday loans

(in Dutch better known as “flash credits”) which are not regulated, and which have APRs

as high as 600%58.

However, when implementing the EU CCD 2008, these forms of credit will become

regulated and the interest rate caps will apply. It is expected that this move will lead to

the demise of payday lending in the Netherlands.

Other forms of credit are also not subject to the general interest rate ceiling, including

business loans, mortgages, pawn broking, stick financing, and government loans.

However, there is a code-of-conduct for Dutch mortgage suppliers, which allows

contractual rates to be used to calculate loan-to-income ratios for mortgages where

these are set to last for at least 10 years. If a mortgage is set to last for less than 10

56 Decree with regard to "transitional provisions with the application of Article of Law on March 7 No .2 108"

(published in G.U. of 29 July 2009), see: http://www.bancaditalia.it/vigilanza/contrasto_usura/ doc_cons/

istruzioni_teg/ istr_usura_doc_consultazione.pdf.

57 “Verranno, tra l’altro, considerati ai fini della definizione del tasso soglia le polizze assicurative, i compensi

per i mediatori, nonché tutte le forme di remunerazione diverse dal tasso di interesse, come le commissioni

di massimo scoperto e quelle per la messa a disposizione dei fondi nei limiti e alle condizioni consentiti dal

legislatore. Per i compensi di mediazione è stata introdotta anche un’apposita rilevazione al fine di fornire un

parametro specifico per valutare l’usurarietà di una componente di costo di rilievo e variabilità considerevoli,

ma sinora priva di limiti definiti”, from August 2009 press release at

http://www.bancaditalia.it/media/comsta/2009/cs_usura_120809.pdf.

58 A June 2009 report (in Dutch) looked at the issue of flash credits and the usury ceiling, available here:

http://www.responsible-credit.net/media.php?t=media&f=file&id=3872.

74 iff/ZEW – Final Report on interest rate restrictions in the EU

years then the contractual rate is not used for this purpose but the lender must instead

calculate the loan to income ratio using a set minimum interest rate. This is intended to

prevent excessive mortgages from being offered based on low contractual initial rates.

1.2.1.3.8 Poland

Rate caps in Poland are subject to regulation in the Civil Code and in the Consumer

Credit Act. The maximum interest rate in Poland is a relative rate ceiling for all types of

credit, calculated by reference to the central Lombard rate multiplied by four. The current

Lombard rate is 5% giving rise to a maximum borrowing rate of 20%59. This mechanism

for limiting interest rates was introduced in 2005 and, in contrast to the other Member

State ceilings, the ceiling is set on the borrowing rate, and thus just on the interest rate,

not the rate representing the total cost of the credit (ie. the APR). In line with this

distinction, in addition, fees and additional charges related to the concluding of the credit

contract are separately regulated as well, and cannot exceed 5% of the amount of the

credit.

The rule applies to all credit types, depending only on central bank decisions (Monetary

Policy Council) and the rates are reviewed monthly. Decisions concerning changes are

published in the statements of the Central Bank’s Monetary Policy Committee.

1.2.1.3.9 Portugal

The recent transposition into the Portuguese law of the CCD 2008 (June 2009) has

established an interest rate cap on credit operations based on the APR. Maximum interest

rates are only applicable to credit granted to consumers, are set quarterly by the

Portuguese Central Bank and are based on the average of the interest rates applied by

credit institutions in the preceding quarter. If these caps are exceeded, the credit is

considered to be usurious and gives rise to criminal liability.

Decree-Law nr. 133/2009 establishes usury ceilings for: Instalment credit; Financial

leasing; Hire purchase financing; Point-of-sale financing (the vendors/service providers

act on behalf of credit institutions); All categories of revolving credit with the exception of

overdraft facilities where the credit has to be repaid within one month.

The rate ceilings also vary in accordance with the purpose of the credit and are currently

set at the levels shown in the table below:

Table 13: Interest rate ceilings in Portugal

Credit categories Interest rate ceiling

(2nd quarter 2010)

Education, health and renewable energy 6.7%

Leasing 7.3%

Other personal credits 18.9%

Auto leasing or hire purchase financing (new vehicles) 7.7%

59 A history of the Lomabrd rate can be found under the following link:

http://www.nbp.pl/home.aspx?f=/dzienne/stopy_archiwum.htm.

iff/ZEW – Final Report on interest rate restrictions in the EU 75

Auto leasing or hire purchase financing (used vehicles) 9.9%

Secured auto loans (new vehicles) 11.1%

Secured auto loans (used vehicles) 15.6%

Credit cards, credit lines, overdraft facility, overrunning 31.6%

Source: Bank of Portugal. Note: These APR usury ceilings shown apply to credit and financial institutions, other

civil and commercial usury ceilings also exist but are not shown.

The ceilings are determined and disclosed by Bank of Portugal, and are set quarterly and

published in the Official Bulletin of Bank of Portugal. However, the ceilings apply only in

respect of credit amounts of between 200 EUR and 75,000 EUR. It is nevertheless

important to point out that the subtypes of credit defined by the Bank of Portugal in

order to fix APR usury ceilings raise a problem of concepts. In fact, there is no definition

or list of “Education/Health” products or services except for tax purposes. Although, in

what renewable energies concern, and also for tax purposes, there is a list (Regulation

nr. 303/2010, from 8 of June 2010), treatment is complicated by the fact that electric

vehicles will fall under the “auto” loan category. The APR usury ceiling that was

introduced in the beginning of 2010 was the result of discussions concerning consumers’

rights that took place during the early months of 2009 and which led, in June 2009, to

the approval of Decree-Law nr. 133/2009 which was in some opinions, seen as hasty and

a little careless. By transposing the CCD 2008 and introducing a new APR usury ceiling

applicable to financial institutions, there is now some ambiguity since there are presently

three usury ceilings in Portugal: (i) civil interest rates usury ceiling; (ii) commercial

interest rates usury ceiling and (iii) APR usury ceiling. Only the APR usury ceiling is

applicable to credit provided by financial institutions, however, it is not entirely clear if

other creditors have to submit consumer contracts to more than one ceiling.

1.2.1.3.10 Slovenia

There are several interest rate related restrictions present in Slovenian legislation. Some

of them are general and refer to all types of credit contracts (contained in the Code of

Obligations, although some of these provisions are further limited to parties to noncommercial

contracts), while others protect only consumers (contained in Consumer

Protection Act and Consumer Credit Act).

Caps of penalty and contractual interest rates are set in the Code of Obligations (and

breach of this provision can lead to contracts being voided). For consumer credits the

interest rate ceiling (expressed as an APRC) is set at twice the average APRC charged by

banks and savings banks (Article 18a) and is published in Official Journal by the Bank of

Slovenia twice a year. The central bank calculates the level of APRs of credit institutions

in accordance with Article 17 of the Consumer Credit Act and the ceiling is legally set by

the Parliament.

Average effective rates are calculated for, and therefore ceilings set for, consumer credits

for the following maturities and amounts: 2 months, 4 months, 6 months, 12 months, 36

months, 10 years; and 200, 500, 1000, 2000, 4000, 20000 Euros.

76 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 14: Interest rate ceilings in Slovenia

Maturity

(months)

Amount

(EUR)

Average

APR

APR

Ceiling

2 200 226.5% 453.0%

4 500 44.3% 88.6%

6 1,000 21.3% 42.6%

12 2,000 12.4% 24.8%

36 4,000 9.6% 19.2%

129 20,000 6.6% 13.2%

Source: Bank of Slovenia. Note: The maximum rate is set at 200% of average consumer credit APRs charged

by banks and savings banks (as of 1. 12. 2009, published by Bank of Slovenia on 31. 12. 2009, OJ 113/2009).

Credits are put in the lowest category possible.

There is also a ceiling on default charges, which is set by statute as 50% above the sum

of the statutory default interest rate of 8% and the ECB’s main refinancing rate60. There

are also specific IRR for housing loans. The national housing saving scheme loan interest

rates are fixed at 75% of the market yield from Republic of Slovenia bonds of maturities

above 9 and a half years. The interest rate changes when the yield differs from base

value by more than 10%. These differences are inspected twice a year (15th May and

15th November), while interest rates are accommodated once a year if in two

consecutive tests yield differs from basis by more than 10%.

Finally, Slovenia also provides a cap on interest rates relating to ecological credits, which

are currently set at 3.20% for maturities of less than five years and at 3.90% for

maturities from five to ten years.

1.2.1.3.11 Spain

Spain has a general prohibition of usury which is not specific to credit.61 Using this

concept courts have sometimes overruled certain loan agreements, if they find that they

breach the law on usury. The Spanish Supreme Court has declared that in the

appreciation of usury it will apply substantive criteria.62 But most court decisions relate

to loans between individuals.63 At a lower level there have been some attempts to apply

the specific IRR for overdraft credit analogously to other credit contract too.64

60 Default interest rate ceilings as opposed to contractual rate ceilings are reported on for all the EU Member

States in a subsequent section.

61 Law of Usury of 23rd July 1908 “Le «Ley Azcárate».

62 26th March 1993 (RJ 1993, 2241).

63 Supreme Court 30th December 1987 ( RJ 1987, 9713), 24th April 1991 (with individual vote) ( RJ 1991,

3025); 8th November 1991 ( RJ 1991, 8148) and 29th September 1992 ( RJ 1992, 7330) among others.

For a contract between a bank and a consumer see Supreme Court 26th March 1993.

64 I.e Audiencia Provincial of Girona, Sección 2ª, (auto) 15th December 2009, Judge Rapporteur: José Isidro

Rey.

iff/ZEW – Final Report on interest rate restrictions in the EU 77

In general there is only the exemption for overdraft credit. Art 19 Consumer Credit Law

1995, provides that current account overdrafts cannot charge more than 2.5 times the

level of the legal interest rate. The legal interest rate is set each year in the Budget Law

and is currently 4%. ACT 7/1995, of March 23, on Consumer Credit (Article 19.4) holds:

The interest applicable to overdraft facilities on consumers' current accounts shall not

exceed, in terms of APR, 2.5 times the legal interest on money. Since Spain refers to the

ancient legal interest rate of the civil codes which is disclosed in the form of a borrowing

rate and since anyhow CCD 1987 did not require an APRC for overdraft credit this is the

only known relative direct IRR which uses the borrowing rate.

There are some special sector specific rules. For instance some caps to applicable interest

rates are found in relation with social housing acquisition (grant aided home purchase of

homes). These form part of special programmes “Plan de Vivienda” which consists in a

subsidised way of home acquisition. All autonomous communities have such

programmes. Spain has a sophisticated system to regulate these interest rates on

protected/social housing.65 In this specific case the interest rates are in the form of the

APRC and are set by reference to indexes, of with the “interés legal” is the index

applicable to default.

The ceilings for protected housing are fixed by the National Minister on Housing, and /or

by the competent regional ministry. As those caps are related to the “legal interest rate”

the Budget Law which sets each year the applicable “legal interest” or “statutory rate”

which currently is set at 4% until 31st December 2010 has to be taken into account. In

long term contracts, the adjustment is done annually.

To fix these rates the Mortgage Loan Order (following order of May 5, 1994 on

transparency of mortgage loans), requires the official setting of the following indexes or

reference rates by the Bank of Spain and published monthly in the "Official Gazette". Its

definition and way of calculation can be found in Annex VIII of the order.

Reference rate for mortgage loans more than three years term for house

purchase, granted by banks.

Reference rate for mortgage loans, more than three years term for house

purchase, granted by savings banks.

Reference rate for mortgage loans, more than three years term, for house

purchase granted by other credit institutions.

Savings banks reference rate.

Domestic income in the national debt secondary market for two to six years term.

One year interbank rate.

If the rate effectively imposed is higher that the rate described in the contract, there

would be sanctions on the lender.

65 Law 2/1994, of 30th March on subrogation and modification of mortgage loans; Law 41/2007, that modifies

Law 2/1981, of 25th March, for the regulation of mortgage market and other laws of the Mortgage and

Financial Systems, for the regulation of reverse mortgages and dependency insurance and which establishes

tax regulations; Law 2/2009, of 31st March that regulates mortgage loan contracts and intermediary

services with consumers.

78 iff/ZEW – Final Report on interest rate restrictions in the EU

1.2.1.4 Interest rate ceiling setup

Because of the possible complexity involved in attempting to control the pricing decisions

of the credit markets, it is important to understand the various mechanisms and

modalities open to regulators which determine how the interest rate restrictions operate

and at what level the ceiling or ceilings will be set.

1.2.1.4.1 Exogenous reference point

A system based on an exogenous reference point is often suggested for a unique ceiling

whereby an appropriate reference should be an index or rate representing the cost of

resources for lending institutions. This can be done by having the system based on the

addition of a flat predetermined margin in percentage points on top of a reference rate

which could either be to the base rate of say the European Central bank (eg. as in the

Netherlands) or to an interbank refinancing rate (eg. as in Belgium). The advantages of

such a system are its simplicity and the level playing field it provides in terms of

competition between different credit types. Proponents of such a system argue that it

would solve the potential problem of the ratchet effect over time that is inherent in

endogenous systems (phenomenon whereby the calculation of the usury ceiling based on

average observable rates in time period t leads to further automatic increase in the

ceiling in t+1 when providers seek to provide credit at prices as close to the ceiling as

possible). A possible variation of such a regime would be to target specific credits with an

extra exogenously referenced condition.

1.2.1.4.2 Endogenous reference point

The majority of Member Sates (see below) have a system that sets the interest rate

ceilings relative to those rates observed in the credit markets in practice. Generally the

preferred method of setting the ceiling is using a multiplication coefficient applied to the

average rates in the market (or more often to sub credit markets), however there is also

the possibility to apply a fixed interest rate margin to the average market rates as

opposed to a multiple. One advantage of specifying a margin fixing the number of basis

points above which a credit is considered usurious is that it limits the proportionality

effects and thus allows providers of small-sized credit to have a more favourable

treatment of their proportionately higher fixed costs vis-à-vis those extending larger

loans.

A possible variation combining both systems of ceiling fixation is a model that

superimposes a second ceiling condition on top of another eg. credit markets or specific

credit categories would see a ceiling set on both average market rates and an

exogenously determined limit. A host of other possible structures could exist. The French

government report investigating possible revisions of the existing system in France also

conducted simulations on a system of authorised coefficient ranges which would then

allow the regulator to decide which categories get allocated which multiplication

coefficient from within the range (with the advantage of allowing some steering in usury

ceilings across categories).

1.2.1.4.3 Interest rate ceiling setting mechanisms in the EU

The majority of Member States with interest rate ceilings have chosen to determine their

ceilings based on a multiplication coefficient. This coefficient varies from country to

country. For those that apply the coefficient to average credit market prices, the

following coefficients are being used: Portugal and France (x1.33); Italy (x1.5);

Germany, Slovakia and Slovenia (x2); and Estonia (x3).

Poland also uses a calculation method based on a multiplication coefficient (of x4) but

this is applied to an endogenous reference rate, namely a central bank rate as opposed

iff/ZEW – Final Report on interest rate restrictions in the EU 79

to rates contracted on the credit markets themselves. Likewise, Spain’s ceiling applied to

overdraft credit (which is being reviewed) uses a coefficient of the Spanish legal rate

(x2.5).

Two countries with relative interest rate ceilings which do not have a system using a

multiplication coefficient are Belgium and the Netherlands. In Belgium where the ceilings

are calculated by category of credit based on their nature and amount, a complex system

using initial absolute ceilings determined at one point in time in the past then uses

diverse elements of reference for each category of credit. These references are

exogenous to the credit market itself and are derived from fluctuations in specified

financial market indexes of either the 3-Month Euribor, or the 1 year or 2 year

government bond rates. In the Netherlands, the usury ceiling corresponds to the legal

interest rate (set on the basis of the ECB base rate) plus a fixed margin irrespective of

the amount of the credit. This margin currently stands at 12 percentage points (reduced

from a previous 17%). The German mechanism of setting the ceiling also has a fixed

nominal margin of 12 percentage points and this serves as a second control to the

relative floating cap by limiting the dispersion of the ceiling from the mean. This is an

extra constraint for times of high market rates which impose the average market rates

plus 12 pp instead of the usual double the average of market rates. Slovakia has also

implemented a double criterion for limiting the ceilings in order to restrain an excessive

spread between the interest rates observed for the different credit categories, namely by

having one unique overarching ceiling for all credit types together and set at four times

the average for all credits (x4).

1.2.1.5 Sanctions

Under Chapter 1.1.4 we have dealt with the general questions of effectiveness with

regard to any regulation of IRR in the law. In this part we will more specifically focus on

those rules which directly affect the contractual interest rate. While it shares the

importance of effective sanctions with all other legal rules, its specific problem lies in the

fact that an interest rate is only a parameter which should “reflect” the true cost of credit

and its burden onto the consumer.

Three types of sanction exist: civil, criminal and administrative.

Civil sanctions include

o the reduction of the interest to either the principal or permitted rate of

interest66, or

o the nullity of the contract with the possibility of judicial allocation of the

obligations under the contract.

Criminal sanctions include imprisonment and/or fines.

Administrative sanctions will generally be loss of a licence to trade.

Some Member States, for example Belgium, France and Italy specifically include all three

potential sanctions67.

66 For example Poland and Portugal (nr 3 art 28 Decree Law nr 133/2009).

67 Belgium: Restriction to cash price or borrowed amount (art. 87, 1 WCK): criminal sanctions (art. 101,§, 4

WCK): administrative sanctions (art. 106,§ 1, art.107§ 1). France Arts. L 313- 4, (reduction of interest to

legal limit, application of excessive payments to interest due and then to capital remaining due: contract not

a nullity) L313-5 (fine up to €45,000 and or/imprisonment for two years with publication of the decision and

temporary or permanent closure of the business). Slovakia: contract is voidable by consumer if interest rate

ceilings contravened; Slovak Trade Inspection may impose a fine up to 500,000 SKK: Poland, excessive

80 iff/ZEW – Final Report on interest rate restrictions in the EU

In civil cases in some Member States the invalidity of the contract must be alleged by the

borrower (eg. Slovakia), but in others it is a matter of ordre public.

A distinction might also be drawn between those Member States that prohibit usury in

the form of judicial interpretation of a general clause. These include Germany, the Czech

Republic, Estonia, Hungary, and Slovenia. In these Member States the consequences of a

finding that an interest rate is “contra bonos mores” may be that the court will declare

the contract null and void. In Germany the courts will permit the borrower to keep the

loan but be absolved from repaying, thus acting as a deterrent to predatory lending more

generally.68

1.2.1.5.1 Public enforcement of usury/interest rate restrictions

Consumer credit institutions may be subject to a variety of public regulators in Europe.

These include:

Through the central bank (Italy, Portugal, France) with lending being restricted to

institutions licensed by the bank.

A licensing authority which may be a Ministry (Belgium) or a specialized agency

(UK, Germany, Netherlands, Estonia).

A consumer protection authority (Ireland, Bulgaria, Latvia) or Market Inspectorate

(Slovenia).

A Consumer Ombudsman along with a Financial Supervisory Authority (Finland).

Countries may have a number of agencies. The Ministry of Economic Affairs in Belgium

authorises consumer credit institutions but mortgage companies are controlled by the

Banking, Finance and Insurance Commission. Financial institutions in Ireland are licensed

by the Central Bank, money lenders by the Financial Regulator. In France the new

Autorité de contrôle prudential (2010) will co-operate with the DGCCRF (Directorate

Générale de la concurrence, de la consommation, et de la repression des Fraude) in

policing consumer financial services.

In many countries therefore there is ex ante control of most credit suppliers entering the

credit market (and this will increase after implementation of the CCD 2008) with the

possibility for the regulator in some countries to stipulate minimum capital requirements

(France), a viable business model (Germany, UK) and to scrutinise the contracts of a

company. The regulator may also have power to suspend the licence of the provider, as

occurred recently in Italy where the Bank of Italy suspended Amex’s licence to issue new

credit cards for alleged contraventions of art. 644 of the Criminal Code.69 There are

however still products which escape ex ante regulation like “express loans” in Lithuania

and unregulated non-banks in Poland.

The UK also has a special illegal lending project dedicated to prosecuting loan sharks in

England’s very poor neighbourhoods. A study of illegal lending in the United Kingdom

estimates its prevalence to be about 165,000 or 0.44 per cent of the population,

although the methodology used in this study has been called into question recently. The

interest is void and maximal rate replaces the contract rate. Italy: consequences of illegal rate is reduction

to legal maximum (art 1419 cod.civ-1815 c.c.), possibility of pecuniary sanctions and possible suspension or

revocation of authorization for supervised institutions (see eg. recent case of Amex suspension of licence).

68 MS Germany see jurisprudence under art § 138.2; Czech Republic (contract void) court rulings 15

December, 2004, no 21 Cdo 1484/2004 and Nr 22 Cdo 1993/2001 from 08. April 2003 No.22 Cdo

1993/2001.

69 See: http://www.financialexpress.com/news/american-express-credit-cards-stopped/602332/.

iff/ZEW – Final Report on interest rate restrictions in the EU 81

average amount lent in the UK illegal lending market is estimated to be very low, at only

£25070.

1.2.1.5.2 Private enforcement/ombudsmen

Ombudsmen are not generally viewed as an alternative means of addressing problems

with high cost credit in those countries with ceilings (eg. Portugal, France, Italy,

Belgium). Although the use of financial ombudsmen seems to be growing throughout the

EU (eg. Slovakia, Slovenia, Spain) there was either not enough experience of these to

comment on whether or not they had an impact on price, or where comments were

forthcoming these indicated that only a modest impact could be discerned. The UK

appears to have the largest Financial Ombudsman service.

1.2.1.5.3 Ability of courts to declare total cost of credit or interest rate excessive and

reopen the terms of the agreement

This power exists in both countries with ceilings (eg. Belgium) and those countries

without ceilings (eg. Denmark, UK). The conditions for doing this generally require a

finding that the terms contravene a general clause such as contra bonos mores

(Germany, Estonia), that the terms are “flagrantly unfair” (Denmark), or unconscionable

(Sweden). Slovakia provides for the possibility of court intervention if the amount of

interest “substantially exceeds the normal interest rate” and the Supreme Court of the

Czech Republic struck down an agreement where the interest was more than four times

the usual interest (Supreme Court of Czech Republic sp. Zn. 21 cdo 1484/2004). A

similar decision can be found in Spain (eg. for a mortgage agreement of 20% when

similar agreements are at 5%: Audiencia Provincial de Madrid, Seccion 12a, Sentencia de

14 July 2009, rec. 634/2007).

Much however depends on the circumstances. An interest rate that is twice or three

times over the average rate charged by financial institutions might not be held to be

unfair. Courts look to all the circumstances of the case and decide based on the fairness

of the term and the market situation of the parties (eg. if the person lacks experience, or

is vulnerable: Sweden, UK, Slovenia).

In Germany it appears that the courts apply an almost irrefutable presumption that one

party is in a weak position where an interest rate is double the average. In the UK

ss140A-D of the Consumer Credit Act 1974 (added in 2006), confer very broad powers

on the courts to reopen and vary terms where there is an “unfair credit relationship”.

However, early case law suggests that very high interest rates will be upheld if the rates

are not different from those in the particular market (see Robert Shaw and Nine Regions

(Log Book Loans) Ltd, where a log book loan of £3000 used to pay for a consumer

vacation at APR of 119.6% was upheld because this was similar to charges made by

pawnbrokers and others operating in the high cost sub markets).

High interest rates might also be challenged under legislation implementing the Unfair

Terms in Consumer Contracts Directives in those Member States that have not included

the limitation in the Directive on review of price clauses.

When the loan – or the interest - is declared void in Spain according to the 1908 law on

usury, the debtor is required to return the principal, but is not required to pay any

interest. This regime is an exception to Art 1303 Spanish Civil Code which, upon default,

imposes full restitution with interests. This interpretation of the usury law as an

exception to the Civil code was upheld by a Decision of the Supreme Court on January

9th 1933, and later Court Cases.

70 See: Policis (2006b) pp 57, 76.

82 iff/ZEW – Final Report on interest rate restrictions in the EU

1.2.1.6 Member States with no special rate ceiling

According to the opinion of the legal experts as well as the responses given by various

stakeholders, thirteen Member States do not have something that could be called a direct

and targeted intervention into the price of credit apart from what has been described by

reference to the general principle of good morals and fairness. These principles as

described above in Section 1.1.3 forbid extortionate pricing in contracts in general if it

amounts to the exploitation of the weakness of the contractual partner. While all Member

States have a potential for IRR only 14 presently use this potential. Whether IRR will be

introduced in the other Member States in the near future depends on political discussions

that are presently taking place, and which are especially strong in the UK. While cultural

and political reasons may play an important role in these discussions, much of it is due to

the state of development of consumer credit in these countries which is discussed under

Chapter 2.3.

Apart from Latvia, Lithuania, and Cyprus, all Member States are reported to have some

form of doctrines which may be used to limit high cost credit in contracts. Romania,

Denmark and the UK use concepts of fairness or unfairness and Austria uses the concept

of unconscionability. Other either discuss its introduction or have a history of regulation

and deregulation in this area.

Romania, following a decree law of 5 May 1938 set the level of the interest rate caps at

4% above the National Bank discount rate (with the possibility of different ceilings set for

different periods and regions). By the decree of 9 August 1954 this was then repealed

and the concept of a “legal interest rate”, which was set at 6%, was introduced in its

place. This law was further amended in 1998 and 2000 when the Romanian Government

decided that the legal interest rate would not be applied to the interest due on any

financial obligations charged or paid by the National Bank of Romania, other banks, the

Office of Savings and consignments, credit cooperative organizations and the Ministry of

Finance. This decision to exclude banks and credit lenders from a legal ceiling has been

criticised on the basis that excessively expensive credits may be used to exploit debtors

and might affect the re-launching of the national economy.

In Denmark, which has historically not been in favour of using IRR as a form of

regulation, there has been debate since early 2009 when the opposition declared that it

would introduce a bill concerning a rate cap corresponding to the central bank base rate

plus 15 percentage points. The bill has not yet been introduced. The discussion has been

reopened in February 2010 (not the least due to the so-called SMS-loans with APR’s at

more than 2,000 percent) with the Danish Consumer Council having expressed a wish for

IRR.

In the UK there have been a number of attempts by consumer advocates to introduce

rate ceilings, including the submission of amendments for this purpose during the

passage of the recent Financial Services Act 2010. However, these were not passed and

Government instead commissioned the Office of Fair Trading to review the case for

ceilings as part of a wider review of high cost credit markets, whose final report was

published in June 201071.

In Cyprus an intense debate took place in 2010 on a legal cap on bank interest rates to

consumers.72 The Cyprus legal environment is characterized by liberalization of interest

rates and usury is not qualified as a criminal offence. The Central Bank of Cyprus in one

of its statements had pointed out that usurious interest rates cannot be regulated and

71 See: Office of Fair Trading (OFT) (2010).

72 Eg. see following media link (in Greek): http://www.mykypros.eu/cgibin/hweb?-A=60149&-V=ikypros&w.

iff/ZEW – Final Report on interest rate restrictions in the EU 83

that consumers and providers have to take their own risks.73 In Cyprus, no primary

sources (legislation, jurisprudence) are available on the issue however information

published in trustworthy press sources refer to the progress of the parliamentary work.74

Though reality shows that interest rates in credit markets in Cyprus are not regulated

(following the previous regime that had a 9% limit which has been abolished pursuant to

the liberalization of interest rates), it is nevertheless legally interesting, since the

phenomenon, which is considered to be a criminal offence in other jurisdictions (usury) is

not currently sanctioned. This however may change pending discussions on a bill that has

currently been drafted to criminalise the lending of money at an unreasonably high

interest rate. The transposition of the CCD is expected to be transposed into national Law

in October 2010. Furthermore, this liberalization of consumer credit was seen by our

expert as not having led to a generalised social problem and is not considered to be a

social issue of high concern. One reason put forward for that is the client due diligence,

as applied by the credit institutions, before the credit decision and the self-binding

conduct of credit institutions during the life of the consumer credit contract appear to be

working rather well.

1.2.2 Default interest rate ceilings

1.2.2.1 Default interest rate regulation doctrines

An important tool for keeping credit contracts alive and for enabling consumers to

resume normal payment of instalments, instead of an acceleration of payments and the

resulting over-indebtedness, are regulations which cap default rate to take away the

incentive of additional earnings in default. This discourages the creditor also from

cancellation of the contract. The same applies for any late payments where default

interest is due. The basic dogmatic assumptions about default interest vary according to

the different legal systems. In the majority of Member States, default interest rates

result either directly or indirectly from negotiations between the contracting parties. In

some Member States, the contracting parties may negotiate freely on default interest

rates and set them in the contract. Some Member States limit the right to define the

default interest rate unilaterally but link its level to the agreed contractual rate plus a

margin fixed by law. Other Member States consider default interest as part of a claim for

damages and therefore limit it by market parameters and not by contractual parameters.

In some Member States, both approaches are used. A legal default rate is applicable

unless the parties agree on a higher default interest rate.

1.2.2.2 Statutory default interest rates

A majority of Member States provide statutory default rules with regard to default

interest rates. These interest rates apply when the contracting parties do not agree upon

the interest to be paid upon default and when the law provides for the right of the

contractor to claim (additional) default interest. Statutory default interest rates and

default interest rate ceilings should not be confused. Member States such as Latvia and

Lithuania make use of statutory default interest rates but do not have explicit default

interest rate ceilings. Conversely, in Bulgaria, the Czech Republic, Estonia, Germany,

73 News report produced by Cyprus’ major electronic news site (Sigma) and accompanied by a statement of

the Director of the Legal Department of the House of Representatives Mr. Jonas Nicolaou and other

parliamentarians explaining that liberalization of interest and non penalization of usury are related (in

Greek). See: http://www.sigmalive.com/news/local/170121.

74 Announcements in September 2009 of the works on transposing the EU Directive show that the expected

Directive is considered to ensure common banking practice, which shows that consumer credit interest rates

are decided and agreed upon on a case- by case basis available here in Greek:

http://www.philenews.com/main/75,1,29,0,17335-.aspx.

84 iff/ZEW – Final Report on interest rate restrictions in the EU

Malta and Slovakia the statutory default interest also provides the default interest rate

ceiling. In Austria and Hungary, default interest rate ceilings and statutory default

interest rates exist, but are calculated differently.

Table 15: Statutory default interest rates in EU Member States

Statutory Default IR Member States

No statutory default interest rate Ireland, Romania, UK

Fixed statutory default interest rate Austria, Belgium, Latvia, Lithuania,

Luxembourg, Malta, Spain

Statutory default interest rate based on a

reference rate

Bulgaria, Czech Rep, Estonia, Finland,

Germany, Hungary, Italy, Portugal,

Slovakia, Slovenia, Sweden

Source: Expert Survey.

Statutory default interest rates based on a reference rate are the most common among

Member States. Eight Member States use an objective reference rate, while in Portugal

alone the statutory interest rate is based on the contractual interest rate.

With regard to consumer loans, Portuguese law establishes a statutory default

interest rate based on the contractual interest rate and the rate is determined by

the addition of 2 percentage points to the contractual interest rate (cf. Decree-

Law no. 344/78.) The parties may stipulate higher rates if they do not exceed the

usury limits as defined under the applicable legal provisions. The relevance of

statutory civil default interest rates is therefore limited.

In Germany, different statutory default interest rates apply to consumer mortgage

lending and to other forms of consumer credit. Both are based on an objective

reference index. Under section 288, 497 German Civil Code, the statutory default

interest rate for consumer credit other than mortgage loans is five percentage

points above the base rate as published by the German central bank. The Base

Rate is based on the main refinancing operation rate as published by the ECB. For

consumer mortgage loans the margin applied is 2.5 percentage points.

Seven Member States provide for fixed statutory interest rates.

The Austrian Civil Code, section 1000, provides for a fixed statutory (default?)

interest rate of four percent per year. This interest rate is only applicable to

consumer credit. With regard to commercial contracts, the statutory default

interest rate is stipulated by the Austrian commercial law code, section 352, as

eight percentage points above the base rate published by the Austrian Central

Bank.

In Lithuania, the general rule is provided in Art 6.73 of the Civil code – the

consumer debtor, must pay default interest at the legal rate of five percent per

annum unless any other rate of interest has been established by law or under the

contract.

Ireland, the UK and Romania do not make use of statutory default interest rates. Though

the UK does control default interest through eg. the penalty doctrine, possibly under

licensing, 140A-D, Unfair Terms in Consumer Contracts Regulations 1999, and FSA

conduct of business rules. Though Romania has got provisions on legal interest rates,

these provisions shall not be applied to the legal interest due for any financial obligations

iff/ZEW – Final Report on interest rate restrictions in the EU 85

charged or paid by the National Bank of Romania, other banks, the Office of Savings and

consignments, credit cooperative organizations and the Ministry of Finance, as regulated

under art. 10 of Ordinance no. 9 of 21 January 2000 of the Romanian Government on the

legal interests due for financial obligations, about the legal interest rate.

1.2.2.3 Default interest rate ceilings

Default interest may be capped either by explicit default interest rate ceilings or by the

ceilings provided by general usury legislation. In only a few Member States, there are

neither explicit nor general ceilings applicable to default interest rates. The majority of

Member States uses explicit default interest rate ceilings. Explicit default interest rate

ceilings provide expressly for the upper limit of default interest rates the lender may

apply. Explicit default interest rate ceilings may be relative or fixed. Relative default

interest rate ceilings are based either on an objective reference rate or on the agreed

contractual interest rate. In Member States were there is no explicit default interest rate

ceiling, or in Member States where the explicit default interest rate ceiling depends on

the agreed contractual interest rate, there may be an (additional) limit set by general

usury legislation. In cases where the default interest rate is fixed or based on an

objective market reference rate, there is no room for an additional usury ceiling, because

the explicit ceilings supersede general usury legislation or jurisprudence.

Table 16: Default interest rate ceilings in EU Member States

Default IR ceiling Member States

No explicit default interest rate ceilings

(usury supervision might be applicable)

Ireland, Latvia, Lithuania, Romania,

Sweden

Explicit default interest rate ceiling based

on contractual interest rate (usury

supervision might be applicable)

Austria, Belgium, Denmark, France,

Finland, Greece, Hungary, UK

Explicit default interest rate ceiling based

on an objective reference rate

Bulgaria, Czech Rep, Denmark, Estonia,

Finland, Germany, Italy, Luxembourg,

Poland, Portugal, Slovakia, Spain

(overdrafts only), Slovenia

Fixed explicit default interest rate ceiling Malta

Source: Expert Survey.

1.2.2.3.1 Member States with no explicit default interest rate ceilings

In Ireland, Latvia, Lithuania, Romania and Sweden there are no explicit default interest

rate ceilings. Apart from Ireland, general usury legislation of individual Member States

applies in this context. Subject to this legislation, the contractors may negotiate the level

of interest due in the event of late payment.

To give an example, in Lithuania the parties may agree a default interest rate in the

contract but the courts have the right to reduce unreasonable or obviously excessive

default interest rates under art 6.73 of the Lithuanian civil code. Until 2008, it was

practice for the courts typically to reduce default interest rates to 0.02% per day.

Currently, there is a trend to move away from this practice and increase the level to

which default rates are reduced

86 iff/ZEW – Final Report on interest rate restrictions in the EU

Even if legislation or the courts do not provide for explicit default interest rate ceilings,

there are often stipulations as to the information required on default interest and the

legal consequences of failure to provide such information. If the default interest rate is

not agreed in the contract, a statutory interest rate is often applied:

In Latvia, section 1765 of the civil code states that the interest rate must be

precisely stipulated in the document or transaction. If this has not been done, and in

cases where the law requires the calculation of interest set by law, the interest rate

must be fixed at six per cent per year However in the event of delayed payment in

relation to the purchase of goods or the provision of services where the consumer

interest rate is not set by law, the interest rate is seven percentage interest points

above the basic interest rate set by Central Bank of Latvia on January 1 and July 1.

Sometimes the creditor loses the right to charge default interest at all if there is no

provision in the contract for default interest.

In Ireland, credit agreements must contain details of any costs and penalties in the

event of breach of the agreement by the consumer. Failure to do so renders the

agreement unenforceable against the consumer under section 38 Consumer Credit

Act.

1.2.2.3.2 Default interest rate ceilings based on the contractual interest rate

In Austria, Belgium, Denmark, France, Finland, Greece, Hungary and the UK, there are

default interest rate ceilings based on the contractual interest rate. In some Member

States the default interest rate is limited by the agreed contractual rate:

In the United Kingdom, the Consumer Credit Act 1974, section 93, states that the

interest may not be increased on default:

“The debtor under a regulated consumer credit agreement shall not be obliged to pay

interest on sums which, in breach of the agreement, are unpaid by him at a rate—

(a) where the total charge for credit includes an item in respect of interest,

exceeding the rate of that interest, or

(b) in any other case, exceeding what would be the rate of the total charge for

credit if any items included in the total charge for credit by virtue of

section 20(2) were disregarded.”

In France, default interest rates are regulated correspondingly. According to Article

L311-30 Consumer Code, in the event of default by the borrower, the lender may ask

for immediate repayment of the capital outstanding, plus accrued interest which is

due but unpaid. Until the actual settlement date, the outstanding sums are liable to

interest on arrears at the same rate as the loan. A similar provision is applied on

Consumer Mortgage Credit in default and not yet cancelled (Article L312-22 of the

Consumer Code).

In addition, the lender may ask the defaulting borrower for compensation which,

depending on the length of contract still left to run, is fixed according to a decreed

scale.

There are also legal systems that relate the default interest rate ceiling to the contractual

interest rate by increasing the contractual rate by a margin.

For example, based on the Act of the Governor of the Bank of Greece 2393/1996 07

15, the default interest rate in credit contracts may not exceed the contractual

interest by more than 2.5 percentage points per year.

iff/ZEW – Final Report on interest rate restrictions in the EU 87

There is also a regulation in the Austrian Consumer Protection Act setting a relative

maximum for default interest rates. According to section 6 paragraph 1 Z 13, the

default interest rate on consumer loans can not be more than 5 percentage points

per annum higher than the contractual interest rate.

1.2.2.3.3 Default interest rate ceilings based on an objective reference rate

The legal systems in Bulgaria, the Czech Republic, Denmark, Estonia, Finland, Germany,

Italy, Luxembourg, Poland, Portugal, Slovakia, Spain (only for overdraft) and Slovenia

provide for an explicit default interest rate ceiling that is based on an objective reference.

Sometimes like in Bulgaria, Estonia or Germany, the default interest rate ceiling is

conformant to the statutory interest rate that is to be provided when no agreement upon

the default interest rate has been made.

In Estonia, any penalty for late payment required from the consumer may not

exceed the amount stipulated by the Law of Obligations Act. The Law of

Obligations Act provides the basis for calculating the legal default interest: the

last interest rate applicable to the main refinancing operations of the European

Central Bank before 1 January or 1 July of each year + 7 per cent per year is the

interest rate on late payment (hereinafter referred to as the legal default interest

rate).

In Germany, there are provisions regarding consumer credit in sections 497, 503

Civil Code. Section 497 (1) “Treatment of default interest, crediting part

performance”, states:

“To the extent that the borrower is in default in making payments owed on the

basis of the consumer loan contract he must pay interest under section 288 (1) on

the amount owed; this does not apply to real estate loan contracts. With regard to

these contracts, the default rate of interest per year amounts to two and a half

percentage points above the base rate of interest. In an individual case, the

lender may prove that the damage was greater or the borrower may prove that

the damage was less.” Section 288 German Civil Code states:

“Default interest (1) Any money debt must bear interest during the time of

default. The default rate of interest per year is five percentage points above the

basic rate of interest.” Borrower and lender may deviate from the provisions

quoted above, but not to the disadvantage of the consumer.

In Spain Art 1108 Spanish Civil Code on “legal interest” relates default interest for

late payment or late fulfilment of an obligation to the legal interest or statutory

interest rate set each year by the Budget law. It is currently 4% until 31st

December 2010. The Law 58/2003 on taxation fixes default interest in relation

with tax (currently at 5%). Law 3/2004 of 29th December to fight against late

payments in commercial transactions, following Directive 35/EC of 29th June 2000

on combating late payments in commercial transactions says that the level of

interest for late payment ("the statutory rate"), which the debtor is obliged to

pay, shall be the sum of the interest rate applied by the European Central Bank to

its most recent main refinancing operation carried out before the first calendar

day of the half-year in question ("the reference rate"), plus at least seven

percentage points ("the margin"), unless otherwise specified in the contract.

There are also Member States where the default interest rate ceiling is calculated on the

basis of the statutory default interest rate, which itself is based on a reference index.

In Slovenia, if the default (or contractual) interest rate exceeds the statutory level

of default interest rate by 50 %, it is considered usurious, unless the creditor

88 iff/ZEW – Final Report on interest rate restrictions in the EU

proves that he did not use the debtor’s emergency or difficult material position,

his lack of experience, frivolity or dependence or that the benefit that he or

someone else received did not significantly exceed the benefits of the other party

(Presumption of usury, Article 377). The statutory default interest rate is basically

17 %, but the Government is authorised to change the statutory default interest

rate, if the economic situation changes (Article 2). The government has so far

carried out several reductions of the default interest rate based on the decision to

change the statutory default rate (OJ 1/07); after the introduction of the euro it

was decided that the default interest rate should be set at 8 % (real interest rate,

determined in Code of Obligations) + ECB’s main refinancing rate.

The default interest rate ceiling may also be based on the usury ceiling which in turn is

based on an objective reference index.

To give an example, in Italy, Law 108/1996 fixes the usury ceiling not only for

contractual interest rates, but also for default interest rates. As mentioned above,

Banca d’Italia calculates the average contractual rates – TEGM – for various types of

credit. Afterwards the Ministry establishes officially the average additional value of

default interest rates in the credit market. The default interest rate ceiling is

therefore based on the TEGM and then increased by the average additional value of

default interest rates in the credit market provided by the Ministry of Economy,

multiplied by 1.5. (TEGM+extra value given as a percentage point; majored by

50%).

1.2.2.3.4 Fixed default interest rate ceilings

Explicit default interest rate ceilings that are neither based on the contractual interest

rate nor on an objective reference point are not very common in Europe.

In Malta, there is a general rule under the Civil Code, Article 1139 which states

that if the obligation has as its object the payment of a determinate sum, the

damages ensuing from a delay in its execution may consist only of interest on the

capital sum due calculated at the rate of eight per cent per annum.

1.2.2.4 Overview: Levels of default interest rate ceilings and statutory default

interest rates

A full breakdown of the levels of default interest rate ceilings as well as of the levels of

statutory default interest rates (that are applied when the contracting parties do not

agree upon the default interest rates) across the Member States is provided in the table

on the following pages. The table does not provide information regarding the average

default interest rates applied on default. However, since default interest rates other than

contractual IRR provide for a rate which is seen as an adequate expression of the

damage the creditor incurs providers in general use the legal default rate as their

contractual default interest rate. This is why the average default interest rates

correspond to the ceilings given in the table. The table also contains calculated examples

for a non-mortgage instalment loan, using reference data as of March 2010, and

assuming contractual interest rate of 8 percent.

iff/ZEW – Final Report on interest rate restrictions in the EU 89

Table 17: Overview of default interest rate regulation in EU Member States

Member State Default-IR Ceiling Statutory default IR

for consumer credit

Default

IR

Ceiling

Example

75

Comments

Austria Contractual IR + 5 pp 4 % (fixed) 13 %

Belgium APR majored by 10

percent; ie. APR * 1.1

3.25 % (fixed) 8.8 %

Bulgaria EURO Credits:

3-month-Libor + 10 pp

Credits in Levs:

BR BCB + 10 pp

EURO Credits:

3-month-Libor + 10 pp

Credits in Levs:

Base rate + 10 pp

10.6 % BRBCB =Base Rate Bulgarian

National Bank

Cyprus penalty fees applied to

overrunning on the

current account are

limited. Also early

repayment fees in

mortgage loans (3%)

no no

Czech

Republic

Repo-rate 2T CNB + 7pp Repo-rate 2T CNB + 7pp 8.0 % CNB = Czech National Bank

Denmark LR + 7pp or contractual

IR;

whichever is higher

n/a 8.05 % LR = Lending Rate Danish

National Bank

Estonia MRO + 7pp MRO + 7pp 8.0 % MRO = Main Refinancing

Operation rate European

Central Bank

Finland First 6 months of default:

BR + 7pp or contractual

IR;

whatever is higher

Default from the 7th

month:

BR + 7pp

BR + 7pp 8.0 % BR = Base Rate European

Central Bank

France Mortgage credits:

Contractual IR + 3pp

None Mortgage credits:

Contractual IR

n.a. 8.0 % There are other stipulations

with regard to rescheduled or

cancelled contracts.

Germany Mortgage credits:

BZ + 2.5pp

None Mortgage credits:

BZ + 5.0pp

Mortgage credits:

BZ + 2.5pp

None Mortgage credits:

BZ + 5.0pp

5.12 % BZ “Basiszinssatz” = Basic

Rate German Federal Bank,

based on Main Refinancing

Operation rate ECB

Greece Contractual IR + 2.5pp n.a. 10.5 %

Hungary Contractual IR + 1/3

BRHNB

BRHNB 9.8 % BRHNB = Base Rate

Hungarian National Bank

Ireland None. The default IR

must be agreed and

stated in contract.

/ / Usury legislation might be

applicable

Italy (TEGM+extrapp) majored

by 50 %;

ie. (TEGM+ extrapp)*1,5

MRO + 7pp 21.9 % TEGM = Basic average IRs

Italian Central Bank

(differentiated by credit

types)

Extrapp = average additional

value of default interests in

the credit market

MRO = Main Refinancing

Operation rate European

Central Bank

75 Non-mortgage instalment credit; contractual interest rate set by 8 percent, reference rates as of March

2010.

90 iff/ZEW – Final Report on interest rate restrictions in the EU

Member State Default-IR Ceiling Statutory default IR

for consumer credit

Default

IR

Ceiling

Example

75

Comments

Latvia None. The default IR

must be agreed and

stated in contract.

6 % (fixed) /

Lithuania None. The default IR

must be agreed and

stated in contract.

5 % (fixed) / Usury legislation might be

applicable

Luxembourg Legal IR + 3pp (where

judgement)

3.5 % (fixed, yearly

amended)

6.5 %

Malta 8 % (fixed) 8 % (fixed) 8.0 %

Netherlands Statutory IR + 12pp 3 % (fixed, amended 5

times since 2002)

15.0 % Default IR ceiling is the same

as Contractual IR Ceiling

Poland SL * 4 n.a. 20.0 % SL = Lombard Rate National

Bank of Poland (“Stopa

Lombardowa”)

Default IR ceiling is the same

as Contractual IR Ceiling

Portugal Average APR majored by

33 %;

ie. Average APR * 1,33

Contractual IR + 2pp 18.9 % APR computed by central

bank

(differentiated by credit

types)

Default IR ceiling is the same

as Contractual IR Ceiling

Romania None None / Statutory default IRs and IR

Ceilings are not applicable for

banks according to art. 10 of

Ordinance no. 9 of 21

January 2000 of the

Romanian Government.

Slovakia MRO + 8pp MRO + 8pp 9.0 % MRO = Main Refinancing

Operation rate European

Central Bank

Spain Overdraft Credit:

Legal IR * 2.5 =10.0 %

4 % (fixed, annually

amended)

/ Legal IR determined on the

basis of financial market

trends

Slovenia Statutory Default IR

majored by 50 %;

ie. Statutory Default IR *

1.5

Statutory Default IR =

Statutory IR (8 %) +

MRO

13.5 % MRO = Main Refinancing

Operation rate European

Central Bank

The calculation method of

Default IR Ceiling is legally

disputed.

Sweden None. The default IR

must be agreed and

stated in the contract.

RRI + 8pp / RRI = Reference Rate of

Interest, based on Main

Refinancing Operation rate of

the Central Bank of Sweden.

Usury legislation may be

applicable under penal law

UK Contractual IR None 8.0 %

Source: Expert Survey.

iff/ZEW – Final Report on interest rate restrictions in the EU 91

1.3 Other cost relevant factors

The question of which costs payable by a consumer should be incorporated into the credit

price under EU consumer credit regulations has been dealt with in another project report

to the European Commission in 1998.76 The information collected from the experts in this

study confirm that there has been an ongoing shift from interest to other cost factors in

consumer credit.

Although a minimum standard in the calculation of the APR has been reached Europewide

there are general systems which allocate cost elements outside the in so far

unchanged definitions of the APRC in Directive 87/102/EEC and 2008/48/EC. While in all

Member States payments concerning services which are directly connected to the credit

like administration and brokers’ fees insurance fees irrespective of the purpose of the

insurance, fees for bank accounts and bank cards are not included in practice.

Endowment life insurance credit has no integrated APRC in which the premiums and

payments concerning the endowment are incorporated. Such products are nearly

unknown in consumer credit as regulated by the CCD 2008 but concern primarily

mortgage loans. In this area EU-law does not require inclusive pricing neither does

national law.

In some countries like Austria, France and Greece, taxes are imposed on the extension of

credit.

Payment Protection Insurance premiums in consumer credit have generally to be

advanced in one single payment. This opens the product for additional financing since the

consumer needs a credit to be able to pay this premiums in a lump sum. This practice

has raised concern especially in the UK. But still the general exemption if the insurance is

not ”non obligatory“ from its integration into the APRC also counts for these finance

charges. Payment Protection Insurance has thus developed into a general outsourcing of

credit risks at the cost of the consumer. The products are disadvantageous, extremely

costly and applied inappropriately and to that extent, far too often. It seems as if the

provisions paid to the banks for the extension of insurance products have become a main

source of additional income for credit providers, escaping competitive forces.

Instead specifically charged bank account fees are not yet an economic problem. But in

credit card credit the enormous fees charged for cash withdrawal with these cards can be

seen as an additional cost for those customers who have no own bank account and use

the cash withdrawal facility to get an easy and immediate short term instalment credit

form the credit card issuer. If for example the fee for cash withdrawal is 3% of the credit

amount this fee adds to the cost of the credit if (a) the credit card account has no assets

and (b) it provides for own credit. Since only a few credit card accounts allow assets and

since those credit cards are most used by low income households for access to small

amounts of money (“payday loans”) such cash withdrawal fees increase the cost of credit

for these borrowers without leaving traces in the APRC.

Combined, endowment products which divert repayments of the credit into a form of

savings agreement with lower interest returns in the savings than is charged in the credit

like „endowment capital life credit“, „secured credit cards“, instalment or overdraft credit

where assets are requested as a security may increase the amount of interest due

through artificial additional demand for credit which the yield of the investment product

does not compensate for. Such products may be as usurious as other products although

its APRC will look significantly better.

76 Reifner, U. et al. (1998).

92 iff/ZEW – Final Report on interest rate restrictions in the EU

Another form of circumvention is zero interest credit extended by banks owned by

automobile companies. They hide that the buyer of the financed cars looses the cash

payment premium which may amount to as much as 15%. One could call this a reverse

cost element similar to a disagio in which interest is defined as capital and thus made

invisible.

Already the 2002 Draft of a Consumer Credit Directive tried to address these problems

by replacing the “voluntary” element as the core condition for “linked products”.77

In its recital to Art. 12 of the draft a vast definition of all cost that should be included into

the APRC had been favoured: “The total cost of the credit must include all costs,

including borrowing rate plus all the other indemnities, commissions, taxes and charges

of any kind that the consumer is required to pay for the credit regardless of whether

these costs are payable to the creditor, to the credit intermediary, to the competent

authority levying the taxes or to any other third party authorised to receive payments

following the brokering or conclusion of a credit agreement or surety agreement.”

Instead Directive 2008/48/EEC referred back to Directive 87/102/EEC where the question

which cost from ancillary services had to be included was made dependent on whether

such contracts had voluntarily been concluded.78 Also circumvention through endowment

credit and bank account fees had been taken into account.

1.3.1 Uniform definition of interest (rates)

Price disclosure and price restrictions have to refer to a price which is uniformly and

objectively defined by law. In price disclosure law as well as in price restrictive law

providers would have an unjustified advantage if they could use a price definition which

would allow them to show a lower price than their actual cost structure would justify.

Especially the possibility to split prices onto different contracts, to exclude certain cost

elements, to use mathematical formulas which lead to incomparable prices etc. are of

concern both for price disclosure law and IRR.

This problem has been acknowledged in all consumer related Directives with a wording

similar to Art. 22 (3) of the CCD 2008 where it reads: “3. Member States shall further

ensure that the provisions they adopt in implementation of this Directive cannot be

circumvented as a result of the way in which agreements are formulated, in particular by

integrating drawdowns or credit agreements falling within the scope of this Directive into

credit agreements the character or purpose of which would make it possible to avoid its

application.” This general principle of all consumer protection law either substantive or

procedural has been further elaborated in the Directive itself with its gradual

improvements in 1998 and 2008.

IRR stand in the legal tradition of price regulations expressed in the laesio enormis.

Usury and good morals as well as substantive fairness principles focus on the

comparatively high amount of money a consumer has to pay for a service or a good. But

77 Proposal for a Directive of the European Parliament and of the Council on the harmonisation of the laws,

regulations and administrative provisions of the Member States concerning credit for consumers (2002/C

331 E/39) COM(2002) 443 final — 2002/0222(COD) (Submitted by the Commission on 11 September 2002)

OJ C331 E/200; for other drafts and alternative proposals see http://www.responsiblecredit.

net/index.php?id=1884.

78 Art. 3 (g) of Directive 2008/48/EEC includes all cost “which are known to the creditor, except for notarial

costs; costs in respect of ancillary services relating to the credit agreement, in particular insurance

premiums, are also included if, in addition, the conclusion of a service contract is compulsory in order to

obtain the credit or to obtain it on the terms and conditions marketed”. Art. 2 (g) of the 2002 Draft read:

(g) ‘total cost of credit to the consumer’ means all the costs, including borrowing interest, indemnities,

commissions, taxes and any other kind of charge which the consumer has to pay for the credit;” and Art. 12

(2) stated: “Costs relating to insurance premiums shall be included in the total cost of the credit if the

insurance is taken out when the credit agreement is concluded”.

iff/ZEW – Final Report on interest rate restrictions in the EU 93

in credit the price in money units is incomparable because it depends on two other

factors which differ from contract to contract: borrowed capital and time. This is why the

price of the credit has to be represented by the interest rate which incorporate time and

capital amount and offers thus a standardised from of prices.

The interest rate is therefore not the price of the credit but a parameter which in the

form of the borrowing rate has been created in practice to calculate those parts of the

credit cost which depend directly on the time of the loan.

Since other fees and obligations add to these costs, CCD 2008 has regulated a second

parameter, the APRC, to give consumers a basis for comparing the different prices of

credit on the market. Recital 19 states: “To ensure the fullest possible transparency and

comparability of offers, such information should, in particular, include the annual

percentage rate of charge applicable to the credit, determined in the same way

throughout the Community.” It should according to recital 20 of CCD 2008 “comprise all

the costs, including interest, commissions, taxes, fees for credit intermediaries and any

other fees which the consumer has to pay in connection with the credit agreement,

except for notarial costs.” Since also the method of calculation can be used to represent

the same cost differently in the APRC also the method of calculation has been

harmonised in Annex 1 of the CCD 2008 with the growth formula79 which in distinction

from the borrowing rate80 also fixes the period where accrued interest should be

compounded.

The table below shows the differences between both rates and reveals that effective IRR

should focus on the APRC instead of the borrowing rate. The table also reveals that this

process from the borrowing rate to the APRC has not yet been fully accomplished.

Traditional civil codes use borrowing rates for their historical regulation of the legal rate

as well as for their rules on anatocism.81 Also default rates are usually defined in the form

of borrowing rates in practice so that its regulation is adapted to it. Old fixed rate ceilings

are mostly related to the borrowing rates since the emergence of a growth rate related

form of calculation as it is present in the APRC has only emerged together with the

spread of computer and their ability of approximations through iteration as described in

the mathematical formula to Annex 1 of the CCD 2008.

79 The growth formula can be written in the form of C1 = C0 * (1 + i)t wherein the initial capital C0 has been

grown into C1 aduring the time t at a rate of (1+i) where i represents the interest rate. The compounding

period of 1 year is hidden in the exponent t since the calculation is based on years and therefore t is divided

by 1 to mark this compounding period.

80 Instead the borrowing rate is mathematically wrongly calculated in the form of i = cost/( C0 *t).

81 See ie. section 246-248, 289 German Civil Code.

94 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 18: Borrowing rate and APRC

Credit Related Cost Borrowing Rate (Art. 3 (j)) APRC (Art. 3 (i))

Elements/parameters:

“Interest” Yes Yes

Closing fees No Yes

Broker fees No Yes

PPI Insurance fees No Mostly no

Finance Charge on

Insurance fees

No Mostly no

Cost of Refinancing No No

Standardised Method of

Calculation

No Yes

Early Repayment

Charges

No No

Compounding period Monthly or arbitrary by

payment periods

Standardised one year

Disclosure % p.a. Yes Yes

Used for:

Legal interest rate Yes No

Usury ceilings No Yes

Default interest rates Yes Rarely

Anatocism Yes No

Variability Yes No

Source: Expert Survey and iff.

But indeed the problems of price disclosure and price regulation are not much different. A

price which omits important parts of the cost prevents a rational choice but also the

equal and just application of IRR. This is why all countries which have introduced direct

IRR on the interest rate level use the APRC and no longer the borrowing rate.

For IRR a clear and comprehensive interest rate is even seen as more important than in

price disclosure law where the underlying idea of capping interest rates is to prevent

insolvency through high cost credit. From an insolvency perspective the monthly rate to

be paid is the main factor that determines the disposable income and household liquidity.

Thus before Directive 98/7/EC harmonised the APRC German courts still included the

premiums of PPI with half of its value into that APRC which had to be compared with the

average market rate while the then existing order for price disclosure excluded such

premiums from it.

1.3.2 Anatocism and compounding

Traditional IRR were based on a quite simple understanding of interest which is still

visible in its standardised form of % p.a. which assumes that the price of a credit can be

expressed simply by relating the cost of the credit to one year and €100. The

mathematical problems of such simplistic assumptions and the modern solutions offered

by the introduction of the APRC are dealt with below.82 Since anatocism is a very old

principle of IRR it is still related to the old form of interest rates now called borrowing

rate in the CCD 2008. For this borrowing rate the period of interest compounding

82 See below at pp 91ff.

iff/ZEW – Final Report on interest rate restrictions in the EU 95

remains undefined if not by special contractual prescriptions. This problem is solved in

the APRC where the mathematical formula assumes a compounding period of 1 year. In

so far all those who use the APRC to define a usury ceiling do not face the problem of

anatocism but even require anatocism as it is the case for Annex 1 of the CCD 2008.

But most systems still use a borrowing rate instead of the APRC to calculate the interest

due. In this case compounding interest at intervals different from one year leads to

enormous differences in the amount of interest due. Besides in default the traditional

interdiction of anatocism keeps the amount of default interest under control. Where the

borrowing rate is used for interest calculation the total cost of credit not only depends on

the size of this rate, but also on the interest calculation method and on the rules

providing for repayment of the debt.

With regard to the interest calculation, rules avoiding or forbidding compound interest

have a great impact on the total cost of credit. Compound interest arises when interest is

added to the principal, so that from that moment on, the interest that has been added

also earns interest. This method is called compounding or anatocism. Anatocism conflicts

with the Roman law principle, forbidding interest charges on interest. The actual position

amongst the Member States is very diverse and depends on individual levels of credit use

and consumer protection. Anatocism seems mathematically outdated since if taken

seriously would provide for an arbitrary compounding period which is the lifetime of a

loan. As this lifetime may vary no loan would be comparable to another loan with a

different lifetime. This has led to its gradual abolition through exemptions for current

accounts, overdrafts, revolving credit and credit-card borrowing. Many ways have been

developed to overcome the irrationality of this principle but as long as instead of the

arbitrary borrowing rate the APRC is not used exclusively also for the calculation of the

interest due the questions remain unsolved.

While mathematical experts83 reject anatocism as an outdated irrational legal form of

mathematics, culturally anatocism is still seen as a means to prevent the exponential

increase of debts through unpaid interest which bear interest again.84 This is why with

regard to contractual interest most countries have gradually abolished its effects or at

least allow forms which could be called circumvention. Instead the focus of this principle

are now default interest where alternatives have already emerged which no longer deny

that any interest calculation needs compounding of interest but achieve the goals of this

debtor protection principle for example by prescribing different accounts for interest and

principal in default as the German expert reports.

In some southern European Member States such as Italy, that law remains in force and

has been revived for consumer protection purposes. Further, rules providing for

amortisation of a debt affect the total cost of credit. Amortisation (from Middle English

“amortisen” – “to kill”) is the process of decreasing an amount over a period of time. If a

debt is to be paid back in instalments, the payments consist of interest and part of the

principal. Therefore, after a certain portion of each payment is applied to the interest on

the debt, any balance reduces the principal. In a situation of default, the instalment may

not be sufficient to cover both interest and the principal. To avoid the accumulation of

interest, some legal frameworks therefore have rules providing for the order in which

interest and the principal have to be charged against the instalment.

83 See for a fierce rejection of the idea of anatocism Seckelmann, R. (1989); Seckelmann was also the main

expert for DG Sanco fort he 1998 amendment to the Consumer Credit Directive.

84 For this discussion see Reifner, U (1992), pp.227-243.

96 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 19: Overview of rules on anatocism in the EU

Member

State

Contractual interest

allowed if convened in

advance

Default interest

allowed

Allowed after

judicial procedures

Austria Yes No (no more than the

debt)

Yes

Belgium No Yes (agreement; 1

year)

Bulgaria Yes (Bank loans only, not

private loans)

Yes

Cyprus Yes (but only twice a year)

Czech Rep No No No

Denmark No

Estonia Yes

Finland No

France Yes Yes Yes (judicial decision

or agreement; 1

year)

Germany No (except for current

account)

No (but not effective) Yes

Greece No

Hungary Yes

Ireland Yes

Italy No Yes (if agreed and six

month)

Yes (if agreed and six

month)

Latvia Yes (but only after one year)

Lithuania Yes (if agreed and not

against good faith)

Luxembourg Yes (but only for one year;

exception current account)

Malta No (but exception if

commercial use and for one

year only)

Netherlands Yes (but limited by the

maximum interest)

Poland No No Yes

iff/ZEW – Final Report on interest rate restrictions in the EU 97

Portugal No Yes (if agreed after

default and for one year)

Yes (for one year)

Romania No

Slovakia Yes

Slovenia No (but higher interest rates

are permitted)

Yes

Spain No

Sweden Yes (but good morals)

UK Yes

Source: Expert Survey.

The following selected information from the experts shows the diversity of answers given

to the old principle of anatocism.

In Austria, in the absence of any other agreement in the contract, section 1335 Austrian

Civil Code (ABGB) applies. The creditor cannot claim default interest exceeding the sum

of the original debt if he does not take legal action. From the moment the creditor takes

legal action against the debtor he can claim default interest again, which can then result

in a sum larger than the original debt.

In Belgium, compound interest is restricted by the Civil Code, the Mortgage Credit Act

and the Consumer Credit Act. Under Article 1154 of the Belgian Civil Code, it is possible

for interest to attract further interest, but this possibility is subject to two conditions. The

first relates to the method to be used. This form of interest may be imposed either by a

Court decision, or by a specific agreement between the parties. The second condition

imposes a minimum time limit of one year over which compound interest is chargeable.

Art 14 Consumer Credit Act establishes that there is an obligation to determine the exact

amount of credit in the credit agreement. Art. 10 Mortgage Credit Act only provides for

interest calculated on the principal sum outstanding. General civil law also provides rules

on amortisation. This provision (art. 1254 Civil Code) applies to all credit types. Unless

otherwise agreed upon by the creditor, no payments can be allocated to the outstanding

principal prior to reducing the interest (art. 1254 BW). For consumer credit, the

regulation is somewhat more complex. As long as the loan is not terminated, the above

rule in article 1254 BW applies, ie. payments should first be allocated to the interest and

only afterwards on the remaining capital. However, if the consumer loan is terminated,

under article 27f, §5 WCK payments may be allocated first to the remaining capital and

then to the interest.

In the Czech Republic anatocism is not allowed. The right to demand an ancillary right

from an ancillary right by a decision of the Highest Court of the Czech Republic from No.

35 Odo 101/2002 because this option is neither granted by the Act No. 513/1991, Coll.,

as amended, the Commercial Code (“Commercial Code”) nor the Act No. 40/1964, Coll.,

as amended, the Civil Code (“Civil Code”).

The Danish expert reported that there are no legal provisions governing the repayment

of debt in general. Restrictions on default interest imply that it is necessary to have a

separate account for default interest. In general the borrower decides how the payments

should be divided between principal and interest (in a very few statutory cases payment

is debited to the interest first).

98 iff/ZEW – Final Report on interest rate restrictions in the EU

In Estonia, the Law of Obligations Act provides regulations governing restrictions on the

calculation method of default interest. A penalty for late payment cannot be required for

a delay in the payment of interest (anatocism). Agreements which derogate from such

requirements to the detriment of the obligor are void. The above does not preclude or

restrict the right of the obligor to claim compensation for damage caused by a delay in

the payment of interest. With regard to amortisation, the Law of Obligations Act also

stipulates the order of the repayment of debts in default. If, on the basis of a credit

contract, a consumer has made a payment which is insufficient for the performance of all

obligations which have fallen due, the payment must cover firstly, the expenses incurred

in collection of the debt, secondly, the principal sum owed, thirdly, interest, fourthly,

other obligations.

In Germany, as in the Northern Member States, anatocism as a historical principle has

been largely abandoned. The Civil Code prohibits a contractual clause allowing interest to

be added “in advance” to the principal sum outstanding. However, some exceptions have

been created. First of all, exemptions apply to current accounts. Another exemption has

been applied to a disagio, a situation where interest is prepaid and thus added to the

principal. As German law allows an agreement that outstanding interest should bear

interest again, it is quite difficult to distinguish whether such an agreement was made in

advance. With regard to default interest alone, the German Civil Code, Section 289

prohibits the compounding of interest on default interest. The right of the obligee to

compensation for damage caused by the default remains unaffected. Notwithstanding

these exceptions, it must be emphasised that Germany has also revived the idea of

limiting pyramid debts through specific legislation governing interest on consumer credit

in the event of default. According to section 497 (2) German Civil Code interest incurred

after default has occurred must be booked to a separate account and may not be paid

into a current account together with the amount owed or other claims of the lender. With

regard to such interest, the lender may claim interest on interest as further damage but

limited to the amount of the statutory rate of interest, which amounts to up to 4 percent

(see section 246, section 289 sentence 2 German Civil Code).

According to the French Bankers’ Association interest capitalization is authorized in

France. However, it is regulated in respect of interest due: this interest may only

produce interest if, by virtue of the agreement, this interest is due for at least an entire

year (Article 1154 of the Civil Code). The draft legislation on consumer credit reform

makes express reference to this rule, which is nevertheless already generally applicable,

for consumer credit. Moreover, there is other legislation in existence capping the costs

that the Banks can collect (incident charges, etc.) but these laws do not relate to interest

rates strictly speaking.

There is comprehensive regulation in Greece restricting anatocism. Credit contracts

made before 1998 may provide that interest in default is anatocised after the first day of

default every 6 months (or longer period if agreed – a shorter period is forbidden). If

there is no agreement in the credit contract regarding anatocism, then anatocism may

take place according to art.296, which provides that interest on interest may be paid only

if agreed or only after issuing proceedings. But in both cases, the interest due may refer

to a period of at least a full year. As an exception to this rule, art.111 of the Introductory

Law to the Civil Code provides for the possibility of anatocism every six months. If they

do not provide for anatocism, anatocism may take place every 12 months. After 1998,

Law 2601/1998 is applicable, (Official Journal A 81/15.4.1998) article 12, under which

interest may be charged on the default interest, if this is agreed by the parties, starting

from the first day of default. The resulting interest is added to the amount of the capital

due at intervals which may not be less than six months. As regards contracts already

concluded by that date, the same law provides for automatic anatocism every 6 months,

even if the contract did not make any provision at all for anatocism.

In Italy, anatocism is forbidden with regard to default interest under art. 1282 civil code

and confirmed in fairly recent but authoritative jurisprudence. Anatocism is permissible

iff/ZEW – Final Report on interest rate restrictions in the EU 99

only in litigation, starting from the date of the judicial summons. Italy has a long history

regarding provisions on anatocism. In fact, art. 1283 of the civil code partially allows

anatocism, as an exception to the general prohibition. This rule provides that unpaid

interest may generate further interest, but only from the date of the summons or, in

relation to a contract, subsequent to the date of default, and only if the interest has been

due for six months. Generally the Italian civil code forbids the compound capitalisation of

interest, but, notwithstanding these rules, in Italian banks the contractual use of

anatocistic clauses has been common for nearly half a century, especially in contracts

governing bank accounts, because this status quo was unanimously supported by case

law. In 1999, the Italian Supreme Court -Corte di Cassazione - changed its approach in

certain decisions referring to contracts governing bank accounts The Court confirmed

that no bank practices can repeal art. 1283 c.c., thereby declaring the most common

anatocistic clauses void. To avoid contradictions between the different rules and credit

operators, the Italian legislator modified art. 120 of TUB (D.Lgs. 385/1993, the main

source of rules for banks and other credit operators) with the D.Lgs. 342/1999,

establishing the principle of equal compounding of interests, both bearing and payable,

saving the effects for past contracts. This was declared unconstitutional at first instance

by the Corte Costituzionale and then corrected by the legislator. The Supreme Court has

confirmed its 1999 change of direction in the interpretation in some important decisions

(Cass. 17813/2002, referring to loans and the Cass. S.U. 21095/2004). The court

considers void every anatocistic clause even in contracts stipulated before 1999, and that

every method of composite compounding interest that has the same purpose of

anatocism is also void.

In Luxembourg, Article 1154 civil code prohibits anatocism when interest is calculated

on a period of less that one year. However, this prohibition is not applicable when carried

out in relation to a consumer’s current account. Furthermore, there is a rule that

repayments in default are first applied to reduce the principal.

In Portugal, anatocism is generally forbidden and it is only allowed by Portuguese law in

two circumstances: as an agreement between the borrower and the lender, posterior to

the maturity of interest, or as a Court notice to the debtor in which he is informed that

he must capitalise interest earned or proceed to payment under penalty of capitalisation.

The law also stipulates that only interest corresponding to a minimum period of one year

may be capitalised. Nevertheless, these rules will only be applicable if they are not

contrary to rules or private uses of commerce (cf. article 560 of the Civil Code) leaving a

possibility for bank anatocism – which is very common practice in Portugal and allowed

by the majority of judicial decisions. Moreover, there is a special law concerning banking

anatocism (articles 5 and 7 of Decree-Law nr. 344/78) which establishes that it is

forbidden to capitalise interest corresponding to a period of less than three months.

According to the Civil Code of Malta, no interest may be charged until the due date and

from the date of legal proceedings claiming such interest, or from the date of a

subsequent agreement. Statute in Malta follows the Code Napoleon (art. 1154) and

allows for no exception to the rule. However, jurisprudence in Malta has always held that

this provision of the Civil Code may be derogated through commercial usage (eg. Edwin

Vassallo v. Salvatore Ballucci 30.04.1947 Civil Court, First Hall). Maltese courts have

however followed French Courts which (except for some dissenting judgements) have

held that the Civil Law rule does not apply to commercial debts. The French Court of

Cassation has held that the capitalisation of interest in a current account takes place de

plein droit or ipso iure when the interest is annual, without the need for a judicial

demand or a subsequent agreement (Dalloz, Compte Courant, judgement of 26.05.1812,

Cour D'Orleans 26.08.1840, Cour de Bordeaux 09.08.1940, para 74 and 96, pages 585

and 590 of Repertoire de Legislation, Vol XI. Paris Edition, Bureau de la Jurisprudence

Generale, 1849). The same principle has been adopted by the Maltese courts - Negte.

Alfonso Ellul v. Negte. Giovanni Mifsud 12.11.1901 Commercial Court, Vol. XVIII.iii.53;

Onor. Alfonso Maria Galea et. Ne. V. Ferdinando Hass, 19.04.1926 Kollez.

100 iff/ZEW – Final Report on interest rate restrictions in the EU

Vol.XXVI.iii.633; Negte. Francesco Tabone v. Luigi Piovano 11.05.1929 Commercial Court

confirmed by the Court of Appeal on 29.06.1929.

1.3.3 Variability of interest rates

The relevance of an IRR on the variability of the interest rates (ie. limits to the variation

of rates) in a given country will first of all depend on the prevalence of credit extended at

a variable rate of interest in the overall credit market. This proportion of variable rate

credit, both in mortgage and non-mortgage consumer credit markets, varies considerably

between Member States, for example, while in Slovakia almost all types of consumer

credit are sold at a variable interest rate, in Portugal instalment credit and revolving

credit is more commonly found to have a fixed interest rate. In the UK, all types of loans,

mortgages, credit cards, personal loans can be sold with caps. There are also variations

within Member States between the different forms of credit, eg. in Denmark, though the

market share of variable interest bank loans is estimated at approximately 90 percent of

all bank loans, only 50 percent of all mortgage loans are at variable rates.

IRR has the same effects on variable rate credit as fixed rate credit. Each variable rate

credit starts at the time of the conclusion of the contract with a contractual interest rate

just as it is the case for fixed rate credit. Equally, in default, the default interest rate

replaces the interest rate for variable and fixed rate credit alike. Since IRR is always

attached to the initial contractual interest rate, there is no difference between the two

interest rate forms with regard to form and effects of IRR.

But this does not make specific regulations on variable rate credit totally impossible.

There is only one example in the EU where a Member State has implemented a special

rate ceiling concerning the variability rate itself. In Belgium, for mortgage loans, the

variable (also called floating) interest rate must not only be linked to a reference rate

(art. 9, §1, 3° WHK) like in many other countries but can only increase by a maximum of

2% during the first 3 years of the mortgage loan (art. 9, §1, 8° WHK), thereby protecting

these borrowers from large shifts in interest rates. If the contracting parties agree on a

floating interest rate, only one floating interest rate per mortgage loan is allowed (art. 9,

§1 WHK). The reference indices must be chosen as a function of the period between two

alterations of the interest rate. The list with the reference indices and the calculation

method is determined by the King (by Royal Decree) after consulting the CBFA (art. 9,

§1, 3° WHK), and is published monthly by the Securities Regulation Fund

(“Rentenfonds”). The floating interest rate is linked to a reference rate, more specifically

to reference indices (art. 9, §1, 3° WHK; cf. infra).

Equally if the change of interest rates for revolving credit in Belgium exceeds 25% in

relation to the original borrowing rate and if the revolving credit account was agreed for a

term of at least 1 year, the consumer has a right to terminate the revolving credit

account within 3 months as from notification by the credit provider (art. 60 WCK).

However, in contrary to fixed rate credit where one rate dominates the lifetime of the

credit, in a credit concluded at a variable rate of interest, the interest rate may be

changed unilaterally in the subsequent period after the conclusion of the contract. Since

only the initial rate is the contractual interest rate the official interest rate ceiling only

applies to this rate. This may induce suppliers to provide so-called teaser-rates where a

variable rate credit carries a low initial interest rate at the beginning which is

consequently increased so that the overall average interest rate of the contract may well

go over the rate ceiling.

This problem is not specific to IRR, and is even more striking with regard to price

disclosure where a provider may cheat competitors by luring consumers into his teaser

rates while exploiting them later. This is why Article 5 (1) (f) CCD 2008 (similar wording

iff/ZEW – Final Report on interest rate restrictions in the EU 101

in Article 6 (1) (e); 10 (2) (f); 10 (5) (e)); addresses “the conditions governing the

application of the borrowing rate and, where available, any index or reference rate

applicable to the initial borrowing rate, as well as the periods, conditions and procedure

for changing the borrowing rate; if different borrowing rates apply in different

circumstances, the abovementioned information on all the applicable rates”. While the

European legislator may have thus already addressed the problem of teaser rates with its

rules concerning the APRC, the rules governing the variability of interest rates however,

do not oblige Member States to regulate “any index or reference rate, and … the periods,

conditions and procedure for changing the borrowing rate” but limits itself to its mere

“disclosure” “where available”. Recital (32) clarifies that:

“this is without prejudice to provisions of national law not related to consumer information

which lay down conditions for, or prescribe the consequences of, changes, other than changes

concerning payments, in borrowing rates and other economic conditions governing the credit,

for instance rules providing that the creditor may change the borrowing rate only where there

is a valid reason for such change or that the consumer may terminate the contract should

there be a change in the borrowing rate or in some other economic condition concerning the

credit.”

These rules can therefore not be called IRR. Furthermore, the national rules on the

adjustment and adaptation of variable interest rates during the lifetime of a contract are

generally not seen as part of IRR and are not made to render rate ceilings more effective,

notwithstanding the limiting effect they will certainly have. These national rules follow

the philosophy that a consumer who has agreed to a certain initial interest rate should

not be subject to unilateral arbitrary increases of this rate. The legal principles which

govern such rules are laid down in the civil codes. For example, Section 315 of the

German Civil Code states that “where performance is to be specified by one of the parties

to the contract, then in case of doubt it is to be assumed that the specification is to be

made at the reasonably exercised discretion of the party making it.” In paragraph 3 of

this section, the final decision whether this discretion has been “equitable” is given to the

courts. German courts still use this rule to require that variable rate credit refer to an

objective reference rate, be adapted in equal terms and under equal conditions.

Since this area does not concern IRR but just a harmonisation of disclosure rules, in the

same way the rules govern the construction of the APRC, we will only give Member State

examples below:

In Austria Section 6 par. 1 No 5 KSchG stipulates that the decisive circumstances for the

variation of the interest must be clarified in the contract. They have to be justified and

not dependant on the decision of the creditor. There are two kinds of clauses which are

often used in contracts and are the subject of many court decisions. Zinsgleitklausel: This

is a contractual arrangement that links the interest rate to a specified variable reference

parameter, so that the interest rate varies automatically if there is a variation of the

reference parameter. Information to the customer about the variation in the interest rate

is only declaratory. The court states that falling interest must be realised within the same

time and in the same amount as the increasing interest. Zinsanpassungsklausel

(Zinsänderungsklausel): Gives creditors freedom to design interest rates. They can adapt

the interest rate if the terms of refinancing change on the capital market. Such clauses

are only effective if they are specific enough to enable the consumer to judge ex ante

within which borders the variation of the interest rate is due. To act arbitrarily at the

expense of the debtor must be impossible (otherwise the clause will not be incorporated

into the contract) OGH 4 Ob 73/03v.

For Belgium (already mentioned on the previous page) mortgage loan agreements must

have a predetermined objective reference rate. In addition they must stipulate that

variations in the interest rate (in minus or in more) are limited to a pre-determined fixed

difference in relation to the original interest rate (art. 9, §1, 7° WHK).

102 iff/ZEW – Final Report on interest rate restrictions in the EU

In Bulgaria Article 5 (1) 10 Consumer credit act requires similar information.

In Estonia, the Law of Obligations overdrafts stipulates that, before entering into an

overdraft agreement, the credit institution must notify the consumer of the interest

applicable at the time of notification and the conditions for changing the interest rate.

Finland forbids changing fixed interest rates in its Consumer Credit Act, but if such rates

are convened the reference rate must be objective.

In Greece Act 2501/1992 of the Governor of the Bank of Greece obliges the provider to

inform the client of the reference interest rate.

In Hungary, recent amendments to Section 210 of Act No. CXII of 1996 on Credit

institutions and Financial Enterprises provide that, in loan contracts with consumers and

in financial leasing agreements, interest, fee or costs elements alone may be increased

unilaterally to the disadvantage of the customer, but other elements of the contract may

not (including the reasons for change). An increase is only possible if an objective reason

has arisen as specified in the contract and reflected in the bank’s “pricing policy”.85

Furthermore, interest rates can be modified when changes in the usury ceilings call for a

change in the interest rate to bring it below the ceiling. For example, in the

Netherlands, if the maximum interest rate is changed, the interest rates must be

adapted unless the loan is due in the next year, in which case the interest rate remains

unchanged.

Ireland only requires according to Consumer Credit Act. Under s. 149, that customers

must be notified of increases in charges.

In Italy the courts use the unfair contract terms rules to limit arbitrary increases of

interest rates.

Latvia’s Consumer Credit Agreements No.692 (2008) Act for overdraft, credit card

credits, mortgage loans, deferred payment in sales contracts provides one statutory

provision – agreements should provide interest rate, interest rate change periods and its

rules.

In Poland, though the Banking Act contains provisions which require customers to be

informed of the mechanism of variations in interest rate, disclosure of some details is not

required (eg. central bank interest rates, currency exchange rates, stock exchange

indexes). However, the proposal by the Polish government for the implementation of the

CCD imposes more detailed information requirements on credit providers.

In Portugal cf. Circular nr. 1/2008/DSB of Bank of Portugal requires that adaptation is

done at a regular period. The need for information and termination by credit types is

subject of Decree-Law nr. 133/2009 and Article 14 of Decree-Law nr. 133/2009.

In Romania, the Emergency Ordinance no. 174 of 19 November 2008 for the

modification and completion of consumer protection laws and regulations and the 2010

Romanian Bill on credit agreements for consumers, require a written consent of the

consumer that the rate may be changed, notification of each modification 30 days in

85 This document, which is not public, though the objective reasons applied by the bank regarding unilateral

changes themselves are, is filed with and supervised by the Supervisory Authority for Financial Institutions

(PSZÁF). The Pricing Policy will be supervised in the light of an official Code of Conduct, adopted on 16

September 2010 and entered into effect on 1 January 2010. The wording of the Code of Conduct is available

in English at: http://www.pszaf.hu/data/cms2043240/codeofconduct.pdf.

iff/ZEW – Final Report on interest rate restrictions in the EU 103

advance and give a consumer 15 days after reception of such notification to accept or

reject these changes.

According to Art. 4 sec. 2 letter h) of The Slovakian Act on Consumer Credit, the

consumer credit contract must contain conditions for change of variable annual interest

rate, as well as index or reference rate applicable to the original variable annual interest

rate. According to the Annex to the Decree of the Ministry of Finance of the Slovak

Republic Nr. 620/2007 Coll., establishing a template for terms and conditions of credit

agreements, in the Formulary about contractual conditions of consumer credit must be

filled some specific information.

In the UK, the new s78A Consumer Credit Act 1974 imposes a duty on creditors to give

information to debtors on changes in interest rates, implementing article 11. Other than

regulations on the unilateral power to vary the contract or unfair terms, the Financial

Services Authority principles of “treating customers fairly” would also apply. There are

also rules that have been recently agreed between government and industry applicable

to changing interest rates on credit and store cards. Other regulation is based on

principles derived from a provider’s duty to execute bona fide contracts. An example

from the UK, where there are regulations on interest rate changes, shows that under

contract law a unilateral power in contract to alter the interest rate is subject to an

implied term that the lender will exercise his discretion “honestly, and not for an

improper purpose, capriciously or arbitrarily”86 and would not act in a manner which no

reasonable lender with a knowledge of the facts would have done.

In some countries, there are further stipulations in relation to non-compliance by

providers with the rules on reference indices or frequency and method of amendment of

interest rates. In Germany, for example, the courts replace the defective term with a

term which complies with the rules, so that the variation is made on the following terms:

the reference index is either the average interest applied to the contract (index-source:

German Central Bank) or the three-months Euribor; the interest rate is adjusted

quarterly (new jurisprudence: monthly); the interest rate is adjusted if the spread of the

index compared to the previous quarter’s index exceeds 0.25 percent (new

jurisprudence: no threshold).

In Finland for example, if the bank’s financial situation is seriously prejudiced, the bank

may, instead of allowing the loan to fall due, raise the interest rates. However, before

this is actually done, the bank must first consult the Finnish Financial Supervision

Authority.

In the present crisis, the issue of interest rates and their variation has been of great

relevance in Spain, where Court decisions are pending with regard to a number of claims

led by consumer associations (such as the Spanish Association of Consumers ADICAE)

involving over 20 financial institutions. These providers issued mortgages with a so-called

“clausula suelo” or “ground clause”, with the effect that interest rates in variable rate

mortgages would not be reduced below certain limits. This concern was also raised

during the research team’s meeting with the FIN-USE, at which a Spanish member

argued that the study of interest rate restrictions should also be concerned with reporting

on regulations controlling the floors as well as the caps. The consumer detriment

resulting from such practices, which have only been found in Spain, are based on the

granting of loans whose essential features on the limits to the adjustment of the interest

rates were not made clear to the borrower.

“The indiscriminate introduction in a surreptitious way by the Spanish banking industry of the

abusive so-called 'Floor and Cap clauses' on thousands (nine out of ten) of mortgages fees

signed from 2008 up to the present moment, preventing consumers and SME to take

86 Paragon Finance Plc v. Nash 2001 Court of Appeal at p 32; Mindy-Chen-Wishart (2008), p 110.

104 iff/ZEW – Final Report on interest rate restrictions in the EU

advantage of the sharp drop in Euribor. Thus, thousands of consumers are suffering the

paradoxical situation that while Euribor is 1.24 % at the current moment, they are facing an

interest rate of the 6.38 %. As a result and depending on the amount of the loan, they are

paying abusive interests (from € 2,500 to 6,000 a year) to banks”.87

The subject of variable rate credit is also under discussion in the context of specific

problems in certain Member Sates. In Austria for example, where the combination of

fixed and variable interest rates in long-term mortgage credit is a popular product (eg. a

20 year loan at a fixed interest rate of 4% for the first six years and 14 years at a

variable interest rate), an ongoing political discussion is taking place. It is based on the

fact that the constellation is seen by some as quite problematic for consumers because of

the potential for significant differences between fixed and variable interest rates, which

are not predictable.

1.3.4 Fees and charges

As already mentioned above88 there are other cost elements like closing and disclosure

fees, broker fees, payment protection insurance (PPI) fees, finance charge on insurance

fees, early repayment charges and commission on linked products that are related to the

cost of credit. The Member State reports from our legal experts indicate a quite liberal

regime regarding restrictions on fees in the European Union. Regulatory gaps seem to

exist especially with regard to PPI, as described in subsection 1.3.4.2 below. Table 20

shows the situation in the EU 27 Member States covering both general restrictions and

payment protection insurance.

Table 20: Restrictions on general fees and charges including PPI in the EU

Member

State

General fees and charges Payment Protection Insurance

Belgium No fees for credit intermediaries for

consumers under article 65, §1 WCK. But

only if credit entered into with his help.

50% of commission must be spread. Only

administrative set-up costs, fees and

costs for valuing the house may be

charged to the borrower.

No obligation to use the sum borrowed

to finance other financial instruments

(article 31, §2 WCK; art. 18 WHK).

Mortgage loans: if the insured dies, the

insured principal must be used to repay

the outstanding balance (art. 6, §3

WHK).

Cyprus No general restrictions. Penalty fees to

overrunning on the current account. A

3% limit for penalty fees for early

repayment mortgage loans but case by

case basis.

Czech

Republic

Contractual penalty if unreasonably high,

the court may reduce it.

Denmark Agreements concerning commission

must be spelled out in advance. The

same goes for the costs and nature of

any linked product.

87 See: FIN-USE (2010), p 4.

88 See Table 18: Borrowing rate and APRC at p 94.

iff/ZEW – Final Report on interest rate restrictions in the EU 105

Member

State

General fees and charges Payment Protection Insurance

Estonia If a contractual penalty is unreasonably

high, the court may reduce it to a

reasonable amount with regard to the

economic situation of the parties.

Germany Fees can be found ineffective if they

unreasonably disadvantage the other

party. Excessive fees for exceeding an

overdraft can be void. Foreclosure fees in

credit contracts without prejudice of the

amount are void.

Payment protection insurance is

intended to lower the default risk for

both lender and borrower and therefore

should influence the usury rate at equal

terms.

Ireland There is a specific prohibition on the

linking of services including payment

protection insurances in the case of

housing loans.

Italy Regulation of insurance brokers

approved obliges insurance companies

and brokers clearly to express the

amount of commission or costs paid in

collective credit insurance agreements.

Malta No specific restrictions on fees, but the

Malta Financial Services Authority (MFSA)

regularly reviews the charges list of the

banks and issues instructions for changes

if the charges are considered too high.

Poland Direct floating ceiling on fees and

additional charges (not included in the

APR) may not exceed 5 percent of the

amount of the loan.

Only Member State where fees and

additional charges (not included in APR)

may not exceed 5% of the amount of

the loan. Insurance is thus regulated.

Portugal Linked insurance premium must

decrease with the residual debt. It is

forbidden to make the credit conditional

upon the purchase of PPI or any other

financial product.

Romania Only the following allowed for a loan: a

credit check fee, a credit management

fee or a checking account management

charge, compensation in the event of

early repayment, insurance-related

costs, penalties, and a single charge for

the services provided at the consumers’

request.

Slovakia The borrower is not liable for payment of

any charges about which the requisite

information is not properly provided (art.

4 sec. 4 consumer credit act).

106 iff/ZEW – Final Report on interest rate restrictions in the EU

Member

State

General fees and charges Payment Protection Insurance

Slovenia The borrower is not liable for payment of

any charges about which the requisite

information is not properly provided (art

6 consumer credit act).

Sweden The borrower is only liable to pay special

compensation for the credit (charges), in

addition to or instead of interest, if such

charges relate to costs which the lender

has incurred for the loan and if the

charges are specifically identified in the

agreement.

UK No PPI sold at the point of sale is

considered. Debtor can only be

contacted for sale of insurance after 7

days. Sale of single premium insurance

also prohibited by the Financial

regulators. Fines on the grounds that a

firm must pay due regard to the

interests of its customers and treat

them fairly.

No

Restricitions

reported

Austria, Bulgaria, Finland, France, Greece, Hungary, Latvia, Lithuania, Luxembourg,

Netherlands, Spain

Source: Stakeholder and Expert survey.

1.3.4.1 General restrictions

Apart from methods of calculation of the APRC, nine Member States (Austria, Bulgaria,

Finland, Greece, Hungary, Latvia, Lithuania, Luxembourg and Romania) report no

restrictions on fees.

As summarised in Table 20, of all Member States, only Poland reports a general direct

floating ceiling on fees to be specified stated in the law. Under art 7a Consumer Credit

Act, fees and additional charges (not included in the APR) related to the conclusion of the

credit contract may not exceed 5 percent of the amount of the loan. The financial

regulator, the Ministry of Finance, reports that, due to the statutory interest rate ceiling,

the level of other charges, fees, etc. resulting from the contract has risen significantly in

Poland. As a result, the overall cost of credit is less clear to consumers and that it might

still exceed the optimum level of interest rates. Although the law on consumer credit sets

a limit for the overall level of fees resulting from the conclusion of a contract, the

definition did not seem broad enough to prevent providers from moving the costs of

credit to unrestricted areas.

In Belgium there are restrictions on fees for credit intermediaries and non-finance

charges. Under article 65, §1 WCK and article 13 WHK, no fee is payable to the credit

intermediary by the consumer (direct or indirect); the credit intermediary is only entitled

to a commission for credit agreements validly entered into with his help. At least 50% of

the commission must be spread depending on the credit type and duration (article 65, §§

3-4 WCK). The following restrictions on non-finance charges are stipulated in article 11

WHK: except for the legal charges on the mortgage and charges due pursuant to other

iff/ZEW – Final Report on interest rate restrictions in the EU 107

legal provisions, only administrative set-up costs, fees and costs for valuing the house

may be charged to the borrower; fees and costs for valuing the house are only due after

valuation of the house. If they are charged, the creditor must provide the borrower with

a copy of these costs in advance; administrative set-up costs are only due after an offer

has been sent to the borrower; no agent or another intermediary may burden directly or

indirectly the credit applicant with costs (art. 48 WHK).

In some Member States the creditor must inform of all fees in the credit agreement and

if not face legal consequences.

For example in Germany, the civil code provides that the consumer credit agreement

must show fees and other expenses payable by the debtor and related to the receipt of

the loan (German Civil Code, section 492). At the time of execution of the agreement the

consumer must also be informed in writing of the fees imposed from the moment of

execution of the agreement, as well as the conditions under which they may be charged.

The borrower is not liable for payment of any charges about which the requisite

information is not properly provided (section 494 German Civil Code). Similar rules are

applied in Slovakia (art. 4 sec. 4 consumer credit act) and in Slovenia (art 6 consumer

credit act). Likewise, the Swedish Consumer Credit Act (1992:830) art 12. Section 12

states: “The borrower is only liable to pay special compensation for the credit (charges),

in addition to or instead of interest, if such charges relate to costs which the lender has

incurred for the loan and if the charges are specifically identified in the agreement.

Few Member States report that certain fees might be reduced or even declared void

either by the courts or by a financial service authority.

In Estonia, the Law of Obligations Act provides for the possibility of applying a reduction

in contractual penalties. If a contractual penalty is unreasonably high, the court may

reduce it to a reasonable amount at the request of the payer, taking into particular

account the extent to which the obligation has been performed by the party, the

legitimate interests of the other party and the economic situation of the parties. The

same method is effective in the Czech Republic. Also in Germany there are restrictions

under jurisprudence (case law), limiting “freedom of the contract” in relation to the type

of fees that might be charged by banks. Section 307 German Civil Code declares a

standard term and condition to be ineffective if it unreasonably disadvantages the other

party. The rule is applied not only to contractual terms and conditions in credit contracts

but also to some fees. Recently the courts have declared fees for exceeding an overdraft

to be void. Furthermore, an ongoing debate was reported as to whether foreclosure fees

in credit contracts without prejudice of the amount are void. In Malta there are no

specific restrictions on fees, but the Malta Financial Services Authority (MFSA) regularly

reviews the charges list of the banks and issues instructions for changes if the charges

are considered too high.

1.3.4.2 Payment protection insurance

One important method of circumventing interest rate restrictions is through “voluntary”

credit insurance or other charges not included within the concept of interest. Experts

report that in Poland, France (over 70% of revolving credit), Germany (over 50% of

instalment loans) and the United Kingdom this kind of ancillary service is widespread.

With regard to Payment Protection Insurance (PPI, Residual Debt Insurance, Outstanding

Debt Insurance) structural features are reported for the UK and the German market

which adversely affect competition. Those features exert little competitive pressure on

the distributor at the key point of sale, the complex nature of the contract, making

comparison difficult, the lack of product information prior to the point of sale, very low

levels of cancellation or switching by consumers, problems faced by stand-alone

108 iff/ZEW – Final Report on interest rate restrictions in the EU

providers in reaching consumers and “vertical integration” (one company offering both

the credit contracts and insurance contracts).89 Furthermore, the conduct of firms has an

adverse effect on competition: the OFT reports that consumers either assume or are told

that buying insurance would positively effect the decision to grant a loan. Poor upfront

information is regarded as having a negative impact on competition, as well as the fact

that the competition is centred on the loan and not on the insurance.90 As a result of the

lack of competition, there is little pressure on the price of Payment Protection Insurance.

This is why those prices differ greatly, even when products with almost identical features

are considered.91 John Fingleton, OFT Chief Executive summarised the findings of the

market investigation as follows:

'Our examination of the evidence presented to date gives us reasonable grounds to

suspect that there are features of this market which restrict competition to the detriment

of consumers. Despite some evidence of a degree of consumer satisfaction with aspects

of the product, the evidence as a whole suggests consumers get a poor deal.'92

According to the Office of Fair Trading93 in the United Kingdom, there are about 20

million payment protection insurance policies currently in force, and about 6.5 to 7.5

million are sold annually. Both the Gross Written Premium (GWP) and the number of

policies indicate a rapidly growing sector between 2000 and 2005. The average yearly

rise of GWP in these years amounted to more then 18 percent. The OFT reports an

average claims ratio94 for all Payment Protection Insurance in the UK of approximately 20

percent and states that these figures are low compared with other forms of insurance.95

In Germany, debtors very often dispose of payment protection insurances on their

loans. According to the BaFin (the German Banking Supervision, Insurance Supervision

and Securities Supervision), at the end of 2008 there were 2.84 million payment

protection contracts, with a total sum insured of about 21 EUR billion96. Although there is

empirical evidence that the main triggers of default and over-indebtedness in Germany

are forced unemployment, forced short-time working, failed self-employment and

divorce97, payment protection insurances very often only cover the life-risk of the

borrower. For this reason, this insurance very rarely assist with the main triggers for

default in Germany, ie. unemployment and divorce. As in the UK, the Insurance Claims

Ratio is very low. Although there has been a rise in credit defaults from 2.3 percent in

2007 to 2.5 percent in 200898, the insurance claims ratio has fallen from 13.92 percent of

GWP to 12.18 percent, as is shown by the following table:

89 Office of Fair Trading (2007), pp 2 ff.; Reifner, U., Knobloch, M., Knops, K, (2010), pp 20 ff.

90 Office of Fair Trading (2007), pp 3 ff.

91 Office of Fair Trading (2007), pp 41 ff.; Reifner, U., Knobloch, M., Knops, K, (2010), pp 47 ff.

92 See OFT press release 15/07, 7th February 2007.

93 All Data from Office of Fair Trading (2007).

94 Claims paid as a percentage of the Gross Written Premium.

95 Comprehensive Motor Insurances: 82 percent of GWP, Medical Insurances: 80 percent of GWP, Pet

Insurances: 72 percent of GWP, Household Insurances: 54 percent of GWP.

96 Data from BaFin, Statistic on life insurances, tables 150(1) and 150(2). The BafIn Statistic contains only a

part of the market of PPI.

97 These triggers are reported in approximately 55 per Cent of all cases as the main trigger of overindebtedness

while debt advisors report only in one of 100 cases, that the death of the partner was the

main factor. See Knobloch, M., Reifner, U., Laatz, W. (2009), p. 23.

98 According to SCHUFA, the biggest German Credit Register (see SCHUFA Kredit-Kompass).

iff/ZEW – Final Report on interest rate restrictions in the EU 109

Table 21: Payment protection insurance in Germany, 2001–2008, EUR (size, claims, claims ratio)

year Insurance claims (EUR) GWP (EUR) Insurance claims ratio

according to GWP (%)

2001 49,000,000 € 288,000,000 € 17.01%

2002 52,000,000 € 408,000,000 € 12.75%

2003 53,000,000 € 391,000,000 € 13.55%

2004 54,000,000 € 292,000,000 € 18.49%

2005 56,000,000 € 283,000,000 € 19.79%

2006 47,000,000 € 307,000,000 € 15.31%

2007 43,000,000 € 309,000,000 € 13.92%

2008 38,000,000 € 312,000,000 € 12.18%

Source: SCHUFA Kredit-Kompass.

Given with the above findings and concerns in relation to the payment protection

insurance market and its lack of competition, there is relatively little regulation of the

prices and fees involved. Nine Member States reported restrictions or regulations on

payment protection insurance. Only one Member State, Poland, reported imposition of a

floating maximum on such fees. In the UK, the Competition Commission banned the sale

of Payment Protection Insurance at the point of sale of the loan outright, but the

Competition Appeal Tribunal revoked the ruling. All other reports (Belgium, Denmark,

Germany, Ireland, Italy and Spain) concerned the regulation of information, such as upfront

information as to the fees involved in related financial products.

In Belgium, neither the creditor nor the credit intermediary may oblige the consumer to

use the sum borrowed to finance other financial instruments (article 31, §2 WCK; art. 18

WHK). If the loan is offered to the consumer with a linked payment protection insurance

policy (“outstanding balance insurance”), with linked income insurance or with linked

supplementary invalidity insurance, the charges relating to these forms of insurance

must be included in the total cost of credit. This article does not apply to loans in which

the amount of credit exceeds 5,000 EUR or if the insurance agreement was signed at the

consumer’s explicit request and after the loan was already signed (article 31, §4 WCK).

Finally the WHK on mortgage loans sets a restriction on the financing of linked

outstanding balance insurance policies. If the insured dies, the insured principal must be

used to repay the outstanding balance (art. 6, §3 WHK). The creditor may only stipulate

that the linked policy be applied for its benefit if the proceeds of sale of the security are

insufficient to repay the mortgage loan (art. 26, §3 WHK).

In Denmark, agreements concerning commission must be spelled out in advance. The

same goes for the costs and nature of any linked product.

In Germany, according to section 492 (6) German Civil Code, the costs of any residual

debt insurance or other insurance taken out in connection with the consumer loan

contract must be set out in the contract. Furthermore, under section 6 of the Statutory

Order on Price Quotation (Verordnung zur Regelung der Preisangaben), “all other costs”,

including payment protection insurance fees, must be included in the APR, but only

where the insurance is a pre-condition of the loan. Because of this exemption, there are

very few contracts where the insurance fees are included in the APR. The Statutory Order

on Price Quotation has been amended recently (effective from June 11, 2010) and now

states that fees for payment protection insurance must be included not only when they

110 iff/ZEW – Final Report on interest rate restrictions in the EU

are pre-condition of the loan but also if they are pre-condition for the intended conditions

of the contract. Payment protection insurance is intended to lower the default risk for

both lender and borrower and therefore should influence the interest rate where the

provider applies “risk-based-pricing”. In spite of this, there is so far no evidence that

there has been a change in providers’ practice of not including the fee in the APR.

Credit intermediaries must be licensed in Ireland (section 144 Consumer Credit Act

1995) and must, in advance of an agreement, disclose in writing to the consumer the

existence of any commission arrangement (section 148). Furthermore, according to

section 127 Consumer Credit Act 1995 there is a specific prohibition on the linking of

services (including payment protection insurances) in the case of housing loans. Where,

in connection with the making or arranging of a housing loan, more than one service is

made available by a mortgage agent or one or more of his subsidiaries, the agent shall

not, and shall ensure that each of his subsidiaries does not, make the services available

on terms other than terms which distinguish the consideration payable for each service

so made available; nor shall any of the subsidiaries make the services available on terms

other than terms which make that distinction.

In Italy, in 2009 a number of actions regarding payment protection insurance policies

have been approved. In particular, the regulation of insurance brokers approved by

ISVAP (Reg.5/2006 as modified by provisions approved in 2009) obliges insurance

companies and brokers clearly to express the amount of commission or costs paid in

collective credit insurance agreements. This is valid only for insurance sold under the

form of collective agreements, but the prospect of extension of the rule is being

discussed.

As mentioned before, Poland is the only Member State that reports that fees and

additional charges (not included in APR) related to conclusion of the credit contract may

not exceed 5% of the amount of the loan. In the cost of credit the cost of insurance and

the cost of establishing collateral are not included.

In what concerns mortgage credit, Portuguese Decree-Law nr. 222/2009 establishes

that the linked insurance premium must be proportionate to the amount of the debt,

which means that the premium must be reduced alongside amortisation of the loan. For

the types of credit subject to Decree-Law no. 133/200999 it is forbidden to make the

conclusion of a credit agreement conditional upon the purchase of any other financial

products. The same rule is established for mortgage loans under Decree-Law no.

51/2007.

Though Romanian regulations introduced a prohibition against new fees and charges

associated with loans in 2008, the Romanian Bill on credit agreements for consumers

2010 states that the creditor may charge only the following amounts for a loan: a credit

check fee, a credit management fee or a checking account management charge,

compensation in the event of early repayment, insurance-related costs, penalties, and a

single charge for the services provided at the consumers’ request.

The UK Competition Commission enacted the Payment Protection Insurance Order 2009

under which the sale of payment protection insurance at the point of sale was banned.

The lender may contact the debtor after 7 days to sell insurance. The sale of single

premium insurance was also prohibited. Barclays Bank appealed this Order to the

Competition Appeal Tribunal and the Point of Sale prohibition was quashed and returned

to the Commission for reconsideration.100 The Financial Services Authority has taken 22

99 Instalment loans, Financial leasing, Hire purchase financing, Point-of-sale financing and all categories of

Revolving Credit except for overdrafts were the credit has to be paid back within one month.

100 See Competition Appeal Tribunal, judgement Case number 1109/6/8/09 from 16th October, 2009,

http://www.catribunal.org.uk/files/Judg_1109_Barclays_16.10.09.pdf.

iff/ZEW – Final Report on interest rate restrictions in the EU 111

enforcement actions against regulated firms for the mis-selling of payment protection

insurance since 2005 and sent a Dear CEO Letter in February 2009 requesting firms not

to sell single premium insurance with unsecured personal loans. The Enforcement Actions

in relation to PPI usually involved contraventions of regulated firms’ fundamental

obligations under the Principles for Business to “take reasonable care to organise and

control its affairs responsibly and effectively, with adequate risk management systems”

and “ a firm must pay due regard to the interests of its customers and treat them

fairly”.101

1.4 Effects of the CCD 2008 on IRR

1.4.1 EU Directives and national IRR

IRR is concerned with the setting of prices. Instead contract law and especially EU

consumer law relates the setting of prices to the parties of the contract while monitoring

the procedures and transparency of the process.

This is why the Consumer Credit Directive102 (CCD 2008) is, as far as prices are

concerned, exclusively focussed on price disclosure. It even explicitly refrains from

harmonising price regulation and in Recital 30 leaves IRR regulation in particular to the

national legislator. While usury had been mentioned in an earlier draft, the regulation

now in force expressly alludes to this restricted purpose more indirectly. For example,

Recital 22 states that for special IRR “prohibiting the creditor from requiring the

consumer, in connection with the credit agreement, to open a bank account or conclude

an agreement in respect of another ancillary service, or to pay the expenses or fees for

such bank accounts or other ancillary services”, the “Member States should remain free

to maintain or introduce national provisions.”

Directive 2005/29/EC on unfair commercial practices defines its area of application in Art.

2 (d) as ‘business-to-consumer commercial practices’ (hereinafter also referred to as

commercial practices)” which “means any act, omission, course of conduct or

representation, commercial communication including advertising and marketing, by a

trader, directly connected with the promotion, sale or supply of a product to consumers”.

While the Directive contains many rules on price disclosure as well as the calculation of

prices or the use of prices in advertisements, the setting of the price itself does not fall

under the notion of “unfair commercial practices” as defined in this Directive as well as in

national law.

The same is true for Directive 93/13/EEC on standard contract terms, “whereas, for the

purposes of this Directive, assessment of unfair character shall not be made of terms

which describe the main subject matter of the contract nor the quality/price ratio of the

goods or services supplied”.

Although the Directives are quite clear in this respect, at EU level experts suggested that

standard contract law may for example be used to qualify default interest clauses as

forbidden penalty clauses or extortionate pricing as “unfair”. In this respect we have

argued that in future IRR may be dealt with as a form of cartel law.103

However, besides these general questions about the appropriate place for IRR in the

body of national and EU law, there is a more or less effective relationship in practice

101 http://fsahandbook.info/FSA/html/handbook/PRIN/2/1.

102 Directive 2008/48/EC of the European Parliament and of the Council of 23 April 2008 on credit agreements

for consumers and repealing Council Directive 87/102/EEC.

103 See 1.1.3.5 Fairness and Good Morals – towards a general principle for IRR in European contract law? at pp

54.

112 iff/ZEW – Final Report on interest rate restrictions in the EU

between IRR and consumer credit legislation in that national legislators tend to provide

harmonised rules for credit products which use the same definitions for IRR as they are

use in general consumer credit law.

It has been shown above104 that the core element of IRR, the interest rate, is largely predefined

by the creation of a special interest rate for price disclosure (APRC) under EU

law. While older IRR rules still refer to the rate of borrowing, the modernisation of

consumer credit law has directly affected this element.

CCD 2008 does not prevent national legislators from using both its definitions and

denominations and its rules on scope and circumvention. Furthermore, questions of how

the restricted interest rate should be calculated both mathematically and legally, whether

small loans of up to €200 (Art. 2 (2) c) or for less than three months (Art. 2 (3)) are

included and whether only consumers (Art. 1) or all borrowers are covered, whether

mortgages (Art. 2 (2) a) or certain micro-loans are excluded (Art. 2 (2) l) must

increasingly be regulated in conformity with general consumer credit law. Since national

credit disclosure law is ruled by the maximum harmonisation approach of Art. 22 (1) of

the CCD 2008, this means that large parts of national IRR are also indirectly regulated by

EU law.

1.4.2 Implementation of CCD 2008

As at March 30, 2010 only a few Member States had incorporated the CCD 2008 into

national law. In most Member States, a draft was in the legislative process. In Ireland

and Poland, details of transposition were still being debated while for some Member

States no proposal or details were available.

Table 22: Transposition of the CCD by March 2010

Implemented

(effective)

Bulgaria (12.5.10); Estonia (1.5.2009), Germany (11.6.10),

Hungary (14.12.09), Portugal (1.7./1.10.09)

Draft, Proposal Austria, Belgium, Czech Republic, Denmark, France, Greece,

Latvia, Lithuania, Netherlands, Slovakia, Slovenia, UK

Under Discussion Ireland, Poland

No details available Cyprus, Finland, Italy, Luxembourg, Malta, Spain, Romania,

Sweden

Source: Expert Survey.

It was reported that only in Portugal had transposition had a direct effect on IRR and

usury ceilings were enacted for the first time along with general implementation of the

Directive. Other direct effects were reported in Poland, where IRR for additional costs of

credit, currently limited to 5% of the loan value, were removed. In Belgium IRR on early

repayment fees for full repayment are now regulated and the calculation of interest rates

was amended in Austria and Denmark.

In the opinion of many of the experts, however, the CCD and the passing of the Directive

on Unfair Commercial Practices 2005/29/EC (UCPD) had at least side effects on the

regulation of specific credit types, on calculation methods for interest rates, and on closer

regulation of early repayment fees. When asked about which credit types should be

104 See p 91.

iff/ZEW – Final Report on interest rate restrictions in the EU 113

regulated within the scope of the CCD 2008, credit card loans and revolving loans were

mentioned.

In some Member States, implementation of the CCD 2008 had an effect on additional

types of credit which will be newly regulated as a result of its implementation. New

regulation affecting IRR was reported in relation to financial leasing (Austria), a €150

limit on micro-credit (Sweden), effects of overrunning a bank account (Austria, Belgium,

Czech Republic, Slovakia), of overdraft credit (Czech Republic, Slovakia), linked credit

agreements (Czech Republic, Slovakia), financial leasing (Austria, Slovakia), mortgage

loans (Lithuania, Romania), credit intermediaries (UK), or new credit classifications from

€21,500 to €75,000 (France).

1.4.3 Small amounts of credit

The CCD excludes credit of less than €200 (Article 2 (2)c) from its scope but allows such

loans to be incorporated into national legislation. (Recital 10).

Only four countries will, after the transposition of the CCD, continue to exclude

such small credit in general from disclosure law as well as from IRR.

Seven countries have, on the other hand, included such small loans into disclosure

law as well as into general IRR, some only recently using the transposition of the

CCD.

Nine countries have a mixed regime.

The Netherlands recently included small loans into price disclosure law, expressly

because it wanted to regulate IRR. This shows that the transposition of EU regulation on

consumer information may also be influenced by IRR legislation at national level.

The following table gives examples of how the exemptions from disclosure law are

applied or not applied in IRR. Since the question was asked in relation to all kinds of IRR,

including the general principle of IRR, the table is especially interesting where the answer

is homogeneous.

114 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 23: Small credit and the CCD

Country Included

Disclosure/IRR

Remark

Austria Yes/Yes

Belgium No/Yes €200 partially excluded from WCK (art. 3, §2 WCK).

IRR apply (default charges, APRC, sanctions).

Bulgaria No/? €200 excluded under Art. 4, (1) 1 Consumer Credit

Act.

Cyprus No/? Excluded €200.

Czech Rep No/No Less than CZK 5000 (approx € 200).

Denmark No/No Excluded for less than €200.

Estonia Yes/Yes Law of Obligations Act § 403 changed in 1.05.2009.

France No/Yes Regulated if for a term of more than 3 months but all

are covered by IRR.

Germany No/Yes €200 excluded but IRR is not affected.

Hungary Yes/Yes

Lithuania Yes/Yes LTL 1000 (approx €290). The proposed Consumer

Credit Law will regulate all loans.

Luxembourg No/Yes Under €185.92 excluded.

Malta No/Yes €202.66 excluded. These rules do not affect IRR.

Netherlands Yes/Yes Exemption only until 11 June 2010 if their term was

longer than 3 months (Wet Financieel Toezicht).

Portugal No/No* €200 excluded, usury ceilings (Decree-Law no.

133/2009) not applicable. But non-bank credit (art.

1146 of the Civil Code).

Romania Yes/Yes Law no. 289 of 24 June 2004 is not applicable to

loans for a total of less than €200 but this will

change in 2010.

Slovakia Yes/Yes Not excluded under civil law (Civil Code - Act No.

40/1964 Coll.), consumer law (art. 52-54) and

contract law (art. 657-658).

Slovenia No Excluded below €170 but after transposition of CCD

at €200.

Spain No/Yes Under €150 excluded.

Sweden Yes/Yes Consumer Credit Act 1.1.2011, micro-credit, shortterm

loans will then be included.

Source: Expert Survey.

1.4.4 Short-term loans

The Directive also addresses short-term loans for less than one month (Art. 2 (2) i) or

three months (Art. 2 (3)) if certain other conditions are also met. This has led to a

number of different exemptions for both credit disclosure law and IRR which, in some

countries, resemble the regulatory provision in Latvia, Estonia, Netherlands, Sweden,

Bulgaria. Czech Republic, Lithuania, Luxembourg for small loans.

In other countries, there are quite sophisticated regulations which take into account the

role of short-term credit in relation to overindebtedness.

In Great-Britain, the provisions of the Consumer Credit Act 1974 apply in particular to

Payday Loans and Home Credit Agreements (both debtor-creditor agreements). Home

iff/ZEW – Final Report on interest rate restrictions in the EU 115

Credit Agreements (weekly collections) are also subject to the Home Credit Market

Investigations Order 2007, which requires home credit lenders (a) to share customer

repayment data with other lenders (b) to publish on a website financed by the main

lenders cost information on their loans and (c) to draw the attention of borrowers to this

website in account statements provided to borrowers.

In Portugal, short-term loans for a term of less than 3 months granted by credit

institutions are included within the scope of Decree-Law no. 133/2009 (which transposes

the CCD 2008) and, for that reason, the usury ceilings set in that legislation are also

applicable to these loans. Special bank accounts exists which are normally conceived as

an overdraft facility on the deposit account into which the borrower’s salary is paid. In

2009, the consumer association magazine, Proteste, (from DECO) published a

comparative study of those accounts, concluding that the APR on them ranged from

12.46%, and 24.16%. Normally, those accounts do not charge maintenance costs and

there are a few other “benefits”, depending on the bank.

In Belgium, consumer credit law distinguishes between revolving credit accounts and

other short-term consumer credit agreements. Revolving credit accounts, which are

repayable within 3 months and the amount of which does not exceed €1,250, are

excluded from the scope of the Belgian Consumer Credit law (art. 3, §1, 4° WCK). Both

conditions (ie. repayable within 3 months and maximum amount of €1,250) are

cumulative. Other credit agreements are excluded from WCK if the loan is repayable

within 3 months (art. 3, §1, 3° WCK). If excluded from the WCK, general civil law

provisions will nevertheless be applicable.

In Denmark, these short-term loans are not covered by the Danish Consumer Credit

Agreement Act. However guidelines have been set up by the Danish consumer

ombudsman and the industry in relation to distance sales of short-term or small loans

(“sms- and web-loans”).

In Finland, if the duration of the loan is less than 3 months and no interest is charged,

the standard form for pre-contractual information and a written form of contract need not

be used (Consumer Protection Act chapter 7 paragraph 3).

In Slovenia, short-term loans are not regulated by the Consumer Credit Act, unless they

exceed 300% of gross personal income for the previous month (or, under the draft new

Consumer Credit Act, unless significant charges are payable). All interest arising from

obligations are regulated by the Code of Obligations (Obligacijski zakonik, OZ-UPB2).

1.4.5 Responsible lending

Also the principle of responsible lending in the 2002 draft of the CCD has not been

overtaken as such into the 2008 final version where Art. 8 refers only to the assessment

of the creditworthiness of the consumer. But recital 26 underlines that “responsible

lending” is a general principle underlying the whole Directive when it says:

“Member States should take appropriate measures to promote responsible practices

during all phases of the credit relationship, taking into account the specific features of

their credit market. Those measures may include, for instance, the provision of

information to, and the education of, consumers, including warnings about the risks

attaching to default on payment and to over-indebtedness. In the expanding credit

market, in particular, it is important that creditors should not engage in irresponsible

lending or give out credit without prior assessment of creditworthiness, and the Member

States should carry out the necessary supervision to avoid such behaviour and should

determine the necessary means to sanction creditors in the event of their doing so.

Without prejudice to the credit risk provisions of Directive 2006/48/EC of the European

Parliament and of the Council of 14 June 2006 relating to the taking up and pursuit of the

116 iff/ZEW – Final Report on interest rate restrictions in the EU

business of credit institutions (1), creditors should bear the responsibility of checking

individually the creditworthiness of the consumer. To that end, they should be allowed to

use information provided by the consumer not only during the preparation of the credit

agreement in question, but also during a longstanding commercial relationship. The

Member States' authorities could also give appropriate instructions and guidelines to

creditors. Consumers should also act with prudence and respect their contractual

obligations.”

Although information and education are seen here as the primary tools, the principle of

responsible credit is seen by both providers105 and consumer organisations106 as much

broader and as including substantive behaviour in terms of the pricing and servicing of

credit contracts.

1.4.5.1 General principles of good morals and bona fide

General principles concerning the need for lenders to exercise good morals or fairness in

contract law, as well as concepts of force majeure and bona fide exist in sixteen Member

States107. In some cases, these have been used by courts in attempts to restrict overindebtedness.

For example, the Belgian courts have held that credit providers violate this

requirement to enter into bona fide contracts if they lend money to people who, at the

outset of the contract, cannot reasonably be expected to maintain the payments.

In other countries (for example, Estonia, Greece, Slovenia, Lithuania, Luxembourg, and

Germany), the concept of good morals appears to be very closely linked to the concept of

usury, and is used for protection against the exploitation of weakness or lack of

experience on the part of the borrower. For example, the Lithuanian Civil Code requires

that contracts be fair and reasonable and gives people a right to apply for the interest

rate on an agreement to be reduced on the basis that it does not conform to ordre

publique and principles of good morals.

In some countries, general principles appear to allow courts to intervene even where the

agreement is not found to be usurious. For example, in Slovenia, the Constitutional Court

ruled in case no. U-I-202/93 (6.10.1994) that, regardless of any indicators of usury, an

agreement under which interest accrued within a short space of time to match the level

of the principal fell foul of the bona fide principle.

However, questions remain as to the effectiveness of these general principles and the

experience across Member States varies. For example, in Italy, social force majeure is a

common topic of debate among legal academics but does not appear to be positively

reflected in court decisions and, in Hungary, it is also noted that the courts rarely apply

the principles directly.

1.4.5.2 Specific provisions

Specific provisions, for example requiring the lender to assess the creditworthiness of the

borrower and to provide an adequate explanation of the credit product, are now found

105 See WOCCU (2008); IFC (2008); Cetelem (2006); UK British Bankers’ Association (2005); BIS (2007);

INGO (2005).

106 Principles N° 3 of the European Coalition for Responsible Credit refers to IRR when it says: “P3 Lending has

at all times to be cautious, responsible and fair. 1. Credit and its servicing must be productive for the

borrower. 2. Responsible lending requires the provision of all necessary information and advice to

consumers and liability for missing and incorrect information. 3. No lender should be allowed to exploit the

weakness, need or naivety of borrowers. 4. Early repayment, without penalty, must be possible. 5. The

conditions under which consumers can refinance or reschedule their debt should be regulated.”

107 The countries are Austria, Belgium, Czech Republic, Denmark, Estonia, Germany, Greece, Hungary, Italy,

Lithuania, Luxembourg, Poland, Romania, Slovakia, Slovenia, and Spain.

iff/ZEW – Final Report on interest rate restrictions in the EU 117

(or are being introduced) in Member States as a result of the provisions of the CCD 2008.

Earlier Directives, for example in respect of Unfair Contract Terms 29/2005 are also

relevant across Member States. However, some countries have additional protections in

place to ensure responsibility in lending. Examples in this respect include:

The UK and Finland have responsible lending rules covering the consumer lending

process from marketing, through loan applications and decisions, to debt

recovery, requiring lenders to consider how they can help people with debt

problems.108 The Financial Services Authority in the UK also provides conduct of

business rules for mortgage lending which address responsible lending issues

including creditworthiness and arrears handling.

o Estonia and the Czech Republic have guidance on this issue, for example

the Czech Republic encourages creditors to investigate the debtor’s

capacity to repay the debt in the Ethical Codex of the Financial Market.

o In Ireland the Consumer Protection Code 2006 contains specific ‘Knowing

the Consumer and Suitability’ provisions that encourage responsibility in

lending.

The UK also provides courts with the power to re-open consumer credit

agreements where these are found to constitute an ‘unfair credit relationship’ (Ss

140A-D, Consumer Credit Act 2006), although there are only a small number of

decisions concerning the interpretation of these sections from the lower courts at

present.

Some countries have also put in place formal limits on loan to value and loan to

income ratios which have turned into classes for risk based pricing so that lower

ratios can lead to lower interest rates but there is a wide variation in practice:

o Austria and Germany had formerly 60% loan to value ratios for mortgage

credit from banks which issue related bonds. But these limits have no

effects on the contracts itself but are only used today for attributing

different interest rates.

o Poland places loan to value and loan to income limits on bank credit only.

o Hungary has issued a Responsible Lending Decree that imposes a debt to

income limit for consumer credits and there are also limits on loan to value

lending for mortgages and car purchase.

o The Netherlands provides guidelines in its Code of Conduct for creditors

concerning the amount of money that should be left following credit

repayments to meet essential household expenditure.

o Romania obliges banks to analyse the repayment capacity of credit

applicants and provides a limit on the maximum level of repayments

relative to income.

108 The licensing regime established by the UK Consumer Credit Act 2006 includes irresponsible lending as

factor in determining whether lender’s conduct is deceitful, oppressive or unfair (s25(2)(B). Lenders are

therefore expected to conform to principle of “fair treatment of borrowers” “Borrowers should not be

targeted with credit products that are clearly unsuitable for them, subjected to high pressure selling,

aggressive or inappropriate coercion, or conduct which is deceitful, oppressive, unfair or improper, whether

unlawful or not”.

118 iff/ZEW – Final Report on interest rate restrictions in the EU

o In Italy there are no formal restrictions but there is an ‘implied limit’ of

one-fifth of monthly income/

However, beyond the requirements to check the creditworthiness of borrowers and the

specific provisions set out above there are no other obligations placed on lenders to

prevent over-indebtedness. Statutory requirements for lenders to act in the best

interests of the borrower are usually regulated in public law and not in the form of a

contractual obligation to the consumer. But some examples of this can be found in

Belgium, where the law does place lenders under an obligation to provide the most

suitable credit product for the customer’s needs (art. 10, 11 and 15 WCK), and in

Finland109 and Ireland110. Suitability requirements are also mentioned in the Czech

Republic’s Ethical Codex for the financial market111 and in some countries has been

established through case-law as a result of the general principles requiring good morals

(for example, Denmark112). In the UK, suitability is also an issue in respect of ‘advised

sales’ of certain financial products including payment protection insurance.113

Finally, it should be noted that in addition to the implementation of the EU Directive on

Unfair Contract Terms, the concept of unfairness has a wider application in the legal

systems of some, but by no means all, Member States. The inclusion of ‘unfair credit

relationship’ in Section 25(2)(B) of the UK Consumer Credit Act 2006 is a case in point,

but other examples include:

In Estonia, ‘unfairness’ occurs if there is an unreasonable imbalance between the

amount of credit and the interest charged and in Ireland, sections 47 and 48 of

the Consumer Credit Act 1995 provide courts with the power to set aside

completely or partially re-open and re-write agreements if a charge is excessive.

In Germany, Hungary, Latvia, Lithuania and the Netherlands the concept of

unfairness appears as ‘good faith’ and ‘reasonableness’.

In contrast, unfairness is not a concept which has a legal definition for the purpose of

credit regulation in Austria (although the concept of unconscionability applies), Bulgaria,

109 Actual bill implementing the CCD 2008 (Directive 2008/48/EC).

110 “Knowing the Consumer and Suitability” provisions of the Consumer Protection Code 2006

http://www.financialregulator.ie/processes/consumer-protectioncode/

Documents/Consumer%20Protection%20Code.pdf.

111 See at http://www.clfa.cz/index.php?textID=94.

112 Executive Order of the Danish Financial Business Act for “good business practice for financial undertakings”.

The executive order contains requirements regarding advice, information on the product, assessment of the

borrower, the suitability of the product, the borrower’s willingness to take risks etc.

113 The vastest reference to responsible lending principles are made in the UK in the bill MCOB 11.3.

http://fsahandbook.info/FSA/html/handbook/MCOB/11/3 New s55A of CCA 1974 implementing 2008

Directive but it requires only advice and information and does not contain IRR as such. It requires creditors

to provide adequate explanations of credit and must include: 55 (A)(2) (a) the features of the agreement

which may make the credit to be provided under the agreement unsuitable for particular types of use, (b)

how much the debtor will have to pay periodically and, where the amount can be determined, in total under

the agreement, (c) the features of the agreement which may operate in a manner which would have a

significant adverse effect on the debtor in a way which the debtor is unlikely to foresee, (d) the principal

consequences for the debtor arising from a failure to make payments under the agreement at the times

required by the agreement including legal proceedings and, where this is a possibility, repossession of the

debtor’s home, and (e) the effect of the exercise of any right to withdraw from the agreement and how and

when this right may be exercised and check creditworthiness which may include checking

databases. Creditworthiness Assessment must be made before making agreement or significantly increasing

credit under the agreement: Creditworthiness assessment must be based on sufficient information obtained

from — (a) the debtor, where appropriate, and (b) a credit reference agency, where necessary (55(B)(3)

The Lending Code (a voluntary code overseen and enforced by the Lending Standards Board—for further

details see http://www.lendingstandardsboard.org.uk/ ) requires that “before lending any money; granting

or increasing an overdraft, or other borrowing, subscribers should assess whether the customer will be able

to repay it.

iff/ZEW – Final Report on interest rate restrictions in the EU 119

France, Malta, and Portugal. In Spain the term is interpreted as ‘abusive’, which indicates

a high threshold in order to be satisfied and this is also reported in Denmark where the

terms of agreements have to be flagrantly unfair in order for the courts to intervene.

1.5 Social impact of high-cost credit

This part of the research focuses on the legal aspects related to high-cost credit. The

expression “usurious” is applied by lawyers and judges where the cost of credit is

unacceptably high. The level of acceptability various significantly in the different Member

States. In the Catholic culture, high interest still tends to be identified with exploitation;

in Protestant cultures on the other hand, high interest reflects high risk. Interest rates of

more than 100% on small loans are thus acceptable in some countries while they are

unthinkable in others.

Both notions “high” and “usurious” are thus more representative of attitudes within the

EU than indicative of an objective threshold. They reflect the impact of higher cost on

liquidity, solvency and the productive use of consumer credit on the one hand, and the

view that access to credit, even at high cost, may help overcome a liquidity crisis or give

access to necessary opportunities for future earnings on the other.

1.5.1 Credit products

1.5.1.1 Non-bank credit and small loans

High interest, especially for small amounts of credit, pay-day loans, credit card credit,

sms credit and other specialist forms of credit have been mentioned by regulators and in

various reports.114

The assessment of the questionnaire responses, including those from experts and

stakeholders, is set out in the table below. Respondents were asked to reply to the

question “How would you assess the different credit forms/types/products with regard to

the attributes/market features below?” by means of a scale between 1 (insignificant) and

5 (very significant). The responses reflect the findings of the French government115 that

prices of small loans are especially high and that revolving credit systems tend to operate

to the disadvantage of consumers.

114 See for the UK: Office of Fair Trading (OFT) (2010); Collard, S., Kempson, E. (2003); DTI (2006); New

Economics Foundation (2009); for France: Inspection générale des finances, Inspection générale des

affaires sociales (2009). A 2006 study in Ireland showed that money lenders were charging high cost credit

with research showing finding that ‘Four out of 10 people on social welfare are in the grip of moneylenders,

who are charging interest rates as high as 188 per cent.’ Research carried out by University College Cork as

sited in Office of Fair Trading (2009) (Annexe B: A report by Europe Economics for the OFT - International

research: Case studies on Ireland, Germany and the United States, December 2009, p 47.

115 Inspection générale des finances, Inspection générale des affaires sociales (2009).

120 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 24: Assumed problems with different credit forms/types/products

High cost

credit

Consumer

detriment

Lack of

transparency

1 SMS 4.5 1 SMS 3.6 1 Pawnbroking 2.3

2 Payday 3.8 2 Credit card 3.4 2 Point-of-sale 2.8

3 Pawnbroking 3.5 3 Other loan 3.3 3 Home loans 2.9

4 Home loans 3.5 4 Point-of-sale 3.3 4 Payday 2.9

5 Point-of-sale 3.4 5 Overdraft 3.2 5 SMS 3.1

6 Credit card 3.2 6 Payday 3.0 6 Overdraft 3.1

7 Overdraft 3.2 7 Home loans 2.9 7 Other loan 3.2

8 Other loan 3.2 8 Auto 2.9 8 Auto 3.2

9 Auto 2.8 9 Mortgages 2.8 9 Credit card 3.3

10 Mortgages 2.4 10 Pawnbroking 2.7 10 2nd charge 3.3

11 2nd charge 2.3 11 2nd charge 2.7 11 Mortgages 3.5

Source: Stakeholder Survey. Note: Scores in the table reflect the mean averages of all responses received to

question SQ 4.3. The scale for answers was between 1 (insignificant) and 5 (very significant).

Reliable empirical data as to the prices of different forms of credit according to size, form

and distribution channel are only partially available and are confined to the UK and

France. The economic part of this report will show the overall cost of consumer credit in

general. Comments from the experts may help to identify possible problem areas.

In twenty-five Member States116, experts reported concerns as to the level of interest and

charges in their respective credit markets. In Bulgaria in particular, high interest in all

sections of the Bulgarian credit market were mentioned. In the other Member States,

high prices were reported only in specific market segments, with considerable variations.

A breakdown of the areas of the credit market causing concern among our experts across

the Member States is provided in the table on the following page. The key observations

are:

In Italy, Malta, and Austria concerns about the price of credit associated with the

purchase of goods (where point of sale contracts may also lack cost transparency)

were reported.

In four Member States (Austria, Germany, the Netherlands and Spain), all

countries where nearly all credit is extended via the banking system, problems

with the cost of unauthorised overdrafts on current accounts were reported. It

was noted that the level of interest charged by banks was sometimes difficult to

determine because the true cost of credit was hidden in other costs but, once

these elements included in the calculation, bank interest rates could be as much

as 30% (Germany) when the cost of refinancing and insurance premiums were

taken into account.

In six countries, revolving credit was reported as a source of high interest. These

include Luxembourg, which does not have a significant problem with high-cost

credit in general, but where the Luxembourg expert considers international credit

card providers as charging high fees when credit limits are exceeded. In Italy, the

report reveals that a credit card provider was prevented by the authorities from

116 No problems with specific products were reported by the expert from Cyprus and no information is available

concerning Greece.

iff/ZEW – Final Report on interest rate restrictions in the EU 121

issuing revolving loan cards because of the high interest rates being charged,

which exceeded the usury rates.

In twelve states, experts reported that problems of high-cost credit have been

identified in the non-banking and specialist lending sectors in particular. According

to the Office of Fair Trading, the cost of short-term loans provided by

moneylenders to lower-income groups are a particular problem in the UK. Similar

information comes from our Irish, Polish, Latvian and Hungarian experts. Payday

loans, auto-title lending, rent-to-own credit, and pawnbroking were also

mentioned.

In four states (Denmark, Finland, Estonia and Slovenia), experts and consumer

organisations were particularly concerned by the cost of SMS loans, which were

observed to charge average rates, up to 2,000 percent in Denmark, for example.

Table 25: Incidence of high-cost credit in EU Member States

No problem reported Cyprus 1

Bank credit: over-running

on accounts

Austria, Germany, Netherlands, Spain 4

Revolving credit France, Germany, Italy, Luxembourg, Portugal,

UK

6

Non-banks/specialist

lenders

UK , Poland, Portugal , Romania , Czech

Republic , Hungary , Ireland, Slovakia, Latvia,

Lithuania, Sweden

12

SMS Loans Denmark, Estonia, Finland, Slovenia 4

Auto leasing/hirepurchase/

point-of-sale

Italy, Malta, Austria 3

Pawn-broking Portugal, Belgium 2

All types of credit Bulgaria 1

Source: Expert survey

High-cost credit is generally linked to sub-prime customers and is therefore often

associated with development of a sub-prime credit market as the following table

reveals.

122 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 26:Development of EU subprime lending markets

1.5.1.2 Revolving credit

The CCD 2008 has given revolving credit a “light regime”. Overdraft credit for less than

one month (Art. 2 (2) e) is totally excluded while all other overdraft credit is privileged

according to Art. 2 (3). In Art. 3 (3), the “tacitly accepted overdraft” is qualified as a

contractual obligation called “overrunning” and not as “default”. Credit card credit and

open-end credit are not specifically addressed as products which would need more

consumer protection.

The features of such products are especially that they are “revolving” which means that

they can exist eternally because either new credit is taken up before the old credit is

repaid or the residual debt is rolled over into a new loan. The risk of market rate change

is shifted to the consumer via variable rates.

The opinions of stakeholders with regard to such products, as revealed in Table 24,117

show that such products as seen as potentially high cost, lead the ranking scale with

products of this revolving form of credit found in second (credit card) and fifth

(overdraft) position on the scale measuring responses to the question of which products

can be assumed to be detrimental to consumers.

Against this background it may be understood why the Dutch legislator removed the

exclusion from this form of credit from interest rate ceilings and why the French

government focussed on such products in their report which we consider below.

The French government report was initiated by the Ministry of Economy, Industry and

Employment and was conducted by both the Finance and Social Affairs departments. Its

117 See p 120.

iff/ZEW – Final Report on interest rate restrictions in the EU 123

task was to analyse existing links between the modalities of the various usury ceilings

and to investigate the exclusion of certain categories of borrower together with the

persistence of high interest rates in certain market segments. The study was specifically

concerned with providing answers to the policy questions of whether existing usury

ceiling categories should be adapted and whether the calculation method of the ceilings

themselves should be adapted118.

The main findings of this French report119 were that the continuing existence of interest

rate ceilings was justified in France in terms of the protection of borrowers in markets

that are not structurally competitive. It also found that the ceilings contribute to the

prevention of exclusion associated with excessive indebtedness by hindering the

emergence of very high-cost credit targeted at the category of people most at risk and

subject to precarity, and who also prevent the highest risk to lenders.

Revolving credit was the main instigator of the French review of usury ceilings, which

followed an analysis of statistics on indebtedness as well as the publication of a study on

revolving credit and responsible credit.120 According to this study, the problems with

usury are currently concentrated in the area of non-mortgage consumer credit, as

opposed to mortgage credit, and are primarily debated within the context of a specific

form of credit (revolving credit121) and its impact on over-indebtedness. Interest rates are

significantly higher for these forms of credit compared with instalment loans. The report

found from an analysis of central bank data that access to these forms of credit is easier

than to instalment loans and that they are present in more than 80% of cases of overindebtedness.

The French government report122 confirms that for France the ceilings are

one lever available to the authorities in the context of responsible credit as envisaged in

the framework of the CCD transposition.

The findings of the government study of revolving credit included the following:

Consumer access to revolving credit faces no restriction whatsoever (judging from

distribution of interest rates collected by the French central bank) and there

remains evidence that specialist lending institutions (non-banks) do not base their

pricing directly on the level of risk presented by the borrower.

Revolving credit and instalment loans differ significantly as to the amounts

borrowed. Although 50% of revolving credit is for amounts of less than €1,524,

28% is for amounts equal to or greater than €3,000 (falling to an 8% share for

amounts greater than €6,000). It is these large amounts of revolving credit that

appear to be substituting themselves for the typically more appropriate instalment

loans that face a competitive disadvantage from the lower interest rate ceilings

applied to those credit types and those amounts.

The supply of small size instalment loans is as of yet underdeveloped with only

4% of personal loans being extended for amounts less than €1,524.123

118 The reform paths contemplated could be either regulatory (eg. modifications in the credit categories and

ceilings) or legislative (eg. modifications to the principle of fixation which are currently based on market

rates existing in practice, or the principle of uniform application of a coefficient which is currently 1.33).

119 See IGF/IGAS (2009), p.3.

120 See Athling (2008).

121 The term 'revolving credit' refers to credit products that are not paid back in a fixed number of payments,

such as credit cards (also refered to as renewable or permanent credit in France – crédit renouveable/prêt

permanent/credit revolving).

122 See IGF/IGAS (2009), p.4.

123 This has led to certain French stakeholders interviewed for the governmental study being in favour of other

modifications to the existing ceilings including ADIE a microcredit specialist who called for a revaluation of

124 iff/ZEW – Final Report on interest rate restrictions in the EU

1.5.2 Consumer welfare

Opinion as to whether high rates of interest are a problem varies significantly according

to differing perspectives and differences in the development of the market in credit

products.

In five Member States, experts are of the opinion that usury is not an issue in their

countries, either because they have effective anti-usury legislation in place (France,

Netherlands, Luxembourg) or because existing laws do not have to be applied very often

and there are few court cases, which suggests that the incidence of usury is low

(Sweden, Austria). As a consequence, they did not express a view about the social

impact of usury. However, experts in the vast majority of Member States did hold views

on this issue. These included some countries with laws designed to restrict usury but

where nevertheless it was felt that usury had not disappeared completely (for example,

Germany and Portugal), as well as countries where there was an absence of usury law.

The main impacts identified were exacerbation of poverty and hardship, increased overindebtedness,

insolvency and homelessness, increased crime and exclusion of young

people from future services.

There were concerns that high-cost lending is particularly targeted at people on low

incomes and that credit is taken out in order to pay for essential items of household

expenditure as a result of inadequate incomes. This was particularly found to be an issue

in the Czech Republic, Slovakia, Slovenia, Ireland, Romania, Poland, and the UK. In these

countries, usurious loans were reported as leading to:

A spiral of increased indebtedness, often because expensive credit was used to

cover fees and charges for prior default.

Inability to maintain essential payments in other areas of the household budget

(eg. rent and utilities).

In Romania and the Czech Republic, the legal experts had concerns that low education

levels amongst the poor meant that they did not fully understand the consequences of

taking out very high-cost credit for their future finances. However, education was not

identified as the main issue in the other countries referred to above. These countries

focused instead on the predatory behaviour of lenders or on the lack of other, more

affordable, credit options for people on low incomes.

Experts in Estonia and Slovakia reported that the cost of usurious loans contributes to

over-indebtedness more generally, and not only in low-income groups. For example, in

Estonia it was reported that much over-indebtedness derives from high interest rates, or

contractual penalties, which leads many people to borrow more, at usurious rates, in

order to cover the debt in default. This has caused a situation in which large numbers of

people have become personally bankrupt, surrendered their homes and sold their

property. In Slovakia concerns about the practice of securing high-cost consumer loans

property, causing people to lose their homes, prompted an amendment to Article 53,

Section 7 of the Civil Code in November 2008, and this now prevents this practice.

Experts in Italy, Greece, Malta, Romania and Slovakia linked high interest rates to crime

and illegal money-lending (Greece, Malta, Southern Italy) or debt recovery practices

(Romania).

the category by amount of the credit from the current €1,524 delimitation (unchanged and un-adjusted for

price inflation since 1990 when it was converted from FRF10,000) to €2,350 (representing their median loan

value). Providers of pawnbroking credit also subject to the same usury ceilings were also in favour of

increasing the amount of credit in this category but for different reasons as they argued that they were

inclined to provide cash for transactions below €1,524 because the authorised ceiling enables them to

charge substantially higher prices than for loan amounts extended above this amount).

iff/ZEW – Final Report on interest rate restrictions in the EU 125

Whereas existing consumer debt, student loans in the UK, unpaid bills in Germany and

overdraft credit in France were seen as detrimental to the creditworthiness of those

people affected, in Finland concern was raised that the use of high-cost SMS loans will

lead to long-term exclusion from mainstream financial services in the future as many

young people are unable to sustain their payments.

1.5.3 Assumed effects on credit markets

While providers put more emphasis on possible negative effects of IRR on access to

credit for persons on a low income124, nineteen of our expert opinions saw positive effects

for responsible credit markets – some country expert views are shown below.

for the Netherlands the presence of IRR are a factor in preventing the growth of a

harmful sub-prime market,

for Denmark and Luxembourg restrictions on default fees are considered to be an

important protection for consumers at a point when their bargaining position is

weak, but the restrictions are not considered to be onerous for providers,

for Estonia, IRR are considered important for setting boundaries on the cost of

credit. Although these may have had an impact on the profit margins of some

providers, there is no evidence that providers have withdrawn from any markets

as a direct consequence of the restrictions,

for Cyprus, a possible introduction of IRR would according to the Cyprian expert

encourage self- regulation by providers.

There were also concerns that IRR could lead to a focus on less transparent products and

to opaque charges for example

in France, IRR are considered to have narrowed the credit options for near-prime

and sub-prime consumers, increasing their reliance on revolving credit,

in Germany, the IRR have caused lenders to hide charges and fees rather than

include them in the interest rate, reducing the transparency of prices for

consumers.

1.6 Stakeholder views on IRR

With regard to mortgage credit, the European Commission had already collected the

views of stakeholders in its consultation on the Green Paper on mortgage credit in

2006125. Three questions were asked in relation to interest rate ceilings: Do usury rules

impact on integration? Should usury rules be examined in a broader, non-mortgage

specific, context? Do caps on compound interest rates impede integration? The responses

were unanimous that “usury rules” should be examined in a broader, non-mortgage

specific context. Thus, although this study is part of DG Market’s work programme in the

context of the follow up work to the White Paper on Mortgage Market Integration, it is

focused on consumer credit in general, with unsecured credit over shorter terms being

the form of credit most affected by interest rate restrictions.

The following arguments were then raised in the responses from provider associations to

support their view that usury rules represent a barrier to integration and competition. It

124 See Section 2.1 Economic Theoretical background at pp 142ff.

125 Feedback on the consultation on the Green Paper on Mortgage Credit (2006), see:

http://ec.europa.eu/internal_market/finservices-retail/docs/home-loans/feedback_gp-en.pdf.

126 iff/ZEW – Final Report on interest rate restrictions in the EU

was felt that they reduce competition in the market; they potentially hide the true price

of credit products and thus lead to the cross-subsidisation of products; they reduce the

products available in the market; they act as a disincentive to product innovation (with

the result that sub-prime borrowers may be excluded from the market).

In contrast, those supporting usury rules made the following points. Usury rules have no

or only very limited impact on cross-border activity; they are considered as part of

general wellbeing and should therefore be respected by all lenders, irrespective of their

country of origin; Caps are an illustration of the social and human aspects of mortgage

credit. These arguments have been examined in detail in our analysis section (Part 2) in

relation to consumer credit in general. In the closing sections of Part 1 below, we provide

additional stakeholder views on issues related to IRR. Further tables with selected

responses can also be found in Annex XI: Tables of stakeholder responses.

1.6.1 Political and legal discussions

While in half of Member States there were no specific political and legal discussions

aimed at making credit more affordable for consumers, in the other half various political

and legal approaches were under discussion and had been partially implemented.

In some Member States the implementation of usury ceilings for consumer credit

have been discussed in the recent past (Slovakia in 2007, ongoing discussions in

Hungary and in the UK in 2004 and in the context of the Financial Services Act 2010).

In Portugal, transposition of the CCD 2008 into national law had the result that, that

since the beginning of the current year 2010, usury interest rates are established

(quarterly), as are rates applicable to some sub-types of revolving credit and

instalment credit.

In France, by contrast, there was discussion of lowering usury ceilings on revolving credit

by modifying existing ceiling categories. Another approach was discussed in Slovakia,

where the idea of a license to provide money was advanced as a means of preventing

usury.

The discussion about the regulation of exorbitant interest rates and fees for consumer

credit, in particular credit card credit and store cards, was interrupted by the national

election in the UK and has not been concluded. At a political level, there has been

discussion of whether usury ceilings lead to greater credit exclusion and a less

diverse set of credit options for lower-income consumers and whether they help

consumers in that situation. The government has currently increased the social fund,

which provides interest-free loans to low-income consumers (70% of payments are made

to lone parents and the disabled). In the UK, there is currently a great deal of focus on

the affordability of credit, and although instant loan companies and payday or doorstep

lenders with high interest rates are seen as problematic by some consumer advocates,

because of the far higher than average interest rates charged to their customers, recent

conclusions from the UK Consumer watchdog the OFT and the Financial Inclusion

Taskforce stress that the context is a market segment that is simply lacking alternatives

from the supply side of the market, which could imply that the persons currently being

served could face other problems should this high-cost source of credit be removed.126

There has also been discussion of specific credit products in the recent past. In Austria

there was discussion of linkage of overdraft interest rates to the Euribor Index in order to

reduce the interest rate. In Denmark SMS- and web-loans have recently led to political

discussions. In Lithuania and Sweden, the regulation of micro-credit and short-term

126 See Financial Inclusion Taskforce (2010).

iff/ZEW – Final Report on interest rate restrictions in the EU 127

loans, which are not regulated by the current CCD, are under discussion at a political

level and short-term loans have been integrated into consumer credit regulation. In

Belgium, there has been debate of the imposition of a limit on the early repayment fee

charged on mortgages.

In Scandinavia and the Baltic states, SMS loans are under discussion.127 On February 25,

2009 Estonia passed an amendment to the principle of good morals contained in Art. 86

of its Contract Law (GPCCA), specifying that a contract is void if one party knew or

should have known that the other party entered into the transaction because of urgent

needs. This principle is similar to that developed by the German Supreme Court in 1981,

which also reversed the burden of proof of the intention to take advantage of the

weakness of another.

In Italy, a number of measures have made specific forms of credit more flexible for

consumers in order to soften the burden of the financial crisis and unemployment.

Following legislation linking state subsidies to more generous conduct towards debtors

affected by unemployment (Tremonti decree), the Italian Banking Association (ABI)

recommended to its members that they should offer a voluntary stay in possession

procedures, or even payment holidays of at least 12 months in relation to mortgage

loans in particular.128

In Poland and the Czech Republic, on the other hand, there is an ongoing debate about

making consumer loans harder to obtain because of existing problems with increasing

credit volumes and over-indebtedness, rather than making credit more affordable.

Table 27: Legal and political discussions

Implementation of usury ceiling Slovakia, Hungary, UK

Specification of usury ceilings (definition, specific

products)

Portugal

Modifying/lowering usury ceilings France

Discussion of exclusion of consumers in case of usury

ceilings

UK

Licence for money lending to avoid usury Slovakia

Indexation of interest rates Austria

Regulation of specific credit types (micro-credit, shortterm

credit)

Lithuania, Sweden

Flexible rules and soft law in case of default Italy

Discussion of whether lending should be constrained to

protect consumers

Poland, Czech Rep.

Source: Expert survey.

127 See Saare, K., Sein, K., Simovart, M.A. (2010), pp 129 -142.

128 Piano Famiglie of December 2009.

128 iff/ZEW – Final Report on interest rate restrictions in the EU

For more information on the level of discussions of IRR in the different Member States

and details on the studies produced on the subject in those respective countries, see

Annex XV: Stakeholder general feedback on IRR. The Annex also sets out some of the

policy concerns identified in Member States without interest rate ceilings as relevant to

the decision not to introduce them. Many of the arguments and recurrent concerns that

introducing interest rate ceilings could reduce access to legitimate sources of credit and

increase the use of unlicensed sources of credit, are set out in detail and evaluated in

Chapter 2 of this report. On the other hand, in four Member States (the Czech Republic,

Denmark, Finland, and the UK) there have been plans to review the position on ceilings

and to reassess the viability of this form of regulation in the future.

Two bills regulating the IRR have not been adopted. Several reasons were

identified as to why interest ceilings are unlikely in the near future: 1) strong

lobbying by the credit industry; 2) liberal position of National Bank (no ceiling is

necessary); 3) the weak position of consumer protection organisations (Czech

Republic).

A ministerial working group is expected to publicise a report analysing the effects

of introducing APR ceilings (Denmark).

Fast loans are harmful, perhaps ceilings and other restrictions are needed

(Finland).

At present there are no plans to change state policy regarding IRR, however if

public pressure for restrictions rises, this might be evaluated and considered in

the future (Latvia).

The position on interest rate ceilings was reviewed by the UK Government in mid-

2009. No changes were introduced. UK credit regulation was fully reviewed and

amended in 2006 and has since been further amended to incorporate the

EU CCD 2008. No further changes are considered to be in consumers' interests.

The new UK Government's five-year programme proposes new regulatory powers

to cap interest rates on credit and store cards although here the primary issue is

probably re-pricing. The new Government will also be influenced by the OFT's

review (UK).

Political pressures were identified by respondents the most likely factors in policy change

and several respondents shared the view that, as elections approach, there is a tendency

for politicians to favour more populist measures. They considered that IRR could be one

such measure.

1.6.2 IRR and consumer over-indebtedness

One of the main objectives of IRR and capped interest rates in particular is seen as the

prevention of overindebtedness.129 There is also extensive literature about the factual

reasons for overindebtedness and its relationship with short term credit.130 In social

science there is a consensus that overindebtedness is primarily related to unemployment,

129 See the official reports in France and the UK cited at FN 24 and 25 as well as the information given above at

Table 4: Main reasons for introducing IRR.

130 A summary of the situation in the EU in 2003 including national legal rules concerning the prevention and

rehabilition can be found in a project report by iff for DG Health and Consumer Protection Directorate-

General in the first half of 2003, see: Reifner, U. et al. (2003). Books with still valid information are Reifner,

U., Ford, J. (1992); Hörmann, G. (1993); Ford, J. (1988); Berthoud, R., Kempson, E.: (1992); Lord

Crowther (1971). Knobloch, M., Reifner, U., Laatz, W. (2008); Knobloch, M., Reifner, U., Laatz, W. (2009);

Holzscheck, K., Hörmann, G., Daviter, J. (1982). Domont-Naert, F. (1993); Niemi-Kiesiläinen, J., Tala, J.,

Wilhelmsson, T. (1991); Caplovitz, D. (1963); Caplovitz, D. (1974); Duhaime, G. (2003).

iff/ZEW – Final Report on interest rate restrictions in the EU 129

loss of income, illness and separation (80%) and secondly to excessive and irresponsible

borrowing (20%). A vast body of rules, in contract law and insolvency law which have

been put together in another project131 have been developed to manage the rising

problems of overindebtedness which are especially visible in countries where consumer

credit has become the basis of individual consumption.

This report draws only the opinions of experts and stakeholders as to the reasons for

overindebtedness, and these are summarised as follows:

A sudden decrease in income because of unemployment, divorce or illness, at the

moment also caused by a shifting economy (Austria, Finland, France, Germany,

Ireland, Netherlands, Portugal, UK);

Low income (Czech Rep, Ireland, Netherlands), increasing living costs (Poland)

and;

Overestimation of ability to repay credit (Czech Rep, Netherlands).

Unemployment was cited as the most important factor in over-indebtedness. The focus

was not on dangerous credit products or supplier or consumer behaviour, but on

changing household circumstances. The answer from the Dutch expert summarised the

point:

There are four causes: 1. Survival debts: the debtor has insufficient income to meet

living expenses; 2. Overspending: the debtor has enough money, but has too many

loans, spends too much and therefore has financial problems; 3. Adaptation Debts:

debts are the result of a significant change in expenditure or income, for example in

situations of divorce, unemployment, disability and so on; 4. compensation debts:

these debts are caused by psychological problems such as drug addiction or

gambling. These causes are the real problem, consumer credit merely exacerbates

existing problems.

In addition to consumer credit, other types of debt were cited as a reason for overindebtedness,

such as energy, healthcare and telephone bills, tax debts and rent arrears.

The official French report and an industry sponsored briefing note on the subject of IRR

also point to specific credit products, particularly small loans and revolving credit, which

are overrepresented in overindebted households.132 This is easy to explain by the fact

that people on low incomes, with no prospects and/or existing debt do not qualify easily

for long-term instalment credit or mortgage loans. If they cannot pay the monthly

instalments they have recourse to small and easy access credit which is provided in the

form of credit card credit, payday loans or overdraft credit. In these segments, providers

are often more generous with regard to creditworthiness since typically only small

amounts are at stake, the sum can be rolled over and the price of the loan is set higher

than in a more competitive consumer credit market.

This is reflected in some responses where strategies and types of providers were

mentioned. Products with significantly higher interest rates as a driver of

overindebtedness were mentioned by the experts in Estonia, the Czech Rep and Austria.

In Ireland, non-banking institutions which lend at interest rates of up to 187% APR were

mentioned and the Danish expert and the Finnish Consumer Ombudsman both cited SMS

loans. In France, revolving loan accounts were specifically named as most used by overindebted

households (10% of over-indebted households have mortgage loans, 91% have

131 See Reifner, U. et al. (2010); Whitford, B., Ramsay, I., Niemi-Kiesiläinen, J. (2009).

132 See IGF/IGAS (2009) and Project Associates (2009).

130 iff/ZEW – Final Report on interest rate restrictions in the EU

revolving credit and 50% have instalment credit). The Belgian expert reported that such

credit contracts have the highest default rates compared with other credit types.

Table 28: Late payments per credit type in Belgium in 2009

Revolving loan accounts 44.9%

Instalment loans 36.3%

Deferred payment in sales contracts 10.6%

Mortgage loans 7.9%

Financial leasing 0.4%

Source: 2009 Statistical report by the Central Individual Credit Register, www.nbb.be, pp.58-61 (figures on the

collective debt settlement procedure).

The assessment of mortgage loans varied. While in Hungary and Ireland, mortgage loans

were named specifically as risky credit products, in France and Belgium default levels in

these segments were said to be lower than average.

Easy access to new forms of credit product was also cited in Estonia and Denmark,

specifically SMS loans and Web loans. It was reported from Latvia that the absence of

regulation and control mechanisms in the past had led to a significant rise in overindebtedness,

which is now a serious problem in that Member State.

The level of consumer credit penetration explains why the experts in some Member

States saw over-indebtedness as more of a problem for the future (Italy, Malta).

Changing bank behaviour in encouraging consumers to borrow money more readily was

seen as a source of concern.

Table 29: Specific types of credit, behaviour and providers in relation to over-indebtedness

Type Example Member States

identified

Credit products with

significantly higher

interest rates

money-lending, SMS

loans, Web loans

Estonia, Czech Rep,

Denmark, Ireland

Non-banking

institutions, sub-prime

sector - significantly

higher interest rates

intermediaries, pawn

brokers

Czech Rep, Austria,

Ireland, Portugal

Revolving credit credit card credit France, Belgium, Portugal

Mortgage loans irresponsible lending, high

loan to value ratio,

subprime second charge

lending

Ireland, Hungary, UK

iff/ZEW – Final Report on interest rate restrictions in the EU 131

Easy access / new

forms of credit

SMS loans, Web loans,

point of sale financing,

credit card credit

Estonia, Denmark

Coping strategies of

consumers

new credit to pay back old

debts etc.

Germany, Italy, Poland,

UK

Behaviour of banks Encourage consumers to

borrow more, increasing

per capita level of

consumer credit

Ireland, Italy, Malta

Lack of regulation in

the past

Latvia

Source: Stakeholder and Expert survey.

Member States where stakeholder respondents did not unanimously see overindebtedness

as being a problem in their country include Belgium, Denmark, Finland,

Italy, the Netherlands, Slovakia, Spain and the UK. Responses identifying the problem as

very severe came from Austria, Denmark, France, Hungary, Latvia, Lithuania, Poland,

Portugal, Spain, Sweden, and the UK.

With regard to the recent trend in over-indebtedness, there has been a clear

deterioration because of the economic and financial crisis (and its negative effect on the

labour market) over the past two years.

However, specific stakeholder responses may be identified with an assessment that there

has been a marked deterioration, and these may be contrasted with responses reporting

a stable or only marginal deterioration. One UK respondent even said that he “cannot

answer this question, because we disagree with its assumption that there is 'a problem'”.

A relatively small percentage of UK consumers have problems with unsecured consumer

credit, but over many years, this percentage has remained remarkably stable. This

reflects the fact that the key driver of problems are unexpected life events. A recent BIS

survey suggests that arrears levels may even have fallen in the last twelve months”.

The following situation described by the expert for Denmark, can be taken as an example

of the general developments. As banks became more lenient in their credit policies in the

years prior to the financial crisis (access to credit was eased for a given level of

disposable income), finance companies had to become even more lenient. In the end

almost anybody could take out consumer credit, regardless of their level of disposable

income. Consumer credit was widely available in retail stores, grocery stores, and

consumers were sent pre-approved credit offers without even asking for them.

Furthermore, credit cards were marketed aggressively by grocery stores, trade unions,

banks, shopping centres etc. As real estate prices were increasing rapidly, people felt

richer and were taking out more credit. This also had a 'keeping-up-with-the-Joneseseffect

and social norms regarding buying goods on credit were changing. Deregulation of

the mortgage market and the lowering of taxes helped fuel the price-bubble resulting in

increasing levels of indebtedness and thereby increasing the risk of over-indebtedness.

Unemployment is now rising and contributing to the increasing numbers of people unable

to meet their financial commitments.

Table 30 shows a few different responses from stakeholders for countries where the

overindebtedness trend is seen as worsening significantly for consumers. Further

132 iff/ZEW – Final Report on interest rate restrictions in the EU

responses on the extent to which over-indebtedness is a problem and/or faces an

increasing trend are available in Annex XI: Tables of stakeholder responses.

Table 30: Member States facing a significantly worsening trend in over-indebtedness

Latvia Government

official

Due to recent financial and economic crisis, unemployment has

increased significantly in Latvia, causing dramatic increase of

indebtedness.

Portugal Consumer

Organisation

Recourse to the courts (insolvency process) and extra-judicial

mechanisms of debt mediation has increased dramatically in recent

years.

Lithuania Other The Authority receives more complains from the consumers in debt.

Czech

Republic

Consumer

Organisation

The Czechs have discovered credit in the last decade. In recent

years, household debt grew fast, but in connection with the crisis

now rising unemployment and declining ability to repay the loans.

Irresponsible borrowing also plays a role. Insufficient regulation of

consumer credits and insufficient regulation of arbitration

proceedings.

Spain Consumer

Organisation

The family overindebtedness has grown systematically to top of the

income in the later years, so the ratio between both variables has

grown from 45% in 1995 or 76.7% in 2001 to exceed 140% in 2008.

United

Kingdom

Consumer

Organisation

The problem has worsened recently because of the effect of the

downturn in the UK economy, with a steep rise in unemployment

through early 2009. However there is some evidence that consumer

credit debt levels were beginning to stabilise just before the

recession (2006-2007) as growth rates in consumer credit lending

slowed significantly and some 'responsible lending' initiatives (such

as better data sharing) arguably began to have some impact on

lending practices.

France Government

official

This problem has improved as data suggest. The system for dealing

with cases of overindebtedness has a single point of entry, the

household debt commissions, whose secretariats are administered by

the Banque de France. During the last few years, the number of

cases these commissions had to deal with increased significantly

(+15% between 2008 and 2009). The number of households

currently concerned by this problem is roughly about 750,000.

Estonia Other Very easy access to different loans during past five years and current

situation of the labour market has significantly worsened the

situation.

Source: Stakeholder survey.

1.1.1 Preferred characteristics of an interest rate ceiling

The details of the interest rate ceiling put in place by the authorities are important and

analysis of the their effect and their administrative complexity will very much depend on

how and at what level they are set and for which credit markets. When stakeholders

were asked about the form of regime which interest rate ceilings should take, there was

a clear majority (68%) in favour of relative rates in preference to a method based on an

absolute fixed rate ceiling. Likewise, 80% of respondents would be in favour of different

ceilings calculated for different credit types as opposed to a single ceiling applicable to all

consumer credit. This reflects the need to understand the differences and consider the

details of the sub-markets when envisaging their regulation. One provider respondent

helped clarify this need to recognise the heterogeneity of credit products as follows:

“We strongly refuse the idea of caps, but if there should be some level of regulation, it is

better to make differences between loans based on their maturity, value, surety, type of

provider and type of customer served. For instance it is impossible to cap mortgages the

same way as credit cards, personal loans, car leasing and revolving credit.”

iff/ZEW – Final Report on interest rate restrictions in the EU 133

Respondents repeatedly stressed that there are clearly distinct sectors and different

products within the market and that this would require different types of regulation.

Whereas respondents emphasised that a mortgage credit is hardly comparable to a

consumer credit (in terms of duration, rates, amounts, other conditions etc.), others said

that the different level of consumer protection between mortgage loans and consumer

loans justifies the difference in treatment. One UK respondent said that “if a rate cap

were to be applied, out preference would probably be for caps to be capable of tailoring

to different market sectors. It is hard to see how a single rate for mainstream mortgage

credit and short term unsecured credit to higher risk borrowers would have any real

meaning”.

A French respondent rightly pointed out that the different ceilings should not be by type

of loan but by size of the loan extended (amount of money borrowed) as this according

to her would be more efficient. Likewise, several respondents agreed that it would seem

that levels need to be much higher for small and short-term loans than for large and

long-term loans (Finland). Below are examples of some of the responses in favour of

different ceilings. The caps have to be different. The credits have a different nature, the

commitments of the consumers have to be different too (France). The market is just the

difference between the offered goods. Unification leads to a reduction in diversity of

supply. The same applies to credit caps (Czech Republic). To reflect the different product

types and how they are operated/used, eg. to reflect the different amounts of credit

typically associated with different product types (UK). We cannot imagine -if at all- the

existence of one unique ceiling level for all types of loans, given the huge variety among

them (Spain). A unique ceiling would be ineffective for large credits and too effective on

small credits (France). One of the classical (but relevant) arguments against any ceiling

is the relationship between security of the loan and the rate level (Czech Republic).

The main arguments put forward for preferring a unique ceiling were related to the scope

of the interest rate restriction in the country of the respondent, whereby mortgage loans

may not be subject to the ceilings affecting unsecured consumer credit. We reproduce

here a few examples of stakeholder comments. Cap should be set for unsecured loans.

Regulation should aim simplicity of understanding (Slovenia). There is no need for

complex and detailed rules (Netherlands). There is no need to have many different levels

(Finland). It is easier to explain to consumers if there is only one level (for non-mortgage

consumer credits) (Netherlands). It is quite hard to distinguish between types of loans

(Netherlands). Unique ceiling otherwise banks will give up the types of loans with the

lowest rates and try to promote the most expensive ones (Belgium). This last point is

exactly what the French experience has shown and which has been thoroughly

documented and researched in the governmental study published last year.133

Figure 3: Preferences with regard to ceiling characteristics

Source: Stakeholder survey. Note: Using survey responses to questions SQ 1.27-31.

133 See Chapter 2.5: Discussion of the hypotheses.

134 iff/ZEW – Final Report on interest rate restrictions in the EU

The precise answers on details of the IRR mechanism will depend on the circumstances

and the details on a case by case basis and one respondent rightly pointed out that no

systematic answer can be given, however these revealed preferences do suggest that an

APR level for the maximum price is the more appropriate measurement to use when

fixing and monitoring consumer credit offers so that they remain below this ceiling

(70%). The borrowing rate was not seen as a sufficiently robust measure of price for the

purposes of interest rate restrictions for the reason that other costs would be passed on

to the consumer with an equivalent effect to an excessive (above the ceiling) cost of the

credit. Examples from those respondents in favour of the APR as a measure of the ceiling

include: “APR is the basic cost of the loan and that is how the consumers could assess

the loan and compare the loans of different banks” (Bulgaria); “If fees and charges are

not included in the cap, rules will probably be very ineffective and easy to circumvent”

(Finland). Furthermore, a number of respondents stressed that the APR is the rate that is

unified and used in all EU Member States and therefore the most appropriate because

fees and charges are part of the cost for the borrower.

A monitoring of a mixture of both APR and the borrowing cost was suggested by 2

respondents in order to facilitate the keeping of a level playing field (Netherlands,

Portugal) and 5 respondents suggested that the ceiling should neither target the APR nor

the borrowing rate but a wider definition than the current APR eg. in France and Italy

were it was mentioned that all costs should be included if excesses are to be seriously

prevented. Furthermore, two Czech respondents pointed to the example of the Polish

model where additional regulation should accompany the ceiling on the price eg. “APR as

it is computed now does not include everything. Laesio enormis - of certain level of

increase should be considered” and “if there would be a reason for the regulation by

some form of cap, then it is necessary to regulate compulsory associated costs (fees and

charges). Again, it depends on the specific reason of regulation”. With regards to

providers being able to avoid breaching the ceiling by introducing additional fees outside

the interest rate used for the ceiling, one provider association from the UK firmly

responded that “There should be no caps. If there are, any fee or charge that is excluded

will simply be used to circumvent the cap. No jurisdiction has a cap that cannot be

circumvented“. One UK respondent mentioned fees associated to overdrafts when he said

“We would support effective controls of ancillary and contingent fees and charges. It is

not clear how these could be easily incorporated into APR or borrowing rates other than

by way of representative examples that may not be particularly realistic (unless perhaps

maximum charge per period limits are applied) There is an argument that charges should

be incorporated into APRs for cost comparator purposes. However the main area of

benefit here would arguably be unauthorised overdraft charges which we understand the

CCD 2008 does not require to be shown in APR examples”.

1.6.3 Importance and adequacy of IRR regulation in the EU

Results from the stakeholder survey show that deregulation is generally preferred by

provider associations and public authorities and others for all of the four forms of IRR

presented in the table below, with the exception of default interest rate ceilings which

half the ‘other’ stakeholders find necessary.

iff/ZEW – Final Report on interest rate restrictions in the EU 135

Figure 4: Stakeholder preferences for regulation (% who answered yes)

Contractual

interest

Default interest

Variability of

rates

Fees

0%

10%

20%

30%

40%

50%

60%

70%

80%

Other Stakeholder

Consumer Org.

Provider Ass.

Note: Answers to the question “With regard to IRR in your country, are you in favour of IRR? How important is

this regulation for? Interest rates should be capped; Default interest should be capped; Variability of interest

rates in variable rate credit contracts should be regulated; Level of fees and charges should be regulated (SQ

1.26).

1.6.3.1 Regulation of contractual interest rate:

Among the recurrent arguments against regulation of interest rates was that interest rate

caps are shown to create a series of adverse unintended outcomes and tend to harm

those they are intended to protect. Provider associations were quoted as saying

“Competition in Belgium is very strong, which means that lenders selling products which

are too expensive or on too severe conditions will automatically set themselves out of the

market” (Belgium) or “the main impact of the caps would be worsening of the access to

credit. A rejection rate in the banks will increase just because of the individual cost of

risk (of worse-profile clients) will not cover maximum rate” (the Czech Republic). “Any

move to regulate the area would represent a significant departure for the mortgage

model in Ireland and is not necessary, given that competitive mortgage interest rates”

(Ireland). UK Provider Associations generally stressed that rate caps may often create a

series of adverse unintended outcomes and tend to harm those they are intended to

protect, and or they believe that “IRR do not deliver sufficient consumer benefit when

viewed against the impact on the competition, product availability, and consumer choice”

(UK). Whereas consumer organisations and regulators have said “it is better to educate

consumers not to spend money they haven't earned yet; save first, spend later. Of

course, a special case is the mortgage credit. Here we have a misbalance in the present

time: consumers had obtained a credit for a house whose price now has diminished”

(Romania), “a point of departure should be that the market determines the prices, but

for compelling reasons restrictions might be considered” (Finland), or “interest rates

should be regulated by market to offer consumer competitive products for adequate

price. Regulation on IRR might be effective in relation to specific credit types, such as

credits for low-income consumers” (Latvia).

The main reason given for the high importance of regulation of contractual interest rates

was obviously because excessive rates should be avoided, but also that low-income

consumers will be most affected because they more than others take credit with high

interest rates. Further stakeholder views in favour of regulation included to take away

the propensity of lenders (and borrowers) to take higher risks (Netherlands) or to

prevent abusive use of their market power by lenders ie. to ensure a fair competition

136 iff/ZEW – Final Report on interest rate restrictions in the EU

(France). Dutch stakeholders generally agreed that although consumer credit interest

rates should be capped to protect consumers, that in mortgage credit markets, these

loans do not need capping because there is enough other regulation for protection

purposes. One German respondent highlighted the complexity of the regulation of

interest because limits to interest rates were needed to exclude a) abusive rates that are

enforced on consumers in need, b) endangering interest rates that are posing a risk to a

consumer to fulfil the contract, c) instruments to ensure that interest rates cannot

develop in a way that it will cause a damage (such as described in b)). Furthermore it

was stressed that even if a real and transparent competition provides for adequate and

re-financeable rates, this should not be taken for granted and effective action by those

authorities in charge of the financial market and competition should be taken to ensure

this ability is really working.

1.6.3.2 Regulation of default interest:

Regulation of interest rates on default are the most accepted of the IRR forms in the

table above by the different stakeholders generally. 3 examples of the stakeholder views

in favour of deregulation include reasons because default interest is already regulated in

consumer credit and limited to 10% of the contractual APRC (Belgium); there should not

be stricter regulation of default interest rates than for the interest rate of the original

loan (Denmark), or simply that default interest should be agreed between lender and

borrower at the outset prior to entering into an agreement (Malta).

In the UK where no ceilings exist, default interest cannot be greater than the interest on

the initial loan. UK respondents from provider associations agreed that this is an

important safeguard for consumers and that it strikes the appropriate balance between

the interests of the lender (who should be able to cover his losses) and those of the

debtor, who needs to be protected against unexpected and punitive action at a time

when he is in difficulties. However, these comments were made in respect of the UK

credit industry generally because in the case of home collected credit, default interest it

not levied on the consumer and hence the above mentioned rule is not applicable for

them. Another UK respondent nevertheless pointed out that despite Section 93 of the

Consumer Credit Act 1974 controlling default interest and Section 86F limiting interest

charges on default sums to simple interest, there is currently no general control of

compounding of default interest and it could also be argued that allowing lenders to

charge default interest at the agreement rate (rather than say the funding cost) allows

additional profit to arise from financial difficulties.

Among the proponents of regulation are respondents from: Austria, where the argument

was often used that if the default interest rate is capped the costs will be more

transparent; Poland where excessive punishment of a delay in payment should be

reduced; and the Czech Republic where this IRR form was generally seen as forming part

of the general prevention on usurious practices. Examples of arguments from

respondents who deemed that it was very important for default interest rates to be

capped include: In order to protect vulnerable consumers it is very important that default

interest rates are capped (Finland); It is necessary that the cost is reasonable and

especially that either propose a solution based on the situation of the borrower (France);

Regulation on default interest rates is important to prevent over-indebtedness of

consumers. However, it should be general enough stating overall rules and boundaries

and not numeral caps (Latvia).

While default interest rate regulation is seen as necessary for cases of inevitable default,

a number of respondents pointed to the need, prior to that, to have providers giving their

customers a greater ability to fix a contract in particular in times and circumstances

where new economic conditions affect the borrowers situation. Fixing undue

compensation was seen by one German respondent as rather making it even more

unlikely that a consumer can be prevented from being pushed into insolvency.

iff/ZEW – Final Report on interest rate restrictions in the EU 137

1.6.3.3 Regulation of variability of interest:

With regards to stakeholder views on the need for regulation of the variability of variable

rate credit, the majority found existing rules such as not permitting unilateral changes by

one of the parties etc.. as appropriate regulation. Stakeholders alluded to already

existing regulation which has proven important in their countries, but one respondent

from Denmark mentioned that they should be equal for banks and for finance houses.

Two Member States said this IRR form was of great importance for the following reasons:

The unsophisticated borrower is not able to perceive the dangers which come with the

gamble on future reference indexes that come with variable rate offers (France) and the

regulation is very important especially concerning how and to what extent the contractual

interest rate is affected by the changes in the benchmark interest rate, so that

consumers know in advance their exposure to interest rate risk (Greece). A different

French respondent nevertheless pointed out that though regulation is needed as a

consequence of the volatility of interest rates, a soft form of regulation has proven

efficient whereby a cap can be negotiated between the lender and the borrower, which in

his eyes is more flexible than a legal cap.

A Polish consumer association respondent stressed the fact that a consumer must know

what the rules of variability actually are. UK respondents stressed this point too, by

saying that the conditions under which rates can be varied must be transparent to the

consumer at the outset with one respondent saying that these variation clauses are a

necessity for loans over long terms. Whereas one UK respondent said that the variability

in interest rates in variable rate credit contracts should be regulated (and indeed it is -

not least through European law on unfair contract terms) to ensure fairness and

transparency, he did not however agree that it requires regulatory thresholds on the

extent of variability. Another UK stakeholder said that current UK case law on lender's

ability to vary loans (fixed term in particular) is very unsatisfactory, despite currently

only limited evidence of widespread abuse by lenders. However in the UK there has been

recent voluntary agreement by UK card issuers on re-pricing credit card debt and more

regulatory action is perhaps likely on this.134

Arguments put forward by those who were not in favour of regulation included: There

should be regulation on the total debt loading burden level, not the form of the loading

(Czech Republic); There does not seem to be reasons to interfere with the market (the

Netherlands); The variability of interest rates should be regulated only by contractual

terms between the bank and its customers (Italy); Variable rate contracts are often

open-ended. Lenders therefore need to be able to vary the rate of charge over time to

reflect both the cost to them of raising money and the borrowers' risk. The latter will

undoubtedly evolve over time, especially for products such as credit cards which may last

for 30+ years. If lenders were not permitted to vary rates in this way, such contracts

would not be offered (UK).

With regards to mortgage credit, a Belgian provider association clearly stated that

deregulation was very important. This is mainly because regulation regarding variability

is very strictly regulated in Belgium with the maximum variability of the mortgage credit

rates is annually (which is interpreted in neighbouring countries as a type of fixed rate),

but also because they have reservations on the way the restriction mechanism operates:

“The reference rate is an index of the assets side of the balance sheet of the lender and

should be an index regarding the debts of the lender. From a prudential point of view, it

makes no sense to provide for caps on variability knowing that the market rates can

fluctuate as strongly as was the case in the early eighties. The calculation technique for

the variability of the rate is incorrect from a financial point of view: the variation should

134 See for instance: http://www.theukcardsassociation.org.uk/files/credit_and_store_card_review_-

_joint_government_industry_statement_15.03.10.pdf.

138 iff/ZEW – Final Report on interest rate restrictions in the EU

follow the fluctuation of the reference rate in the sense that a contractually fixed margin

will have to be added to the reference rate. The restrictions on the annual variability are

unacceptable from a prudential point of view.

1.6.3.4 Regulation of fees and charges:

Below are examples of responses from the different stakeholders regarding their views

on regulation of fees:

An Austrian government official tends to find a regulation concerning the level of

fees and charges welcome but also mentioned that such a regulation may

minimize the competition. From the consumer's view it would be important to

regulate a maximum level of fees and charges.

A Belgium Provider Association mentioned that competition was very strong.

Considering this, lenders selling products which are too expensive or on too

severe conditions will automatically set themselves out of the market. The

information disclosure (prospectus, tariff of the interest rates and the costs)

seems important to them.

A Czech Provider Association agreed that regulation should not be implemented

on costs and fees. Another even pointed out that according to their opinion a

regulation of prices would not necessarily result in a lesser payment for

consumers but rather the contrary would be the case. Furthermore, they doubt

that a regulation would mean wider access to credit or a wider range of products

offered.

A Bulgarian Provider Association stated that competition between the banks would

ensure adequate fees and charges as well as adequate service quality. According

to them a possible regulation may have a negative effect on the free market, and

that there should be, however, a requirement for full disclosure of fees and

charges related to every loan product, in order to eliminate the possibility for

banks to collect hidden charges.

A Provider Association from Malta stressed that the level of fees and charges

should depend on market forces as well as the cost structure of the lending

institution, a view shared by the UK Provider Associations who also favoured

deregulation because while the focus in the UK is on providing transparent

information to consumers, lenders should nevertheless be able to set charges and

fees to reflect the administration costs involved. Others in favour of deregulation

of fees and charges but also supporters of an open and comprehensible

calculation of these include a Latvian government official and a Slovenian

consumer respondent who was in favour of regulation by market measures alone

because it felt that caps on the APR are a more effective method of regulation.

A Danish Consumer Organisation pointed out that only an APR-ceiling would be

effective. A sole cap of interest rates would result in higher fees and charges.

They already experience that some providers on the 'second market' charge

excessive fees and/or charges. This makes the loan appear cheaper if the

borrower pays attention to the interest rate only. As many consumers fail to

understand the concept of APR, or how to use it for comparison of different credit

offers, this practice is well established. Finnish regulators also deem a cap on

charges and fees alone as insufficient.

According to an Estonian government official, their market is too small for

effective competition. Therefore providers are in no need to lower interests or

charges.

iff/ZEW – Final Report on interest rate restrictions in the EU 139

Most Provider Associations object an IRR and a cap on fees and charges. From the

received questionnaires also many other stakeholders like banking or financial regulators

from newer EU Member States oppose the idea of a statutorily regulated IRR. Consumer

Organisations seem to be more sceptical concerning the regulation through the free

market alone. As one Spanish Consumer Organisation pointed out that a legal solution

would lead to a balanced society rather then diminish the offer made by lenders. Without

the IRR certain social groups would be excluded from credit means.

1.6.3.5 Adequacy of regulation facing providers

When asked about the adequacy of IRR in terms of the nature of the provider extending

the credit, the table below shows how stakeholders have scored financial institutions and

moneylenders

Table 31: Adequacy of IRR regulation facing banks and non-banks

Banks Non-banks

Overall 3.1 2.5

Consumer Organisations 2.6 1.9

Provider Associations 4.0 3.8

Other stakeholders 3.0 2.4

Source: Stakeholder survey. Note: Average answers from all stakeholders to the question: “How would you

describe the adequacy of the actual level of regulation of IRR faced by banks and non-banks?” (Q1.25 - the

scale for answers was: 1 - very insufficient; 2 – insufficient; 3 - just adequate; 4 - more than sufficient; 5 –

excessive).

We see from Table 31 above showing average scores from responses, that as expected

non-banks are seen to have less adequate IRR regulation than banks, and that provider

responses overall see the level of IRR regulation as adequate. Regarding banks, an

insufficient level of interest rate regulation was reported from Denmark, Finland, France,

Germany, Lithuania and Spain, whereas respondents from Belgium, Czech Republic,

Estonia, Italy, Netherlands, Poland, and the UK judged the level of regulation to be more

than sufficient or even excessive. Respondent views on level of adequacy in the

regulation of IRR affecting non-bank lenders shows a similar grouping of countries where

it is less than adequate with the addition of the Czech Republic to this group and the

removal of Germany. The scores provided for the sub groups making up the non-bank

provider group show that finance companies and mortgage specialists were seen

marginally more adequately regulated than moneylenders overall (2.8 versus 2.2

respectively). Whereas the Belgian provider association affirmed that there was an

efficient legislation of credit in Belgium, it also expressed a strong view that mortgage

credit is ruled by an outdated regulation. Other respondents remarked that some banks

were only banks in name and concern was expressed that any IRR should really need to

apply to all loans granted to consumers irrespective of the type of the creditor.

1.6.4 IRR and other regulatory options

Findings of the survey have also helped to situate IRR with respect to other potential

alternative forms of regulation of credit markets and the overall results give an indication

as to the average regulatory usefulness as seen by the stakeholders surveyed (which are

equally balanced in terms of responses used between provider associations, consumer

associations and other stakeholders made up primarily by the public authorities).

Rankings of results to SQ 1.23 asking which of the following seven regulatory activities

140 iff/ZEW – Final Report on interest rate restrictions in the EU

would have the most pronounced effects on the four desired outcomes shown in Table 32

below.

Table 32: Ranking of IRR as a policy measure alongside other alternatives

In reducing cost of credit In reducing

overindebtedness

1 IRR 3.9 1 Responsible lending 4.4

2 Responsible lending 2.8 2 IRR 3.6

3 Limiting rolling-over 2.6 3 Disclosure 3.5

4 Disclosure 2.5 4 Bankruptcy (general) 3.3

5 Bankruptcy (general) 2.2 5 Limiting rolling-over 3.2

6 Bankruptcy (easier insolvency) 1.8 6 Bankruptcy (easier insolvency) 2.6

7 Bankruptcy (earlier discharge) 1.8 7 Bankruptcy (earlier discharge) 2.4

In improving credit access In widening variety of

products

1 Disclosure 2.5 1 Responsible lending 2.6

2 Responsible lending 2.4 2 Disclosure 2.4

3 IRR 2.2 3 IRR 2.4

4 Bankruptcy (general) 2.1 4 Limiting rolling-over 2.1

5 Limiting rolling-over 2.0 5 Bankruptcy (general) 2.1

6 Bankruptcy (easier insolvency) 1.9 6 Bankruptcy (easier insolvency) 1.8

7 Bankruptcy (earlier discharge) 1.8 7 Bankruptcy (earlier discharge) 1.7

Source: Stakeholder survey. Note: The values are mean averages of answers given to SQ 1.23 (Scale: 1 -

opposite effect; 2 - no effect at all; 3 - very little effect; 4 - some effect; 5 – strong effect).

Interpretation of the ranking above should be subject to caution because the answers

received have a strong element of subjectivity and the choices do not capture the

complexity of the issues. For example, when assessing the impact which interest rate

restrictions may have on the different variables, answers will hopefully have been based

on the assumption of setting ceilings at a reasonable level. The replies will therefore

often reflect more of a personal opinion of the respondent rather than report on the

observed effects locally as in many cases the stated scenarios have not actually been

experienced locally.

The tables nevertheless indicate that IRR are overall recognised as a useful regulatory

option to reach certain policy objectives and especially seen as having an impact on the

cost of credit relative to other regulatory options in consumer credit markets. Some of

the additional comments stressed that

If it were easier to get out of debts it would make it easier to have a fresh start

but that on the other side access to credit would be more difficult. To make

insolvency easier and enable earlier discharge would have a positive effect on the

level of over-indebtedness for obvious reasons but it would make the cost of

credits increase because they would become riskier for the banks (the consumer

would be easier allowed not to pay back). Allowing for easier insolvency processes

were not overall seen as having too much of an effect on either cost, access,

over-indebtedness or the variety of credit types offered, although it was reported

to mean better protection for consumers that are passively indebted,

guaranteeing the coverage of basic needs and facilitating and easing repayment of

debts. Likewise, it was seen by some as uncertain as to whether banks would

significantly change their credit policy should insolvency or bankruptcy become

iff/ZEW – Final Report on interest rate restrictions in the EU 141

easier. Furthermore, it was also specified that insolvency is a method to fight

existing over-indebtedness. Although it may very well improve proceedings to a

more responsible lending as lenders have to care about a success of contract

much stronger than before, this however, does not need to have direct link to

reduced costs or a wider variety of products though it may.

Opponents to regulation generally have reported that all these supposed

protections should end up harming consumers with one respondent stating that

each of the mentioned proposals is likely to lead to a contraction in supply of legal

credit pointing out hat the reasons differ slightly: For instance, in his view rate

caps and responsible lending rules have direct rationing effects. By contrast,

'easier' bankruptcy forces up lender losses, which produce indirect rationing

effects (arising because the bankruptcy losses increase lender costs and make

more customers unprofitable to serve). The same respondent also states that

overall consumer detriment would be the most likely outcome, for the following

reasons: (a) on overindebtedness, a consumer can be 'overindebted' even though

he has no credit debt. Such a consumer's position is worse if he has no access to

the smoothing capabilities that credit provides; (b) on overindebtedness, a

consumer can also be overindebted to an illegal lender. It is worth stressing that

illegal lenders are not concerned with legal controls or protections for consumers;

(c) on overindebtedness, a consumer can be overindebted on retail credit

liabilities (even on '0%’ credit) even though he may have no outstanding cash

credit; (d) on reduced cost of credit, rate caps create displacement effects

whereby credit costs are, for instance, recovered via fees or charges other than

those included in the APR, or displacement effects occur as the market shifts

towards retail credit as a way to avoid the effects of the rate cap. Finally, charges

on illegal loans far exceed any market norms; (e) in terms of access to credit, the

reality is that rate caps, responsible lending rules, constraints on refinancing and

easier bankruptcy all cause credit supply to contract. Reduced supply means it

becomes harder, not easier, for consumers to get credit; (f) all these proposals for

regulatory action would be likely to reduce the variety of products - particularly

small-sum products - available in a market.

With regards to the variety of products, it was generally seen that the regulatory

interventions listed would not increase this and as one respondent said “only

deregulation would achieve that”. In terms of IRR and wider variety of products in

the mortgage credit market, there was a noticeable increase in variety following

IRR regulation in Belgium but one respondent specified that “this was the

experience of the Belgian lenders in the nineties in the mortgage credit market,

but this is not a definite outcome as it depends on the way the restrictions are

formulated”.

IRR were also reported to generate big costs for legitimate lenders as they would

need to take into account the added compliance costs, however the costs of

alternative regulation was not referred to it such answers. Furthermore, one

respondent said that “reduced levels of overindebtedness can only be achieved by

promoting smaller fixed sum loans as opposed to large loans or running account

credit yet all consumer credit regulation actually penalises small fixed sum loans

and promotes running account credit and large loans thus increasing

overindebtedness, not decreasing it”.

142 iff/ZEW – Final Report on interest rate restrictions in the EU

2 Economic Part

2.1 Theoretical background

2.1.1 Outline

This part of the study aims at discussing relevant aspects of the functioning of credit

markets and the role of interest rate restrictions in a comprehensive way. It provides the

economic framework for the hypotheses which are tested in the subsequent chapters of

this study. The theoretical discussion proceeds along the two dimensions individual

choices and market outcomes.

2.1.1.1 The individual choice of credit - an overview

In a first part (2.1.2) we discuss the role of credit from the perspective of economic

decision-making by an individual consumer.135 Acknowledging that consumers make

choices according to their own preferences, we demonstrate in a classical framework how

a decision to finance consumption by credit arises from the preference to smooth

consumption over time (2.1.2.1). As modern economic research has made serious

objections to some of the underlying assumptions in the classical approach, we also

discuss cognitive biases and irrational decisions of the customers in more detail

(2.1.2.2).

Overall, the results of part (2.1.2) can be summarised as follows:

When households decide about their consumption, saving and borrowing, they not

only consider their current income but also their expectations about their future

income.

From an individual’s point of view, credit access may be desired as it allows

households to bring forward future income and compensate for sudden drops in

current income.

Even if interest rates are high, these costs are incorporated into the consumers’

choice of their optimal consumption behaviour. From the perspective of an

individual, interest rate restrictions may be considered to be harmful as they may

make credit less accessible to certain groups of households.

Behavioural biases may lead to systematic (and predictable) deviations from

rational behaviour, which lead to suboptimal consumer decisions. These biases

(such as wishful thinking or underestimation of exponential) challenge the view

that borrowing decisions are an unambiguously rational consumption optimisation.

As a consequence of wishful thinking, consumers’ beliefs about event risk may be

distorted: they may be willing to borrow money even when it is rather unrealistic

that future income will be sufficient to repay the debt.

Consumers may also underestimate the true cost of borrowing. This might occur

because they hold erroneous beliefs about the actual time period during which

they will use the credit or its actual cost (ie. interest rates plus other costs). It

might also stem from conceptual difficulties to understand the effect of interest

rates compounding over longer horizons (underestimation of exponential growth).

The resulting consumption decision might be distorted as consumers do not

135 Different views originate from, eg. Friedman (1957) and Modigliani (2005) on the one hand and, eg.

Kahneman/Tversky (1974), Fehr (2002) on the other.

iff/ZEW – Final Report on interest rate restrictions in the EU 143

properly take into account how much loan repayments reduce their future

consumption.

According to this view, consumers’ credit decisions are not necessarily optimal. Firms

may exploit the consumers’ cognitive biases to increase profits. Limitation to credit

access (eg. through interest rate restrictions) might thus be beneficial to protect

consumers from voluntarily taking decisions that may actually decrease their welfare. As

a consequence of these various aspects, there are competing explanations for overindebtedness:

According to the permanent income lifecycle hypothesis, only unexpected adverse

shocks should lead consumers into over-indebtedness. Such shocks can affect a

consumer’s total resources, as well as on a consumer’s expenditures.

In reality, however, consumers´ consumption plan may a priori be unsustainable

due to limited rationality or irrational behaviour. Further prominent explanations

for why households end up in over-indebtedness are moral hazard (due to

consumer insolvency regimes), market failure (information asymmetry between

lenders and borrowers), lack of financial literacy and supply-driven

over-indebtedness.

The theoretical focus on consumption smoothing as the most important driver of

borrowing takes account of the fact that this study primarily focuses on consumer credit,

rather than business loans.

In the context of businesses, in particular small and medium size enterprises

(SME), credit serves to finance a profitable investment when internal funds do not

suffice.136 Among consumers, in contrast, credit is a matter of preference (under a

budget constraint) rather than of positive investment returns. For companies, the

investment opportunity decides whether or not a credit (for micro companies even

high cost credit) yields profitable returns to the borrower.

However, there are both practical overlaps as well as differences between the

uses of consumer credit on the one hand and business loans on the other.

2.1.1.2 Interest rates and capital allocation - an overview

In a second part (2.1.3) we discuss market outcomes of interest rate restrictions, in

particular with respect to capital allocation and interest rates. We sketch the choice set of

the supply side and demonstrate how lenders allocate their capital to different subgroups

of consumers and why they charge different interest rates for each of these groups. We

discuss the role of interest rate restrictions in different settings. We show that, from a

theoretical point of view, interest rate restrictions may serve as a device of consumer

protection if a policy maker seeks to restrict credit access for specific subgroups of the

population. The results of part 2.1.3 may be summarised as follows:

The level of interest rates charged on consumer credit depends on the market

level of interest rates, the bank’s margin and a component which compensates

the lender for the risk of borrower’s default, which in turn depends on the

collateral, the credit history and the income/wealth situation of the borrower.137

Due to the fixed costs to each loan, small amounts of credit may be relatively

136 According to the pecking order theory, companies first attempt to refinance their investments by internal

sources,before turning to - more expensive - external sources. Among these sources, debt refinancing is

considered to be less expensive than equity refinancing.

137 Galor and Zeira (1993).

144 iff/ZEW – Final Report on interest rate restrictions in the EU

expensive. As the risk of low-income borrowers is perceived to be high, lenders

charge these customers higher interest rates.

Legal interest rate restrictions reduce the lender’s opportunity to charge riskadjusted

costs. Obviously, this decreases the willingness to lend.138 As a

consequence, high-risk borrowers may be denied credit access in the presence of

legal interest rate restrictions.

As lenders are not always fully aware of the borrowers’ riskiness (information

asymmetry), they are unable to accurately estimate risk-adjusted costs. Thus, it

is rational for lenders to keep the interest rate level low and simultaneously

reduce the amount of loans offered in the market. Legal interest rate restrictions

are only effective if they are below such market-based interest rate ceilings. They

are beneficial if the welfare gain from lowering interest rates for those who are

served outweigh the welfare loss due to a reduction of credit availability.

Banks frequently do business with private consumers, many of whom only have

limited experience in financial affairs. Hence, banks can be seen as the more

sophisticated contractor. Recent models have therefore discussed the case in

which banks have an informational advantage over their customers. In this case,

they could lend more aggressively. If aggressive lending takes place in a specific

market, interest rate restrictions can be beneficial, as they reduce interest rates

as well as the group of consumers which obtain credit although they would not do

so if they were fully rational.

2.1.2 The individual choice of credit

2.1.2.1 The Classical view on consumer credit demand

2.1.2.1.1 Permanent income hypothesis and consumption smoothing

The classical economic view on consumer behaviour stresses the idea that households

are rationally anticipating their future income and future financial needs. They do not

know these financial variables with certainty, but are at least able to form reasonable

assumptions about them. Consumers also have a realistic view about the uncertainty

they face in the future.139

As a consequence of this behaviour, households do not necessarily spend the exact

amount of their income they have just earned. Instead, they save money to transfer

current income to the future, and take out loans to dispose of future income to meet the

needs of today’s life circumstances.140

138 Villegas (1982).

139 Following Friedman (1957) and Modigliani (2005), consumers form their expectations regarding future

income based on the expected value of the probability distribution of income in each period. For an

amplification on the measurements of expected income, see Friedman (1957, pp.23-25) and Modigliani and

Ando (1963).

140 See Friedman (1957, p.7).

iff/ZEW – Final Report on interest rate restrictions in the EU 145

Figure 5: The benefits of borrowing and saving

Figure 5 illustrates how consumers may take advantage of transferring or borrowing

money. In this stylised example, households live in two periods, “today” and “tomorrow”.

They have preferences about an ideal combination of consumption today c1 and

consumption tomorrow c2, which are given by the specific curvature of the indifference

curve U. However, households have to take into account how much they earn today and

tomorrow to determine the budget constraint they may not exceed. For this reason, the

position of any feasible optimal combination of c*1 and c*2 must not be to the right of the

straight line B. Now, imagine the household earns P1 now and expects to earn P2 in the

future. Given the ability to save, the household will transfer money to tomorrow so that it

can consume today and tomorrow. Similarly, if the household earns today but expects to

earn P’2 tomorrow, it will borrow money to come to the same combination of

consumption today and tomorrow. Note that, if the household lacks the opportunity to

borrow or save today, it cannot reach the optimal consumption point (c’1,c’2), but will end

up on a lower indifference curve.

It is obvious that this example with two periods is a strong simplification of consumers’

life cycles. Nevertheless, the core idea holds in more complex settings: consumers can

save and borrow whenever they feel that their current income is different from (the

current value of) their expected average future income, regardless of whether a business

cycle shock, a sudden job loss or the stage in the life cycle (eg. student age) has caused

this deviation. Taking account of the complexity of real life circumstances, economists

have also incorporated further aspects into their models, such as

…the role of uncertainty about future income: These models typically imply

that consumers increase their savings (“precautionary savings”) to have some

buffer in the case of adverse events.141

…the role of the life cycle: These models bring a more realistic structure into the

typical consumption pattern of households: Households do not necessarily want to

have an equal amount of consumption in all periods, but consume according to

their needs (eg. raising children).142

…the role of investment into durable goods: At certain points in life, people

tend to spend money on specific goods they will take advantage of for many years

141 See Modigliani and Brumberg (1954) or Lusardi (1998) for an empirical confirmation.

142 See Deaton (1992), pp. 5-6, for an analysis of the influence of external influencing factors on consumption;

and Zeldes (1989) for a detailed description and classification of external influencing factors.

146 iff/ZEW – Final Report on interest rate restrictions in the EU

(eg. housing).143 As consumers prefer to purchase these goods earlier, they tend

to increase borrowing at the early stages of life.

While these models come closer to a more realistic consumption pattern of households,

they still rely on a crucial assumption: Households base their decision on their

(reasonable) expectation of their life-time income, which they consider their overall

“budget constraint”. As a result, they show some pattern of consumption smoothing,

such as that represented in Figure 6:

Figure 6: The idea of consumption smoothing

2.1.2.1.2 The interest rates level and consumption smoothing

Interest rates are both the compensation for saving money and the cost of borrowing. As

a consequence, it is intuitively plausible that consumers borrow less when the interest

rate level is high, and more when the interest rate level is low.144

Figure 7: The effect of lower interest rates

143 See Deaton (1992), pp.10ff.

144 Rising interest rates make consumption today more expensive relative to consumption tomorrow

(substitution effect), and the same burden of interest payments tomorrow finances less consumption today

(income effect). Hence, both the substitution and the income effect lower borrowing when interest rates rise

(see Varian, 2003, pp.137ff).

iff/ZEW – Final Report on interest rate restrictions in the EU 147

Figure 7 illustrates this idea. It follows the above example (Figure 5) and shows two

groups of consumers with identical preferences but different life-time incomes (“rich” and

“poor” consumers) and, as a consequence of their credit rating, different borrowing

interest rates charged by banks. The different interest rates are reflected by the

differently sloped budget constraints: for the poor borrower, it is more costly to borrow

against future income and to consume now. When both groups expect their future

average income to be higher than the current one and prefer to spend more than the

current income, both groups of consumers will borrow money today. However, the rich

group will borrow a larger fraction of the current income than the poor group, as it is less

costly to do so. The distance “a” denotes the amount of money the rich group borrows in

addition because it faces lower interest rates than the poor group.

This example reveals an important point: In the classical view, rational consumers take

the level of interest rates into account when they decide how much of their future income

they currently want to consume. From the consumers’ perspective, high interest rates

make consumption smoothing by borrowing more costly but not necessarily undesirable.

And consumers may be willing to give up a substantial fraction of their total consumption

if it enables them to consume at the “right” time.

2.1.2.1.3 The desirability of credit access from the classical perspective

According to the classical view, households decide about their consumption, savings and

borrowing behaviour according to their preferences, the interest rates level and their

expectation about their future income. This optimal choice may only be feasible when

they have access to the credit market. Following this idea, any institution restricting the

access to credit leads to an inferior situation for the households. It has been frequently

argued (see next chapter) that interest rate restrictions have this effect.

Figure 8: Disadvantage of poor borrowers without credit access

Figure 8 illustrates why limited credit access is problematic from the classical point of

view. If, due to interest rate restrictions or any other reason, poor consumers are

excluded from credit markets, they have to rely solely on their current income to finance

their current consumption. As a consequence, consumption levels may vary significantly

over time (as in the illustration above). As consumers would prefer more balanced

consumption, they are deprived of utility (as represented by the lower indifference

curve). Following this rationale, policy measures to enhance credit access are beneficial

to consumers, no matter how high the charged interest rate.

148 iff/ZEW – Final Report on interest rate restrictions in the EU

It has to be stressed, however, that this model assumes that no borrower ever borrows

more money than she can reasonably expect to pay back in future periods. While

permitting the compensation of temporary negative income shocks by means of credit,

the model does not provide a rationale for credit as a means to “make ends meet” for

deprived households who do not have a realistic perspective of an improvement of their

situation. Hence, permanent financial shortages (compared to desired expenditures)

cannot just simply be set equivalent to the need for credit.

2.1.2.2 A Behavioural View on Consumer Demand

2.1.2.2.1 Behavioural economics

Inspired by findings from psychology, behavioural economics explains the economy by

models which assume that people are not fully rational. In particular, it challenges the

classical view that people form their beliefs correctly when they receive new information.

It also argues that people make choices which are not in line with a maximisation of

expected utility.145 Behavioural economics aims at making more accurate predictions by

incorporating more realistic assumptions about economic actors.146

Psychological experiments have shown that people rely on heuristics which allow them to

reduce the complexity of a problem. While this natural behaviour is useful in many

aspects of daily life, it sometimes comes with cognitive biases. These biases may lead to

systematic (and predictable) deviations from rational behaviour.147 This fact can

sometimes be exploited eg. by corporate managers who adapt products and marketing

strategies according to consumers’ biased behaviour, in order to increase their profits.148

When producers cause substantial harm to consumers by taking advantage of their

irrational behaviour, action of regulatory authorities or policymakers could be advisable.

The optimisation of consumption over time, which has been discussed in Chapter 2.1.2.1

as a rationale for consumer credit, relies strongly on the assumption of consumer

rationality. Besides other aspects, we have assumed that…

…people know their future preferences and take them into account when

making choices in the present.

…people form their expectations about their future income and its variations

properly and take them into account for their budget constraint.

…anticipate the cost of borrowing correctly.

As we will detail in the following paragraphs, these assumptions cannot be taken for

granted with universal validity. Even if there are some customers whose behaviour is

close to rational, there may still be others for whom this is not the case. For the latter,

however, the interpretation of borrowing as a device of consumption optimisation may be

misleading. Instead, borrowing might be triggered by cognitive biases and lead to a

suboptimal strategy from the consumers’ perspective.

145 See Barberis and Thaler (2003).

146 See Rabin (2002).

147 See Kahnemann and Tversky (1974).

148 For example, Stango and Zinman (2010b) discuss how lenders can exploit the fact that consumers

persistently underestimate the APR in credit contracts when it is not explicitly detailed.

iff/ZEW – Final Report on interest rate restrictions in the EU 149

2.1.2.2.2 The role of unrealistic optimism (wishful thinking)

Psychological studies have documented that people tend to process information in a way

which makes their beliefs overly optimistic.149 In particular, people believe that negative

events are less likely to happen to them than to others. The opposite holds true for

positive events.150 People underestimate event risks such as becoming unemployed or

severely ill, and thus overestimate their expected future income. Conversely, in case of

financial distress, wishful thinking might lead to the perception of a consumer that her

income will recover in the future.

As a consequence of wishful thinking, consumers may be willing to borrow money even

when it is rather unrealistic that future income will be sufficient to pay the instalments.

Such behaviour, however, contradicts the idea of consumption smoothing by the means

of borrowing: this concept necessarily implies that households are borrowing within the

scope of their expected lifetime income. They borrow today because they may rationally

expect future income to be sufficient to pay back the loan. This idea might appear

realistic, in particular, for young households who expect their income to rise in the

future. However, when notorious low-income households overestimate their future

income growth, they may find themselves in a situation in which they lack liquidity to pay

back their loan. In the worst case, this initial unrealistic optimism leads to the necessity

to refinance old debt with new debt, although it remains unrealistic that the income

situation will improve in the future.

Other aspects of consumers’ financial behaviour have already been discussed with

unrealistic optimism as a potential cause: some consumers tend to choose credit card

contracts with high interest rates and low fees, although it would be more advisable for

them to choose a different contract with lower interest rates and higher annual fees. It

has been argued that wishful thinking makes these consumers erroneously expect that

they will not overdraw their credit card. As a consequence, they assume that they will

not have to pay interest and do not take the cost of credit into account when choosing

the credit card contract.151

Similarly, it is perceivable that some consumers are overly optimistic with respect to the

time in which they will have repaid the loan: as they wish to repay their debts as soon as

possible, they do not expect their debts to persist.152 Consequently, they do not properly

incorporate high levels of interest rates into their consumption smoothing decision and

underestimate their true cost of borrowing.

2.1.2.2.3 Impatience

Other studies have documented that people are overly impatient when it comes to

deciding between small benefits in the present and large benefits in the future.153 They

exhibit preferences which are not consistent in a dynamic way: they have relatively high

discount rates over short horizons, whereas the discount rates are rather low over long

horizons.154 Those people who prefer present gratification take out more credit than they

149 Eg. Brown and Taylor (1988).

150 See Weinstein, 1980.

151 See Yang, Markoczy and Qi (2003). Ausubel (1991) even argues that this behaviour of a subgroup of

consumers provides a rationale for interest rate stickiness in credit card markets. Calem and Mester (1995)

provide empirical evidence on this conjecture for the US market.

152 This behaviour is not limited to private households: Landier and Thesmar (2009) document that overly

optimistic entrepreneurs are more likely to take short-term loans (instead of long-term loans) than realistic

entrepreneurs.

153 See Fehr (2002) for a discussion of this phenomenon.

154 This phenomenon is called Hyperbolic Discounting, See Laibson (1997) for a standard reference.

150 iff/ZEW – Final Report on interest rate restrictions in the EU

rationally should. As a consequence, they face higher costs of their consumption as

expected.155 Some forms of credit take advantage of this psychological bias, encouraging

this behaviour:156 Examples for this are all kinds of “buy now, pay later” schemes, such as

low upfront interest rates and increasing interest rates at a later stage of the loan cycle,

a scheme which is also known as “teaser rates”. Also, offers of revolving credit can

encourage consumption which would otherwise not have occurred.157

2.1.2.2.4 Discounting and the role of interest rates

In Chapter 2.1.2.1.2 we discussed the effects of the level of interest rates on the

consumption smoothing decision. Like most classical theories, it imposes that consumers

correctly perceive the decline in future consumption which results from the interest

payments on the loan. Recent evidence, however, reveals that this assumption generally

does not hold: the effects of interest rates are difficult to evaluate as people tend to

underestimate exponential growth.158 As a consequence of this cognitive bias, people tend

to underestimate how quickly the interest rates on an outstanding balance compound.

This leads to an underestimation of the future value of a given present value.159 It has to

be noted that the extent of this bias differs from one person to the next. There is

empirical evidence from the United States underlining the relevance of the exponential

growth perception bias for household decisions on credit finance.160

As a consequence from these effects, consumers may not be able to anticipate the exact

cost that will be charged over the years for early consumption based on borrowing.

Hence, consumers do not fully take into account how much their future consumption will

be reduced when they take out a loan to finance current consumption. Due to the

exponential growth of interest rates, this bias has particularly severe consequences when

interest rates are high:161 when, for instance, short-term loans with high interest rates

are rolled over several times, interest charges quickly accumulate in addition to the initial

principal. Given the exponential growth bias, consumers are unaware of this when

signing a loan contract for the first time. This may lead to a level of credit which is not

sustainable in the long run.

2.1.2.2.5 The desirability of credit access from a behavioural perspective

As discussed in the preceding paragraphs, psychological and cognitive reasons challenge

the view that people are generally able to anticipate future income fluctuations as well as

the cost of credit in a realistic way. These insights imply that borrowing decisions are not

always optimal (as described in Chapter 2.1.2.1), but based on misconceptions and result

in doing harm to some consumers in the long run. This view of imperfect consumer

155 Heidhues/Köszegi (2010) discuss the welfare implications of this behaviour and conclude that a ban of large

penalties of deferring credit could be welfare enhancing.

156 Vandone (2009, p. 20).

157 Kilborn (2005 p. 22) advances the opinion that the effect of hyperbolic discounting is “especially pronounced

in the average consumer credit transaction, particularly those involving revolving credit sources like credit

cards. Consumer credit facilitates, indeed, enhances consumers’ susceptibility to the bias toward present

consumption and against delayed gratification. Hyperbolic discounting explains (at least in part) why

consumers can only be expected to overvalue the benefits of ‘buying now’ and downplay the costs of ‘paying

later’.”

158 See Wagenaar and Sagaria (1975) as a fundamental reference.

159 See Stango and Zinman (2010a) for a detailed discussion. Lusardi and Mitchell (2007) provide independent

evidence of the poor ability to compound interest rates among older US citizens (50+).

160 Stango and Zinman (2010a) demonstrate that those people with severe biases have substantially larger

short-term-debt-to-income ratios. In contrast, long-term debt ratios are not significantly affected by the

severity of the bias.

161 See Eisenstein and Hoch (2005).

iff/ZEW – Final Report on interest rate restrictions in the EU 151

rationality is also backed by recent contributions underlining a low level of understanding

in personal financial issues (financial literacy).162

The effect of these misconceptions is most adverse for customers with low income and

wealth and little opportunity of improving their fortune.163 These vulnerable customers

may not necessarily benefit from improved credit access. Rather, it may be advisable to

restrict credit access for those groups for which irresponsible borrowing would bear

particularly damaging (private as well as social) consequences, such as (irreversible)

over-indebtedness. As interest rates restrictions rule out those credit options which would

otherwise be offered to vulnerable customers, they can be targeted at reducing credit

access. In this sense, interest rate restrictions could play a desirable role, as they reduce

the occurrence of credit finance among these particularly vulnerable consumers164 On the

other hand, this obviously implies that also those low-income borrowers who would not

borrow irresponsibly or in an irrational way (with respect to future income expectations)

will suffer from reduced credit access to credit and consumption smoothing.

2.1.3 Interest rate restrictions and capital allocation

2.1.3.1 Risk-adjusted interest rates and the effects of restrictions

Interest rates charged on consumer credit are essentially composed of four elements: the

market interest rate, operational costs, the bank’s profit margin and a compensation for

default risk.

Operational Costs Lenders have to borrow funds in capital markets to provide loans to

borrowers. They also have to cover their expenses for doing business (operational costs,

cost of equity). Interest rates and fees charged on loans thus have to cover fixed and

variable costs of providing credit services. Based on this aspect alone, interest rate

restrictions (IRR) may have two effects (operational cost considerations): first, since the

interest margin decreases, the lender may exploit other sources of income to earn her

required return, eg. fees. Second, if the overall volume of credit decreases due to IRR,

overheads have to be distributed to a lower number of total credit, increasing the

average interest rate.

Risk Adjustment This view, however, ignores important aspects of credit lending:

lenders also have to protect themselves against financial losses due to the customer

defaults. Wherever possible, lenders evaluate the riskiness of their customers based on

financial features, such as their income and assets. In particular, they consider the buffer

between income and essential expenses a borrower has. The larger the buffer, the less

likely borrowers will default on their loans. As low income customers only have a small

buffer, lenders face an increased risk of default. To compensate for higher risk, lowincome

borrowers are required to pay a higher interest rate compared to average

customers. From an economic point of view, this kind of price discrimination is efficient

and maximises credit supply in the market.165

162 Lusardi and Tuffano (2009) find a poor understanding in fundamental concepts of debt among US

households. Miles (2004) also documents central misunderstandings among borrowers in UK with respect to

debt interest rates. Atkinson et al. (2006) document that some UK customers have bought financial

products in an unwise manner.

163 See also Lusardi and Tuffano (2009).

164 In this vein, IGF/IGAS (2009) describe the exclusion of specific groups of the population as one of the

purposes of interest rate restrictions.

165 Smith (1970) points out that, in the case of inelastic demand, banks do not have to charge preferential

rates which take account of risk characteristics. Rather, they could charge a higher interest rate (potentially

up to an interest rate ceiling) for all loans. Smith demonstrates, however, that lenders do offer preferential

152 iff/ZEW – Final Report on interest rate restrictions in the EU

If legal regulation enforces interest rate restrictions, banks are only able to charge

interest rates which compensate up to a specific level of risk. All customers beyond this

risk level cannot be served with credit at the legal maximum of interest rates. In this

setting, interest rate restrictions may exclude borrowers with higher risk from being

provided with credit.

In this context, it has to be noted that the exclusion of high-risk borrowers from the

credit market alters the market conditions (market equilibrium considerations): interest

rates restrictions may affect the amount of credit available to less risky customers:166 in a

competitive market, prices of credit fall, and low-risk customers can borrow more money

not at higher, but even at lower prices. Figure 9 illustrates this idea: The left figure

shows that the quantity available for high-risk borrowers drops; those who still receive

credit are charged lower interest rates (the interest rate cap). The right-hand figure

shows that the amount supplied by banks to less-risky customers increases. As demand

is assumed to be constant, this group is able to borrow more at lower prices. It has to be

noted, however, that the total amount of credit decreases as the riskier customers are

rationed out.

Figure 9: Effect of IRR on high-risk (left) and low-risk (right) borrower

Whether average interest rates increase (as suggested due to the considerations of

operational costs) or decrease (as suggested by the market equilibrium considerations)

after the introduction of interest rate restrictions is ultimately an empirical issue. We

return to this aspect in Chapter 2.5.

Profit Margins Under perfect competition, lenders are not able to earn profits

higher than the costs they are facing for operating in their business and bearing the risk,

However, when markets are not fully competitive, profit margins can be remarkably

higher, as prices (in this case: interest rates) are not competed down in the interplay of

supply and demand.167 In this context, interest rate restrictions could redistribute banks’

rates and concludes that borrowers’ demand is sufficiently elastic to justify risk-adjusted pricing where

possible.

166 This is demonstrated by Blitz and Long (1965), who also discuss the case of a monopolistic bank: when the

bank is able to differentiate between customer risk types, interest rates restrictions are not suitable to force

the monopolist to reproduce the competitive outcome. The allocation of capital and the respective cost of

capital is ambiguous in this setting.

167 Note that even very high interest rates are not necessarily a sign for excess profit of a lender, which could

only arise in the case of lacking competition: Skiba/Tobacman (2007) demonstrate that even very high

iff/ZEW – Final Report on interest rate restrictions in the EU 153

profits to borrowers. However, it has been argued that this policy instrument has

difficulties reproducing the competitive outcome when a monopolistic bank is able to

differentiate between customer risk types.168 Rather than curbing monopoly rents,

antitrust policies have to ensure a sufficient level of competition in the markets

irrespective of the restrictions on interest rates.

Convergence to the cap level When demand for consumer credit is fairly inelastic in

markets with oligopolistic players, interest rate restrictions might serve as a focal point

for collusive lenders.169 Hence, even if the ceiling is initially not binding interest rates

would increase and converge to the level of the ceiling.

Note that this behaviour arises in the case of fixed rate caps or in the case of fluctuating

rate caps which are tied to a specific external reference, such as an interest rate decided

on by a central bank. In contrast, reference rates which are endogenously determined

(such as average observed interest rates of previous periods) respond to the pricesetting

behaviour of lenders: even if suppliers’ prices converge towards an existing rate

cap, the next period’s cap will be accordingly higher; as a consequence, there is no focal

point for collusion in the long run.

Tacit collusion distorts existing competition and results in higher average interest rates.

Whether this behaviour, which is possible from a theoretical perspective, occurs in reality

in the presence of interest rate restrictions, is again an empirical issue. (We come back

to this point in Chapter 2.5.)

Further note, that collusive behaviour is weaker the higher the cap is set above the initial

market interest rate hence it could cause only a small, if any, upwards fluctuation of the

average interest rate.

2.1.3.2 The effect of asymmetric information (supply side)

So far, we have assumed that lenders are actually able to discriminate between riskier

and less risky borrowers. However, this need not be the case. It might be that lenders

cannot (fully) observe the quality of borrowers (asymmetric information). In the extreme

case, they have to charge a single interest rate for all customers. If low-risk borrowers

are not willing to pay the market clearing interest rates, the lender would only attract

high-risk borrowers (adverse selection). To avoid this, it is rational for lenders to keep

the interest rate level low (to attract low-risk borrowers as well) and to lower the amount

of loans offered in the market (credit rationing).170 This is how an interest rate ceiling as

well as credit rationing may naturally arise from market forces.171 This implies that legal

APRs of US payday lenders may be reconciled with competitive markets, as these corporations face high per

loan and per store fixed costs.

168 See Blitz and Long (1965): the allocation of capital and the respective cost of capital is ambiguous in this

setting.

169 See Knittel and Stango (2003).

170 This idea of asymmetric information in credit markets and potentially resulting credit rationing has been

developed by Stiglitz and Weiss (1981).

171 Note that these considerations are not a rationale for why interest rate ceilings arise automatically by

market forces in all types of credit: for example, asymmetric information may lead to a floor in interest

rates for credit types which are closely linked to payment devices. Ausubel (1991) discusses this case for

credit cards, assuming that some low-risk borrowers falsely expect not to use their credit card as a source

of credit and are thus insensitve to high interest rates.

154 iff/ZEW – Final Report on interest rate restrictions in the EU

interest rate restrictions can only be considered effective when they are below marketbased

interest rate caps.172

This idea is illustrated in Figure 10. Note that, in the left figure, the supply curve is

“backward bending”. If the demand function looks like D2, some consumers are denied

credit access although they may be willing to pay higher interest rates than the market

rate. The right figure illustrates that an interest rate restriction R will only be effective if

it is below the equilibrium interest rate cap.

Figure 10: Credit markets under asymmetric information

Effective interest rate ceilings have two implications in this setting: they lower the

average interest rates and exclude more potential borrowers from being served with

credit. These effects are only jointly beneficial if the benefits from lowering interest rates

for those who are served outweigh the welfare loss due to reduction of credit contracts.

It has to be noted, however, that according to this model, interest rate restrictions do not

alter the credit rationing behaviour of lenders as such: if interest rate restrictions are

ineffective, the market-based interest rate cap remains; if they are effective, credit

rationing even occurs in a more pronounced way.173

When lenders find ways to overcome information asymmetry, they are ultimately able to

supply more credit. Collaterals can be particularly important in this respect:174 banks can

offer credit contracts with as well as without collateral requirements. Low-risk borrowers

are more strongly inclined to choose a contract with lower interest rates and stricter

collateral requirements, whereas those who are more likely to fail choose the contract

without (or weak) collateral requirements instead. Contracts which are dissimilar enough

in terms of collateral requirements and interest rates enable lenders to set discriminate

prices and to ultimately increase the level of credit provided. In this context, it has to be

noted that interest rate restrictions may again result in lower total credit supply, as they

weaken the lenders’ ability to differentiate their products in a sufficient manner.

172 This is also pointed out by Villegas (1989). He also shows that, if the interest rate restriction is below the

market interest rate, funds are shifted to countries with rationed borrowers in countries without interest rate

restrictions. In this case, the conclusions made in Figure 5 do not hold any more.

173 Coco and Meza (2009) discuss these effects in detail and argue that moderate interest rate restrictions

below the market outcome are beneficial.

174 The role of collaterals as a separating device to overcome asymmetric information has been proposed by

Bester (1985).

iff/ZEW – Final Report on interest rate restrictions in the EU 155

2.1.3.3 The effect of asymmetric information (demand side)

In the preceding Chapter 2.1.3.2, we have assumed that borrowers are better informed

about their own riskiness than lenders. However, this need not be true in all segments of

consumer credit: Instead, it is perceivable that lenders have past experience within a

local region or among a group of potential borrowers, whereas the borrower is rather

inexperienced in financial affairs.175 In this case, it is possible that lenders approve credit

although they know that it is against the households’ interest (aggressive lending).

If lenders are too aggressive in a market segment, low enough interest rate restrictions

can have positive effects on consumers’ welfare: they constrain lenders’ aggressiveness

to a smaller fraction of the consumers, as would be the case in the absence of IRR. The

remaining borrowers would face lower interest rate payments. Note that this argument in

favour of interest rate restrictions implies that fewer households are served when legal

interest rate restrictions are effective. This illustrates why, in the context of aggressive

lending, it can be seen as socially desirable that less households encounter the problem

of being provided with credit access detrimental to their interest.

It has to be noted, though, that under certain circumstances aggressive lending does not

necessarily arise when lenders are better informed than borrowers. Under circumstances,

lenders may even act too conservatively and offer less credit than desirable.176 This

situation could again be worsened by IRR. Unlike in the case of aggressive lending, one

would not see merits in reducing the number of households served with credit, but would

rather find further credit supply advisable.

Hence, the desirability of interest rate restrictions does not ultimately depend on the

question of whether or not lenders may be better informed about default risk than

borrowers. Rather, it depends on whether these lenders are in a position to offer their

credit contracts in an aggressive way. Deciding about the usefulness of interest rate

restrictions thus requires deciding whether the amount of credit supplied in a market

segment is considered to be sufficient, non-sufficient or even exaggerated.

2.1.4 Resulting Hypotheses

Whether or not the theoretical considerations above are relevant for the realworld

consumer credit market is ultimately an empirical issue. To make the ideas of

Chapters 2.1.2 and 2.1.3 tractable, we formulate them in the form of several

hypotheses. This allows discussing specific findings of existing literature as well as of our

data analysis and the responses from the questionnaires in this study under the guidance

of theory.

The following hypotheses concern the discussion about credit access and credit supply in

Chapter 2.1.3. For some issues, they have different implications for average borrowers

as well as low-income borrowers.

The first hypothesis is a central argument which is common to theories assuming that

banks have the ability to differentiate between different customer types.

175 This paragraph follows Inderst (2009). He argues that the underlying assumptions are realistic when the

borrowers are households. One may expect to find different levels of this demand side information to be a

disadvantage for different customer groups within the household sector. Bond et al. (2009) make similar

arguments for mortgage markets and discuss the role of collaterals.

176 Inderst (2009) points out that the market structure determines whether an informed lender behaves in an

aggressive or conservative manner. If she has monopolistic power, aggressive behaviour is rational; if she

faces (uninformed) competition, conservative behaviour is more likely. Bond (2009) also concludes that

competition lowers aggressive lending in most cases.

156 iff/ZEW – Final Report on interest rate restrictions in the EU

H1: IRR reduce credit access, in particular for low-income borrowers.

As a consequence to the exclusion of some subgroups, the overall amount of credit might

drop:

H2: IRR lead to a decline in the volumes of consumer credit granted.

A related aspect is that, in the presence of interest rate restrictions, banks are not able

to offer certain types of products in the market. For example, the presence of interest

rate restrictions could imply that lenders are not able to cover the fixed costs of small

amount credits when they face limits to interest rates. We will therefore consider the

hypothesis

H2a: Without IRR, more product types exist in the market.

If H1 can be confirmed, this will have two potential implications for low-income

households: On the one hand, interest rate restrictions might prevent credit access for

high-risk borrowers, leading to a lower level of over-indebtedness (see H5). On the other

hand, consumers might try to find other, not regulated, sources of the desired credit

supply. This leads to the hypothesis

H3: IRR lead to credit from non-bank sources, such as paying bills late.

In the same vein, it is also sometimes argued that

H4: IRR lead to a substantial illegal market in lending.

In the debate on interest rate restrictions, it is also frequently argued that

H5: The lack of IRR leads to a high level of over-indebtedness,

which ultimately draws on behavioural assumptions as discussed in Chapter 2.1.2.2: in

this context, the motivation of this hypothesis would be, for example, that people may

underestimate the true cost of borrowing and are not able to sustain the high levels of

debt. Thus, we also consider a sub-hypothesis of H5:

H5a: The lack of IRR has particularly adverse effects on default rates/over-indebtedness in the

presence of negative shocks (eg. recessions) to the economy.

This could hold true due to overoptimistic behaviour of consumers: in good times, they

might underestimate the risk of future negative shocks in their credit decisions.

Some theories imply that average borrowers – unlike high-risk borrowers - face lower

interest rate charges when there are interest rate restrictions. We therefore investigate

whether

H6: The average consumer (or even more so: the low-risk consumer) would be granted cheaper

credit in the presence of IRR.

Note that, as discussed in Chapter 2.1.3.1, one might as well expect the opposite

conclusion from an increased need to cover operational costs, which would result in an

opposing hypothesis.

Some potential mechanics on the supply side of the market are rather obvious. Lenders

could try to circumvent the regulation, such that

H7: IRR lead to increased charges as providers will try to compensate the reduced interest

revenues by increased charges.

As different regulations across various countries cause additional costs for those who

consider entering a foreign market, we also hypothesise that

iff/ZEW – Final Report on interest rate restrictions in the EU 157

H8: IRR represent barriers to consumer credit market integration.

Decreased profit opportunities in the consumer credit segment could have the effect that

H9: IRR lead to lower competition in the consumer credit industry.

Finally, it has been argued that interest rate restrictions could serve as a focal point for

implicit collusion, which could imply that:

H10: IRR lead to a convergence of all consumer credit interest rates at the level of the interest rate

cap.

2.1.5 Credit to consumers vs. credit to businesses

The content of the hypotheses in Chapter 2.1.4 reveals, once again, that this study takes

a focus on consumer credit rather than credit to business (eg. SMEs, self-employed,

microenterprises).

In general, business loans and consumer credit are conceptually distinct products:

rational borrowers will only borrow if the expected return from their investment will

exceed the level of the cost of credit they need to pay.177 In contrast, consumers do not

expect a monetary gain from their investment, but borrow to increase their utility level

by smoothing consumption over time. Consumer credit is, unlike business credit,

ultimately a question of preference (given a budget constraint) which cannot be judged

on the basis of profitability considerations. Note that in this context, consumer credit is

also very dissimilar from micro investment credit in developing countries (note that there

is also micro credit for consumption purposes in those countries). As a consequence,

credit access to small investment credit in developing countries and to consumer credit in

the EU Member States are likely to have different macroeconomic effects. While the

former is found to promote economic development directly by creating entrepreneurship,

the latter may only affect the economy’s demand side through the consumption channel.

The theory presented in this section and the hypotheses on consumer credit take account

for this fact.

Nevertheless, other aspects described above also remain valid in the context of business

loans: entrepreneurs can also be subject to some behavioural biases (eg. over-optimism)

inducing over-borrowing. However, unlike in the case of private consumers, legislators

typically do not see the need to protect businessmen against their behavioural biases.

When interest rate restrictions are implemented, they are typically targeted at the

protection of consumers rather than SMEs, self-employed or micro enterprises178.

While interest rate restrictions primarily target credit to private households, one has to

acknowledge that consumer credit is also used to finance part of the businesses of selfemployed

persons. However, it is impossible to determine to what extent the reported

figures in the remainder of this report contain credit to businesses. Still, it is important to

note that, when talking about consumer credit in this study, this discussion has

implications for the level of financing of self-employed and small enterprises. In this

sense, our findings on the hypotheses also apply to small (micro) enterprises and the

self-employed.

177 See also Fernando (2006).

178 With the exception of Italy where the ceilings apply to both business loans and consumer loans, and France

where ceilings still apply to overdraft credit granted to businesses.

158 iff/ZEW – Final Report on interest rate restrictions in the EU

2.2 Existing studies on IRR

2.2.1 Outline

Numerous empirical studies have been devoted to determining the economic impacts of

interest rate restrictions. Ideally, two identical countries should be compared which only

differ solely in terms of their interest rate restrictions in order to find unambiguous

empirical support for any of the hypotheses discussed above. As this is not feasible in

reality, researchers try to find situations which come close to this requirement. However,

one has to be careful about the external validity of these findings, which might still be

idiosyncratic to the considered time or country. This chapter provides an overview of

these studies and their findings in the context of our hypotheses. It can be summed up

as follows:

Carefully observed natural experiments date back to the 19th century or earlier. They

suggest that interest rate restrictions reallocate capital in favour of socially superior

groups.

Comparisons of different states in the United States of America with different interest

rate regulations typically suggest that tighter interest rate restrictions lower credit access

for low-income customers as well as total consumer credit. Some studies also indicate

that small amounts of credits are less often available in the presence of IRR. However,

they tend to deny a relationship between interest rate restrictions and the interest rate

level for average (non-high-risk) customers.

Studies on payday loans, in particular in the USA, point at two further aspects: credit is

not per se beneficial (especially in the long run). Furthermore, when thinking about

banning a financial product, it is crucial to take into account the evasion strategies of

potential borrowers.

Unlike in the United States, any study which compares EU Member States will face severe

problems to identify the exact effects of interest rate restrictions, as observations are

also determined by a multitude of other economic and regulatory factors. Existing policyoriented

reports on countries of the European Union argue therefore either rather on

theoretical grounds, or provide empirical facts which cannot unambiguously attributed to

IRR.

While the studies from the US are the only ones which provide solid evidence on the

causal effects of IRR, they have the disadvantage of looking at low levels of interest rate

caps, while typical caps in the EU are higher. The results are thus only partly transferable

to European legislation.

2.2.2 Natural experiments

There are only few natural experiments from situations in which effective interest rate

restrictions have been introduced from one day to the next, and where data is available.

Where these natural experiments exist, they may give us some hints about the

mechanisms. Fortunately, the effect of interest rate restrictions have been fruitfully

investigated by economic historians analysing changes in legislation which had occurred

more than a century ago: Bodenhorn (2007) analyses the 7% interest rate ceiling in

mid-nineteenth century New York. He finds support for the hypothesis that, in the

presence of interest rate restrictions, the level of illegal lending is high (supports H4). He

further concludes that the average loan size increases while the average maturity

decreases. This study also questions the effectiveness of usury laws, as it suggests that

customers decide to pay an illegally charged usurious rate to a bank because they fear to

lose the benefits of the long-term relationship if they do not pay. However, it is obvious

that the latter point is strongly dependent on the legal system, the enforcement of

iff/ZEW – Final Report on interest rate restrictions in the EU 159

interest rate restrictions, as well as the effectiveness of fighting illegal lending by the

government. Going even further back in time, Temin/Voth (2007) investigate the effects

of a legal change in Britain in 1714. They find that a decrease of the interest rate

restriction from 6 to 5 percent lead to an increase in minimum and average loan size and

improved credit access for nobles (as in line with H6). Similarly, Benmelech/Moskowitz

(2010) find that the interest rate restrictions imposed in different US states in the 19th

century are in the interest of wealthy borrowers, since these can escape credit

constraints due to their reputation.

Although, obviously, these results have to be understood in the context of the social

structure of those days, they make an important point: restricting interest rates could

reallocate credit supply at the expense of socially and economically inferior households

(H1).

2.2.3 Comparison of different States within the USA

With respect to more recent regulations and social circumstances, academic studies

typically focus on specific markets in the USA. The advantage of this approach is that

they can compare legally and economically relatively similar entities (ie. the U.S. States)

in which the regulation of interest rates can differ from State to State. The observed

differences between the considered U.S. states can then -with some caution- be

identified as the effects of interest rate restrictions.179 However, it is problematic for the

purpose of this report that these studies look at interest rate caps which are relatively

low (eg. 12%). In Europe, the interest rate caps are frequently at higher rates.

Goudzwaard (1968) analyses consumption credit by 32 “State Small Loan Licensee

Reports” of lender operations in 1964. He finds a relationship between charged interest

rates and the riskiness of the lenders’ portfolios. He also shows that the lenders in the 11

states with the highest interest rate ceilings have loss rates which are 50 percent above

those of the providers in the states with the lowest interest rate ceilings. This finding

provides evidence for the conjecture that credit access for high-risk (low-income)

borrowers is higher when interest rate ceilings are higher (H1), but also to the

hypothesis that the existence of high cost credit increases credit default, which might be

interpreted as over-indebtedness (H5). Wolken/Navratil (1981) consider the introduction

of a 12 % interest rate ceiling for some parts of US Credit Union sector. They find lower

average interest rates (H6), but at the same time reduced credit supply (H2).

Villegas (1982) considers consumption credits taken out to finance a vehicle and

analyses the characteristics of people who are rationed out of the market. Based on more

than 1,000 observed loans in 1973/1974, he demonstrates that interest rate

restrictions are effective in this specific market. He shows that households with low

expenditure or who want to borrow small amounts face the highest interest rates, but are

also most likely to be rejected (H1). The probability of rejection decreases when interest

rate ceilings are set at higher levels. He rejects the idea that, in the presence of interest

rate ceilings, lenders charge lower interest to an identical loan compared to a situation

without interest rate restrictions. Rather, lower average interest rates in countries with

IRR arise from the fact that high-borrowers are excluded from the market. Based on data

on 250 auto loans from the 1983 Survey of Consumer Finances, Villegas (1989) finds

that low-income consumers tend to have higher consumer credit in states without IRR

(H1). He further demonstrates that middle-income households also hold less credit after

the introduction of IRR. As a consequence, he argues that interest rate ceilings lead to an

overall drop in available credit (H2). He rejects the hypothesis that distributional effects

of interest rate ceilings lower the interest rates paid by low-risk borrowers (H6). Villegas

179 It has to be noted that such a comparison across EU countries is problematic, as these are much more

heterogeneous than the US states.

160 iff/ZEW – Final Report on interest rate restrictions in the EU

(1987) analyses the same dataset and demonstrates that usury ceilings lower the

probability of low-income borrowers holding revolving credit (H1), whereas they do not

affect high-income borrowers. More recently, a stricter legislation on (subprime)

mortgage credit in North Carolina has been found to reduce credit access of high-risk

borrowers (H1), but not that of low-risk borrowers (Elliehausen 2004).

As can be seen, the experience from the USA over the last decades confirms that IRR

reduce credit access for low-income borrowers. (While these empirical findings are in line

with theory and can be assumed to hold for European countries as well, it remains

unclear how to evaluate the consequences of reduced credit access.) Typically these

studies find that credit access does not improve (at lower cost) for average or low-risk

borrowers when interest rate restrictions are present.

2.2.4 Payday loans: the discussion of the desirability of credit access

While most of the studies cited above are based on the notion that credit access is per se

beneficial, the considerations about consumer rationality in chapter 2.1.2.2 challenge this

view: those customers who tend to underestimate negative consequences of repaying a

loan could be better off when not taking out a credit in the first place. Attanasio et al.

(2008) find strong evidence that low-income customers are very insensitive to the level

of interest rates. It is therefore perceivable that they are willing to accept credit even at

conditions that turn out to be adverse in the long run.

A very recent strand of economic literature dealing with payday loans discusses the

pros and cons of increased credit supply in a more balanced way.180 It is still debated

whether payday loans are viewed as a tolerable high-cost form of emergency short-term

credit, or […] a highly addictive source of easy money that hooks the unwary consumer

into a perpetual cycle of debt.” (Stegman 2007). Zinman (2008) investigates the effect of

the interest rate cap on payday loans in Oregon.181 He finds that payday credit access is

strongly reduced (H1) in Oregon compared to Washington state. He reports that due to

the introduction of the interest rate cap, the share of respondents reporting difficulties

obtaining short-term credit rose by 17-21 percentage points. Zinman explicitly notes that

the welfare implications of this are unclear and not testable by the underlying data, as

they depend on the model of consumer choice. He emphasises that, as a consequence,

potential payday borrowers increasingly turn to suboptimal substitutes, such as paying

bills late or overdrafts on the checking account (H7). Zinman demonstrates that the

financial condition of borrowers suffers as a consequence of an interest rate cap. He

admits, however, that this finding is a short-term effect and does not necessarily reflect

the long-term consequences of denied credit access.

To fill in this gap, Karlan/Zinman (2009) identify a positive role of increased consumer

credit access (which implies that H1 is valid) to households in South Africa over a 27-

month horizon. They see this finding as evidence against the hypothesis of negative longrun

effects of a “debt trap” when credit is approved. It remains questionable, though,

whether these 27 months are sufficient to capture all long-term effects and to what

extent this experience from South Africa can be transferred to European countries.

Casting doubts on financial wisdom of consumers, Agarwal et al. (2009) document that

US payday borrowers tend to have unused liquidity by their credit card provider. Given

the availability of liquidity and the costs of payday lending, this contradicts a

conventional understanding of financial planning. The authors also describe that payday

180 Payday loans are typically defined as a small amount, short term credit (only for a few days until the

following payday). The lender receives, in return, a post-dated cheque. Payday loans are controversial due

to their high annualised interest rates charges.

181 Zinman (2009) reports that under this regulation, the maximum combination of finance charges and fees

that can be charged to Oregon borrowers is $10 per $100, with a minimum loan term of 31 days.

iff/ZEW – Final Report on interest rate restrictions in the EU 161

borrowers typically have experienced declining credit card liquidity during the six months

preceding their first payday loan. While these findings may be seen in the context of

consumer irrationality, they do not reveal anything about a scenario without payday

lending (which could arise when regulators introduce low interest rate caps). Some

studies point to the fact that interest rate restrictions targeting payday lending are only

effective for some financial products. Alternative sources of finance with inferior

conditions to low-income customers (eg. paying bills late) could still remain accessible for

low-income customers (H3). Some studies on payday loans in the US therefore discuss

whether or not it is desirable to ban payday lending given the alternatives in place. Some

argue that high cost credit, eg. payday loans, may be “well suited to the specific needs of

high risk borrowers” (Policis 2006a, p.3): this kind of credit can be taken for a very short

time span and on a small credit amount. This implies that borrowers have to substitute

them with less suitable (longer term, higher amount) credit if payday loans are banned.

However, evidence from other studies suggests that the actual needs of many payday

borrowers are not ultimately met with short-term credit: based on a survey in 2000/2001

among US payday loan customers, Ellihausen/Laurence (2001) report that 40 percent of

payday loan customers rolled over more than five loans in the preceding 12 months. 10

percent of the borrowers even rolled over an existing loan more than 14 times.182

Stegman/Faris (2003) point to a direct relationship between industry revenues and the

incidence of repeat borrowing of payday loan borrowers, which they quantify based on a

sample of suppliers from North Carolina in 1999/2000. They conclude that legal action

should not aim at a prohibition of payday loans, but limit the rollover of payday loan

debt.

While these contributions highlight the circumstances and implications of a specific US

product, they underline two aspects: credit is not per se beneficial, and, when thinking

about banning a financial product, it is crucial to take into account the evasion strategies

of potential borrowers.

2.2.5 Reports about IRR in EU Member States

As studies on the US markets can not necessarily be transferred to the European

situation, some policy-oriented reports have taken on the task to deliver empirical

evidence from European countries in that issue. However, a challenge they face is that

European countries are more heterogeneous than US states, which complicates the

identification of causal effects compared to the settings in most of the studies cited

above. A study by Policis (2004) on behalf of the Department of Trade and Industry

(DTI) in the United Kingdom aimed at providing an evidence-based analysis of regulatory

conditions and its effects for the UK, USA, France and Germany. Policis (2004)

presupposes that the demand for consumer credit is equally prevalent in all countries,

irrespective of the regulatory or cultural context”. As a consequence of this assumption,

Policis (2004) interprets low level of credit as an indication for low levels of credit

supply. It points out that low income borrowers prefer “readily accessible cash” without

delivering collaterals and seek low transaction costs and low efforts to obtain a loan. The

presence of IRR is described as reducing the diversity of credit products available to lowincome

borrowers (H2a). In particular, interest rate ceilings are seen to cause the lack of

a subprime credit market in Germany and France: according to the study, IRR in

Germany are used as a policy instrument distorting risk-adjusted pricing by lenders with

the purpose of excluding higher-risk groups from credit (H1). Policis (2004)

demonstrates that interest rate restrictions are effective in the sense that providers of

loan products for low-income households withdraw from countries with restrictions on

interest rates. Referring to their initial assumption that credit demand is universal in all

182 Similar behaviour is also reported for Florida and Oklahoma payday borrowers by the figures in Veritec

Solutions (2005a) and (2009), respectively.

162 iff/ZEW – Final Report on interest rate restrictions in the EU

countries, Policis (2004) concludes that reduced credit access leads to an increased

usage of “second-best” options, such as paying bills late (H3) and illegal lending (H4).183

Interestingly, Policis (2004) bases its assessment of credit demand on the lack of

opportunity of households to spend a larger amount without borrowing. Policis (2004)

does not, however, take into account that borrowing in the context of optimal

consumption smoothing not only implies a shortage of funds (“need” for credit),184 but

also the capability of paying off the debts later: if someone is not able to save €500 for a

major expenditure, it is also very likely that he or she will be in arrears on a loan for this

expenditure. We will return to this argumentation by Policis (2004) when we discuss H4

in Chapter 2.5.5 on page 269.

In a different study, Policis (2006a) attributes different macroeconomic developments in

France, Germany and the UK to different regulatory conditions in consumer credit

markets. It states that the German regulation hinders providers from lending to lowincome

households, while the French legislation can be circumvented. Policis (2006a)

hypothesises that the tough regulation in Germany leads to a lower level of consumer

credit (H2), which is ultimately interpreted a cause of its lower economic growth.185

Contrary to the relationship discussed in our theoretical part (H5), Policis (2006a)

hypothesises that a lack of suitable credit supply for low-income borrowers increase the

likelihood of over-indebtedness, as alternative sources of credit (including from illegal

sources) are more damaging to consumers. We will provide more evidence on the latter

point in Chapter 2.5.6.

More recently, the Office of Fair Trading (OFT 2010) has issued a report on high cost

credit in the UK. It summarises that high cost credit is primarily demanded by lowerthan-

average levels of income and people with poor credit history. It concludes that high

cost credit markets function well in the sense that they meet the demands of their

clients, but that there are low levels of competition in those markets. As the lack of price

competition in some of these markets is found to lead to excessively high prices, the

report discusses several potential policy measures to bring prices down. It concludes that

price controls (ie. interest rate restrictions) are not appropriate: the report hypothesises

that the suppliers would respond to price controls by restricting the “type and risk of

consumers they are willing to supply” (H1). It also argues that “suppliers could cease

offering a particular product” (H2a) and that suppliers could try to circumvent specific

price controls by imposing unregulated fees (H7) OFT (2010) describes recipients of high

cost credit to have limited options and to need the money for non-discretionary

expenditure. It also acknowledges that some stakeholders question whether some

consumers groups should have access to credit, but see this point outside the scope of

their report.

Bialowolski (2009) discusses the introduction of interest rate restrictions in Poland in

2006. He points out that interest rate ceilings lead to decreasing credit access (H1) and

argues that this reduces welfare: assuming that current interest rates reflect equilibrium

rates, he estimates a credit demand curve from an overlapping-generations model as

well as a credit supply curve from survey questions to suppliers. For the latter, he finds

that a reduction of the interest rates due to a legal obligation by one percent leads to a

decrease in credit supply by 15 percent, which he estimates to generate costs for both

households and financial institutions of 100 mio PLN and 500 mio PLN respectively.

Bialowolski (2009) demonstrates that the size of this effect depends on the elasticity of

183 Based on the identical data source, the same line of reasoning is also made in Policis (2006b).

184 The argumentation by Policis (2004) understands credit access to be desirable. Albeit not explicitly

mentioned, it draws on the idea of optimal consumption smoothing in the neo-classical sense, see Section

2.1.2.1.1.

185 See Chapter 2.5.2 for a detailed discussion of this point.

iff/ZEW – Final Report on interest rate restrictions in the EU 163

demand. It should be noted, however, that like any quantification of welfare effects,

Bialowolski (2009) relies on specific assumptions about consumer and supplier behaviour

to estimate the demand and supply curve: for example, the exact results also depend on

the calibration of the underlying macroeconomic model. The need for such assumptions

complicates the analysis, particularly in heterogeneous markets, where the estimation of

demand curves is even more cumbersome, if not “impossible” (Smith 1970). Bialowolski

(2009) ignores the heterogeneity with respect to the riskiness of loans as well as the

elasticity of demand for different consumer types, and assumes that there is a single

interest rate prevailing for all kinds of consumer credits. As discussed in Chapter 2.1.3,

however, it would be more appropriate to consider various interest rates depending on

the individual situation of the customer. The (static) welfare considerations of Bialowolski

(2009) are also ignorant of potential long-term effects of credit access, which materialise

when a customer defaults on his or her loan.

IOO (2009) investigates the Dutch market for Flitskrediet, which are loans of less than 1

month to maturity. The report points out that this market segments exists due to its

exemption from interest rate regulation and that the extension of existing interest rate

restrictions from other market segments would not allow this market type to survive.

This line of reasoning is in line with H2a.

IGF/IGAS (2009) investigates the functioning of the French interest rate restrictions and

suggests technical changes. Equivalent to the view in our study, the report points out

that the issue of interest rate restrictions primarily affects, if at all, consumer credit other

than mortgages. It compares the average level of effective interest rates across

European countries and concludes that the existence or non-existence of interest rate

regulations is not correlated with the interest rate level, as H6 would suggest. The study

points out that reducing the level of credit access to specific consumer types is one of the

purposes of interest rate restrictions, and discusses the trade-off between the

effectiveness of the measures with respect to the targeted group on the one hand and

excluding excessively many consumers from credit on the other. The study stresses that

the existing interest rate restrictions lead to substitutions among different types of credit

favouring revolving credit. It documents that low-income borrowers are more likely to

use revolving credit rather than instalment credit. The study also describes that, for small

revolving credit, it appears that typical interest rates cluster at the level slightly below

the usury ceiling, as H10 suggests.

2.2.6 Summary

The following table summarises the results of existing empirical studies with respect to

our hypothesis in Chapter 2.1.4. There appears to be a consensus that interest rate

restrictions lower credit access for low-income customers (H1), lower total credit (H2)

and lead to less product choice (H2b). It is more controversial whether credit costs

decrease for average customers (H6), and whether credit access is per se desirable or

leads to increased levels of over-indebtedness (H5).

Table 33: Overview of the hypotheses

Study Data Hypothesis Confirmed?

Bodenhorn (2007) New York 19th century H4 YES

H2b YES

Temin and Voth (2007) Britain 1714 H2b YES

164 iff/ZEW – Final Report on interest rate restrictions in the EU

H7 YES

Benmelech and Moskowitz

(2010)

US 19th century H7 YES

Goudzwaard (1968) US 1964 H1 YES

H5

Wolken and Navratil (1981) US Credit Unions 1970’s H7 YES

H1 YES

H2b YES

Villegas (1982) US 1973/1974 (Vehicle

finance)

H7 NO

H1 YES

H2 YES

Villegas (1989) US 1983 (Survey of

Consumer Finances)

H7 NO

Villegas (1987) US 1983 (Survey of H1 YES

Consumer Finances)

H7 NO

Elliehausen (2004) North Carolina (Mortgage

credit)

H1 YES

Zinman (2008) Oregon/ Washington

Payday loans

H1 YES

Karlan and Zinman (2009) South Africa H1 YES

Ellihausen and Laurence

(2001)

US payday loan customers H5 YES

iff/ZEW – Final Report on interest rate restrictions in the EU 165

Policis (2004) UK, USA, France, Germany H1 YES

H4 YES

Policis (2006a) France, Germany, UK H5 NO

H2 YES

Bialowolski (2009) Poland H1 YES

IOO (2009) Netherlands H2a YES

IGF/IGAS (2009) France H6 NO

H10 YES

166 iff/ZEW – Final Report on interest rate restrictions in the EU

2.3 Credit market overview

2.3.1 Outline

This part of the study provides facts and figures about the markets of credit to

households. We distinguish the markets for consumer credit (which do not include

mortgages) and the markets for consumer mortgage credit. This chapter first gives a

comprehensive overview of the credit markets in all 27 Member States of the European

Union over the period 1995-2009. For some aspects, it also considers specific subgroups,

such as the EU 25, the EU 15 and the New Member States.186

As the analysis in this section shows, we observe several trends during the period 1995-

2008.

Households’ credit is an important element of the EU economy. At the end of

2008 the estimated volume of total credit to households (outstanding) stood

at EUR6,781.88 billion or about 54.3% of EU GDP.187

The six countries with the largest volume of total lending to households in

Europe in 2008 were Germany, the UK, France, Italy, Spain and the

Netherlands. Together they accounted for nearly 80% of the overall volume of

total credit outstanding in EU 27 at the end of 2008.

The importance of the total lending to households across countries varies

significantly: from 128% of GDP in Denmark, to 18.8% of GDP in Slovakia.

The six countries with the largest volume of consumer credit in Europe in 2008

were the UK, Germany, France, Spain, Italy and Poland. These countries

represent 79% of the overall volume of consumer credit outstanding in EU 27

at the end of 2008. As is the case with total lending to households, the relative

importance of consumer credit shows significant variability across EU

countries: from 57% of GDP in EU 15 to 26% in the New Member States.

In 2008 for the first time since 1995, after more than a decade of robust

growth, consumer credit in the European Union slowed down and declined by

about -5%. Notwithstanding, consumer credit in the New Member States

demonstrated double-digit growth (17.9%).

Consumer credit plays an important role in financing current consumption of

European households. On average, consumer credit accounted for 15.2% of

final consumption expenditure across the EU countries at the end of 2008.

The level of indebtedness as measured by the ratio of consumer credit

outstanding relative to the disposable income was 15.4% on average across

the EU 27 countries.

186 EU 25 includes EU 27 countries except for Bulgaria and Romania. The EU 15 comprised the following 15

countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg,

Netherlands, Portugal, Spain, Sweden, and United Kingdom. New Member States include Bulgaria, Cyprus,

Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, Slovakia and Slovenia.

187 Data source: ECRI (2009), “Lending to Households in Europe. ECRI Statistical package 1995-2008”,

European Credit Research Institute (ECRI), Brussels. It should be pointed out that this estimate is likely to

be lower than the true size of the consumer credit market as not all institutions providing consumer credit

are covered by the available statistics.

iff/ZEW – Final Report on interest rate restrictions in the EU 167

The first look at the differences in consumer credit interest rates reveals

differences across the New Member States and the EU as a whole with the

nominal levels of interest rates being almost double the EU average figures for

most of the types of credit.

2.3.2 Data availability and definitions

The problem of availability and heterogeneity of existing data on consumer credit in the

EU is well known.188 The European Central Bank (ECB) uses monthly data provided by the

national central banks to calculate average figures for the Euro area and the EU. This

bears two problems: the figures are not necessarily harmonised across countries, and

they do not necessarily capture the entire credit markets.

As a result of the lack of harmonised data on consumer credit in the EU, in this study we

rely on data from a number of sources, including European Credit Research Institute

(ECRI), national associations of providers of consumer credit, European and National

Central Banks. In this chapter we rely heavily on the dataset prepared by ECRI (2009).

This dataset provides information on total credit to households (includes mortgage credit,

consumer credit and other credit) for the EU and a number of other developed and

developing economies during the period 1995-2008.189

All these data originate from national central bank statistics, who publish figures referring

to consumer credit granted by monetary financial institutions (MFIs) only. Unfortunately,

this fact implies that other financial intermediaries which service private sector in general

and households in particular, such as financial vehicle corporations, hire-purchase

companies, leasing and factoring companies, insurance, loans and securities dealers are

usually not included in the statistics provided by the official sources. A report prepared

for the European Commission – DG Health and Consumer Protection carried out an

extensive survey of the financial regulators, banking associations, creditors associations

and consumer organisations in the 27 Member States of the European Union.190 It

concluded that a majority of industry associations do not collect any statistics on

consumer credit provided by non-MFIs. Only three financial regulators out of eleven

which responded and five banking associations out of sixteen which responded to the

GHK survey collect some data. This implies a significant gap in the data on consumer

credit markets. As a consequence, the parts in this study relying on official data are

expected to underestimate the true size of the consumer credit markets in some market

segments as not all institutions providing consumer credit are covered by the available

statistics.

To fill in this void, the ECB has recently launched the Eurosystem Households Finance

and Consumption Network, which will conduct the Survey on Household Finance and

Consumption. Among other goals, the survey will provide harmonised EU-wide

household-level data on access and use of credit by European households (ECB, 2009).

As the data collection for the first wave of the survey only started in 2009, this source

cannot yet be incorporated into this study.

188 See, eg.Vandone (2009).

189 Please see Annex V for details of the main definitions used in relation to ECRI dataset.

190 GHK (2009), “Establishment of a Benchmark on the Economic Impact of the Consumer Credit” Directive on

the Functioning of the Internal Market in This Sector and on the Level of Economic Protection (European

Commission, DG Health and Consumer Protection, Brussels).

168 iff/ZEW – Final Report on interest rate restrictions in the EU

2.3.3 Overview of household credit in the EU 27

2.3.3.1 Total credit to households

As can be seen from Figure 11, the volume of credit outstanding varies substantially

among the EU 27 countries. This is natural as it reflects the size of the population and

domestic economies. In Germany, which had the highest volume of total credit in

absolute terms in 2008, it stood at the level of EUR 1,406 billion. The smallest volume of

total credit in absolute terms is registered for Malta at the level of EUR 3.2 billion. The six

countries with the highest volume of consumer credit in the EU 27 group are (in

descending order): Germany (EUR 1,406 billion), the UK (EUR 1,096 billion), France

(EUR 921 billion), Spain (EUR 880 billion), Italy (EUR 515 billion) and the Netherlands

(EUR 429 billion).

Figure 11: Total credit to households in EU 27 countries, 2008; EUR billion

Table 34 shows that the ranking of the countries with the largest volume of total

consumer credit has remained largely the same since 1998. Interestingly, however, the

dynamic of credit growth differs substantially: since 2003, for instance, the volume of

total credit to households in Italy has grown by an astounding 170% in nominal terms. In

2008 the six countries mentioned above (Germany, the UK, France, Italy, Spain and the

Netherlands) accounted for nearly 80% of the overall credit to households.

Table 34: Countries with the highest volume of total credit to households, EUR billion

Country 1998 Country 2003 Country 2008

Germany 1,179 Germany 1,426 Germany 1,406

France 417 UK 1,224 UK 1,096

Italy 205 France 583 France 921

Spain 202 Spain 409 Italy 880

Netherlands 190 Netherlands 343 Spain 515

iff/ZEW – Final Report on interest rate restrictions in the EU 169

Denmark 126 Italy 327 Netherlands 429

Total, EUR billions 2,319 4,312 5,247

Total EU 27, EUR

billions

2.583 5,121 6,782

Total, % of EU 27

total

89% 84% 77%

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations.

Note. The ranking for 1998 does not include the UK due to lack of data in the ECRI statistical package.

As Figure 12 and Figure 13 show, total credit to households grew rapidly during the last

two decades across the European Union. Average annual growth rates in the EU15 and

New Member States during 1996-2001 stood at 11% and 29% respectively. While growth

in EU15 countries slowed down during the next seven years to 8% per annum, it

accelerated to 39% per annum in the New Member States. In effect, in three New

Member States, Romania, Latvia and Lithuania, total credit grew by more than 50% per

annum (89%, 66% and 52% respectively). Countries with the lowest growth rates during

2002-2008 were Netherlands, UK and Germany (3.6%, 3.6% and -1.6% respectively).

Figure 12: Total credit to households in EU 15 countries, average annual real growth rates; in %

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations

170 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 13: Total credit to households in New Member States, average annual real growth rates; %

Figure 14 shows annual growth rates across EU 15, EU-25, EU 15 and New Member

States. The graph shows that the growth in the older Member States slowed down

substantially during 2007-2008, the years of financial turmoil. In contrast, growth in the

New Member States persisted. In 2008 EU 27, EU-25 and EU 15 all documented negative

growth of -5%. However, growth of total credit in the New Member States, albeit slowed

sown since 2006, was still at significant 17.9%.

Figure 14: Real growth rates of total credit to households in EU countries; 1998-2008; %

2.3.3.2 Total credit to households per capita

This ranking is different if one takes into account the size of population: looking at the

volume of total credit to households on per capita basis in Table 35, we find the six

countries in which household credit per capita was the highest in 2008 to be Denmark

iff/ZEW – Final Report on interest rate restrictions in the EU 171

(EUR 55), Luxembourg (EUR 36 thousand), Ireland (EUR 31 thousand), the Netherlands

(EUR 26 thousand), Cyprus (EUR 25 thousand) and Sweden (EUR 21 thousand).

Table 35: Countries with the highest total credit to households per capita in EU 27; 1000 EUR

Country 1998 Country 2003 Country 2008

Denmark 23.756 Denmark 33.722 Denmark 54.595

Luxembourg 15.652 Luxembourg 22.053 Luxembourg 36.402

Germany 14.366 Netherlands 21.165 Ireland 31.322

Netherlands 12.159 UK 20.546 Netherlands 25.664

Belgium 9.915 Germany 17.282 Cyprus 25.080

France 7.144 Ireland 17.221 Sweden 21.448

Median, EU 27 4.744 8.124 12.471

Standard Deviation,

EU 27

6.489 9.048 12.678

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations.

As Figure 15 demonstrates, the dispersion of total credit to households on per capita

basis is substantial among the EU 27 countries. The country with the highest credit per

capita in 2008, Denmark, shows a total credit per capita of EUR 55 thousand, while

Bulgaria, the country with the lowest level of consumer credit per capita shows EUR 1.22

thousand.

Figure 15: Total credit to households per capita in EU 27 countries, 2008, EUR billion

Source: ECRI (2009), “Lending to Households 1995-2008”.

Denmark and Luxembourg remained the countries with the highest per capital credit

during the period from 1998 to 2008. The Netherlands belongs to the top six countries of

both total credit as well as per capita credit. For Ireland, it is interesting to point out that

high growth rate of credit to households propelled it from number six to number three on

172 iff/ZEW – Final Report on interest rate restrictions in the EU

the list within the period of just five years. Table 31 details the growth rates of credit to

households per capita during the two sub-periods: 1998-2003 and 2003-2008.

2.3.3.3 Housing loans (mortgages)

At the end of 2008, the volume of housing loans in the Member States of the European

Union stood at EUR 4,653 billion. The six countries with the highest volume of housing

loans are (in descending order): UK (EUR 798 billion), Germany (EUR 787 billion), France

(EUR 687 billion), Spain (EUR 650 billion), the Netherlands (EUR 382 billion) and Italy

(EUR 263 billion) (Figure 16). Together, these six countries account for EUR 3,568 billion

or 77% of the overall volume of housing loans in the European Union. Notably, the ten

New Member States that joined the EU in May 2004 (the Czech Republic, Cyprus,

Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia, and Slovenia) account for

only EUR 126 billion or less than 3% of the overall volume of housing loans in the EU.

Figure 16: Housing loans in EU 27 countries, 2008; EUR billion

Source: ECRI (2009), “Lending to Households 1995-2008”.

As Table 36 demonstrates, the ranking of the top six countries has remained nearly

unchanged throughout the last decade and included the UK, Germany, France, the

Netherlands and Spain. However, the fraction of the top six countries in the overall

housing credit market had decreased from 92% in 1998 to (a still substantial) 77% a

decade later.

iff/ZEW – Final Report on interest rate restrictions in the EU 173

Table 36: Countries with the highest volume of housing loans; EUR billion

Country 1998 Country 2003 Country 2008

Germany 527.268 UK 901.620 UK 797.937

France 259.212 Germany 744.714 Germany 787.339

Netherlands 160.671 France 383.407 France 687.217

Spain 123.256 Netherlands 300.901 Spain 649.849

Italy 63.466 Spain 275.958 Netherlands 382.190

Belgium 60.575 Denmark 154.664 Italy 263.072

Total, EUR billions 1194.448 2761.264 3567.604

Total EU 27, EUR

billions

1,297.46

9

3,333.52

7

4,653.340

Total, % of EU 27

total

92% 83% 77%

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations. Note. The ranking for 1998 does

not include the UK due to lack of data in the ECRI statistical package.

In 2008, the largest housing debt per capita in the EU is found in Denmark (EUR 46

thousand), Luxembourg (EUR 31 thousand), Ireland (EUR 26 thousand), Netherlands

(EUR 23 thousand), Spain (EUR 14 thousand) and Sweden (EUR 14 thousand). As with

total credit, Ireland has quickly moved from position five (in the ranking in 2005) to

position three in just five years. During this time, the housing loans taken out per capita

in Ireland nearly doubled from almost EUR 14 thousands to EUR 26 thousand.

Figure 17: Housing loans in EU 27 countries per capita; 2008; 1000 EUR

Source: ECRI (2009), “Lending to Households 1995-2008”.

174 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 37: Countries with the highest housing loans per capita in EU 27; 1000 EUR

Country 1998 Country 2003 Country 2008

Netherlands 10.264 Denmark 28.727 Denmark 46.131

Luxembourg 10.148 Netherlands 18.546 Luxembourg 30.661

Germany 6.427 Luxembourg 17.323 Ireland 26.031

Belgium 5.931 UK 15.139 Netherlands 22.880

France 4.439 Ireland 13.867 Spain 14.245

Ireland 4.285 Sweden 9.334 Sweden 13.989

Median, EU 27 2.942 5.589 8.343

Standard Deviation,

EU 27

3.317 11.598 10.678

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations.

Like in case with the total credit to households, housing loans have grown substantially

during the last two decades (Figure 18-Figure 19). Average annual growth rates for EU15

and New Member States were 33% and 23% , respectively, during 1996-2001. During

2002 to 2008 average annual growth rates stood at 29% and 23% respectively.

Accordingly to the European Mortgage Federation factors such as increasing house prices,

a healthy labour market, increasing incomes and, tax-deductability of mortgage interest

payments in some countries have all contributed to enormous growth in mortgage

lending in recent years (European Mortgage Federation, 2009). The highest growth rates

are found, again, in the New Member States: housing loans in Bulgaria, Cyprus and the

Czech Republic have grown by more than 50% in each year between 2002 and 2008

(63%, 56% and 50% respectively). The lowest growth rates in this period were observed

in Slovenia, Slovakia and the UK (3.8%, 3.6% and -0.1%).

Figure 18: Housing loans in EU15, average annual real growth rates; %

Source: ECRI (2009), “Lending to Households 1995-2008”.

iff/ZEW – Final Report on interest rate restrictions in the EU 175

Figure 19: Total credit to households per capita in New Member States, average annual growth

rates; %

Source: ECRI (2009), “Lending to Households 1995-2008”.

2.3.3.4 Consumer credit

At the end of 2008, the total volume of outstanding consumer credit in countries of the

European Union stood at EUR 1,094 billion. The countries with the highest overall

volumes of consumer debt were: UK (EUR 245 billion), Germany (EUR 224 billion),

France (EUR 156 billion), Spain (EUR 102 billion), Italy (EUR 102 billion) and, somewhat

surprisingly, Poland (EUR 33 billion) It has to be noted that Poland first appeared in this

ranking in 2008.

Figure 20: Outstanding consumer credit in EU 27 countries; EUR billion

Source: ECRI (2009), “Lending to Households 1995-2008”.

176 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 38: Countries with the highest volume of consumer credit; EUR billion

Country 1998 Country 2003 Country 2008

Germany 216.637 UK 256.312 UK 245.217

UK 150.741 Germany 230.913 Germany 224.046

France 93.797 France 127.695 France 155.733

Spain 36.652 Spain 55.529 Spain 102.331

Italy 27.202 Italy 50.109 Italy 101.825

Austria 14.825 Austria 20.906 Poland 32.803

Total, EUR billions 539.854 741.465 861.954

Total EU 27, EUR

billions

590.804 854.021 1093.847

Total, % of EU 27

total

91% 87% 79%

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations.

When we look at consumer credit on per capita basis, the composition of the top

countries is rather different. This finding suggests that the occurrence of consumer credit

is highly diverse across the Member States of the European Union Figure 21).

Figure 21: Outstanding consumer credit in EU 27 countries; 2008; per capita

0

1

2

3

4

5

6

CY IE UK DK AT EL DE FR LU FI ES BE IT PT SE SI NL HU PL RO MT BG EE CZ LV LT SK

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Consumer credit per capita (left axis) Consumer credit per capita (right axis)

use left axis use right axis

Source: ECRI (2009), “Lending to Households 1995-2008”.

The countries with the highest consumer credit share per capita are Cyprus (EUR 5.6

thousand), Ireland (EUR 4.1 thousand), UK (EUR 4.0 thousand), Denmark (EUR 3.5

thousand), Austria (EUR 2.9 thousand) and Greece (EUR 2.8 thousand). Denmark

entered the list in 2003, while Cyprus and Greece were the “new-comers” in 2008. This

suggests a rapid growth of consumer credit in these three countries (Table 39).

iff/ZEW – Final Report on interest rate restrictions in the EU 177

Table 39: Countries with the highest consumer credit per capita in the EU 27; 1000 EUR

Country 1998 Country 2003 Country 2008

Germany 2.641 UK 4.304 Cyprus 5.581

UK 2.578 Ireland 2.847 Ireland 4.107

Luxembourg 1.949 Germany 2.798 UK 4.015

Austria 1.858 Austria 2.575 Denmark 3.462

France 1.606 Luxembourg 2.354 Austria 2.876

Ireland 1.361 Denmark 2.246 Greece 2.812

Median, EU 27 0.695 1.097 1.453

Standard Deviation, 27 0.853 1.150 1.318

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations.

The last two decades saw a rapid expansion of consumer credit in the European Union.

As is the case with total credit and housing loans, consumer credit grew more rapidly in

the New Member States than in older Member States (Figure 22-Figure 23). Average

annual real growth rates in Member States between 1996 and 2001 and 2002 2008 were

16% and 37% respectively. Figures for EU15 for the same two periods are 11% and 8%

respectively. Consumer credit grew the fastest in the Baltic States and Romania and

Hungary, with all these countries having annual growth rates above 40% during 2002-

2008. France, Germany and Austria showed the lowest growth rates during the same

period (4.1%, 0.7% and 0.1% respectively).

Figure 22: Consumer credit to households per capita in EU15, average annual growth rates; %

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations.

178 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 23: Consumer credit to households per capita in New Member States, average annual

growth rates; %

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations

2.3.3.5 The diffusion of credit to households

In previous parts we have described the volumes outstanding of credit to households and

its main trends in the last decade. These figures, however, say little about their

affordability. In a further step, we thus turn to the question about how important credit is

relative to the sizes of domestic economies. When looking at the ratios of respective

measures of credit to GDP, Figure 24 shows that during the last eight years the ratio of

total credit to GDP in the EU 27 has fluctuated between 47% and 57%. The ratio showed

a slight, yet distinct, upward trend until 2006 after which it slowed by less than 1% in

2007 and by about 2% in 2008. We do not observe abrupt changes in this indicator and

may say that during the last eight years, total credit to households accounted for about

half of the GDP in the European Union on average. The figure shows that with respect to

this variable, the situation in the EU is rather different from that of the USA, where the

proportion of the credit to households in GDP has been steadily increasing in the last

eight years and has grown by about 25%: from 69% in 2000 to 94% in 2008.

iff/ZEW – Final Report on interest rate restrictions in the EU 179

Figure 24: Development of total credit to households in % of GDP; EU 27 and USA; 2000-2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

However, the average figures for the whole of the EU conceal differences between the old

and new members: in particular, we observe that the importance of the credit relative to

GDP is still significantly smaller for the New Member States of the EU. Only after 2003

did the consumer credit reach more that 10% of GDP of these countries and at the end of

2008 this figure stood at about 26% (Figure 25).

Figure 25: Development of total credit to households in % of GDP for different EU country groups;

1995-2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

180 iff/ZEW – Final Report on interest rate restrictions in the EU

Credit to households plays the most significant role with respect to the overall economies

in the following of the EU countries (in descending order): Denmark (128%), Cyprus

(113%), Spain (80%), Portugal (80%), Ireland (75%), the Netherlands (73%) and the

UK (72%) (Figure 26).

Figure 26: Total credit to households outstanding, % of GDP; 2005-2008

Source: ECRI (2009), “Lending to Households 1995-2008”. Top seven countries with the highest level of

diffusion of total credit to households relatively to GDP are highlighted in red.

2.3.3.6 Role of credit for household income and consumption

At the end of 2008, consumer credit accounted for an average of 15.4% of the

disposable income among the EU 27 countries. Comparing this figure with that of the US

(where at the end of 2008 it stood at 24.4%), this level can be considered moderate.

However, it is noteworthy that we observe opposite trends in Europe and in the US after

2003. While it decreased slightly in the USA between 2003 and 2008 (from 25.8% to

24.4%), it grew slightly in Europe (from nearly 13% to 15.4%) (Figure 27). Again, while

the levels were still lower among the New Member States, those countries exhibited the

strongest growth since 2003.

Figure 27: Consumer credit in the EU 27 and the US as a % of Disposable Income; EU and USA; %

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

US EU 27 EU 15 New Member States

Source: ECRI (2009), “Lending to Households 1995-2008”.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

110%

120%

130%

140%

AT BE BG CY CZ DE DK EE EL ES FI FR HU IE IT LT LU LV MT NL PL PT RO SE SI SK UK

2005 2006 2007 2008

iff/ZEW – Final Report on interest rate restrictions in the EU 181

These figures indicate that consumer credit plays an important role in consumption

expenditure of European households: at the end of 2008, consumer credit as a

percentage of final consumption stood at an average of 15.2% across the 27 EU

countries (Figure 28). This figure is very close for the subgroups of EU-25 and EU 15

countries. It should be pointed out that for the New Member States this figure has

converged in the last thirteen years to the EU 27 level, as that figure rose from slightly

more than one percent in 1995 to 15% in 2008.

Figure 28: Consumer credit as a percentage of final consumption; Expenditure of households; %

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

16.00%

18.00%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

EU 27 EU 25 EU 15 New Member States

Source: ECRI (2009), “Lending to Households 1995-2008”.

2.3.3.7 Credit to households: distribution by category

Figure 29 shows that, compared to the US, both consumer credit and housing loans

represent lower fractions of credit to households in the EU: housing loans account for

about 69% in the EU 27, while these are nearly 78% in the USA. US consumers also rely

heavier on consumer credit than their European peers: the fraction of the consumer

credit in the total credit to households stood at 16% in the EU 27 and at 19% in the USA.

Figure 29: Distribution of total credit to households by MFIs in Euro area 16; EU 27 and US

14.82% 16.13% 19.35%

67.50% 68.61%

77.91%

17.68% 15.26%

2.75%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Euro Area 16 EU 27 US

Consumer Credit Housing Loans Other Loans

Source: ECRI (2009), “Lending to Households 1995-2008”.

182 iff/ZEW – Final Report on interest rate restrictions in the EU

Loans to households are also more important for the US economy than for the European

economy when they are measured as ratios relative to the corresponding GDP. Housing

loans account for about 73% of GDP in the USA and only about 37% in the EU 27.

However, even in the EU these numbers are far from being insignificant. Consumer credit

accounts for about 18% of the GDP in the USA and only about 9% of EU GDP.

Figure 30: Lending to households by credit type, % of GDP; EU 16; EU 27 and US

9.50%

36.27%

7.96% 8.28%

37.23%

8.75%

2.58%

73.28%

18.20%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Consumer Credit Housing Loans Other Loans

Euro Area 16 EU 27 US

Source: ECRI (2009), “Lending to Households 1995-2008”.

Table 40 highlights some of the trends in terms of development of total credit to

households and its two major components, housing loans and consumer credit, as a

percentage of GDP during the period 2000-2008. Several major trends should be pointed

out:

There is a substantial and monotonic increase in the importance of the total credit

relatively to the overall economy in all of the three considered EU country

groupings (EU 27, EU 15 and New Member States), and particularly so among the

New Member States, which saw more that five-fold increase in this indicator.

A similar trend is taking place with regard to housing loans, with the housing

loans to GDP ratio among Member States increasing by more than 21 times.

There is stabilisation in the importance of consumer credit among the earlier EU

members and a two-fold increase among the New Member States.

Table 40: Development of total credit to households as % of GDP in the EU 27; 2000-2008

Total Credit Housing Loans Consumer Credit

1998

EU 27 39.47% 20.36% 7.71%

EU 15 40.99% 21.11% 7.81%

New Member

States

4.69% 0.64% 4.23%

iff/ZEW – Final Report on interest rate restrictions in the EU 183

2003

EU 27 50.78% 33.23% 8.51%

EU 15 52.92% 34.55% 8.74%

New Member

States

9.54% 4.88% 3.63%

2008

EU 27 54.25% 37.23% 8.75%

EU 15 56.70% 39.24% 8.74%

New Member

States

25.67% 13.73% 8.88%

Source: ECRI (2009), “Lending to Households 1995-2008”, own calculations.

Figure 31 highlights the distribution of total credit across the individual EU 27 countries in

2008. In that year, countries with the highest proportion of housing credit were: the

Netherlands (89%), Denmark (85%), Luxembourg (84%), Ireland (83%), Estonia (81%)

and Latvia (79%). Countries with the highest proportion of consumer credit were:

Romania (74%), Bulgaria (51%), Hungary (44%), Poland (37%), Slovenia (37%) and

Greece (31%). Finally, Cyprus (33%), Italy (29%), Germany (28%), Sweden (28%),

Austria (24%), and Malta (20%) show the highest proportions of the other loans to

households.

Figure 31: Distribution of lending to households; 2008

0%

20%

40%

60%

80%

100%

AT

BE

BG

CY

CZ

DE

DK

EE

EL

ES

FI

FR

HU

IE

IT

LT

LU

LV

MT

NL

PL

PT

RO

SE

SI

SK

UK

HR

TR

CH

IS

AUS

CAN

JP

US

Consumer credit Housing loans Other loans

Source: ECRI (2009), “Lending to Households 1995-2008”.

2.3.4 Interest rates on credit to households in the EU

Changes in interest rates affect the cost of capital and thus influence investment and

saving decisions by households. The interest rates charged by financial institutions, in

combination with the volumes of lending and borrowing, may help to shed light on

184 iff/ZEW – Final Report on interest rate restrictions in the EU

structural developments in the consumer credit markets and can provide important

information for the analysis of stability and integration in this sector.

2.3.4.1 Definitions

In this section we provide a snap-shot of interest rates on consumer credit products

charged by monetary financial institutions (MFIs). The statistical requirements for

interest rate statistics produced by MFIs are described in Regulation (EC) No. 63/2002 of

the ECB of 20 December 2001 (ECB, 2001, “MIR Regulation”). This regulation defines the

statistical standards according to which MFI interest rates should be collected and

produced.191

The interest rates data presented in this section are collected from the National Central

Banks (NCBs) and the European Central Bank (ECB). The interest rates data available

from these sources refer to new business and outstanding amounts. The new business is

defined as all financial contracts, terms and conditions that specify for the first time the

interest rate of the loan, and all new negotiations of existing loans. Outstanding loans

cover all loans used and not yet repaid by customers in all the periods up to and

including the reporting date, although excluding bad loans and loans for debt

restructuring at rates below market conditions (ECB, 2003b). In the present section we

provide the data on the interest rates in reference to the new business.

Depending on the choice of NCBs, the interest rates are either annualised agreed rates

(AAR) or narrowly defined effective rates (NDER) and compiled either as a snapshot of

end-month observations or as implicit rates referring to the average of the month. The

interest rate that is individually agreed between the reporting agent and the household or

non-financial corporation for a deposit or loan, converted to an annual basis and quoted

in percentages per annum. The AAR covers all interest payments on deposits and loans,

but no other charges that may apply. The NDER refers to an annual basis and is defined

as the interest rate that equalises the present value of all commitments other than

charges (deposits or loans, payments or repayments, interest payments), future or

existing, agreed by the credit institution or other institutions and the household or nonfinancial

corporation. The NDER is equivalent to the interest rate component of the APRC,

ie. it does not take into account the component of other charges (ECB, 2003a).192

Interest rate statistics for the euro-area refers to instrument categories rather than to

individual products. These categories (in case of new business statistics which is relevant

for this section) are presented in Table 37 below.

Table 41: Lending to households: MFI interest rates (AAR/NDER) on new business

Type of Instrument Initial rate fixation

Bank overdraft

Lending for consumption - Variable rate and up to 1 year initial rate

fixation

- Over 1 and up to 5 years initial rate fixation

- Over 5 years initial rate fixation

191 This regulation is explained in detail in the “Manual on MFI interest rate statistics” (ECB, 2003a), which

clarifies and illustrates the statistical requirements.

192 As the data coverage differs substantially across AAR/NDER and APRC, we provide data on the former two

types of interest rate.

iff/ZEW – Final Report on interest rate restrictions in the EU 185

Lending for house purchases - Variable rate and up to 1 year initial rate

fixation

- Over 1 and up to 5 years initial rate fixation

- Over 5 and up to 10 years initial rate fixation

- Over 10 years initial rate fixation

Lending for other purposes - Variable rate and up to 1 year initial rate

fixation

- Over 1 and up to 5 years initial rate fixation

- Over 5 years initial rate fixation

Source: ECB (2003a).

2.3.4.2 Overview

Table 42 provides a snapshot of the interest rates charged on the various types of

housing loans and consumer credit in the 27 countries of the European Union. It should

be pointed out that the data in the tables need to be treated with caution due to

differences in the levels of macroeconomic risks (such as default risk, currency risk,

inflation risk, economic growth risk) and cost of capital across the EU countries. For

example, both level of risks and cost of capital are usually higher in the New Member

States and some particular earlier members of the EU. This fact results in higher levels of

interest rates charged on credit products in these countries, including the consumer

credit products. However, it should be noted that the interest rates are not fully

comparable due to different rates of inflation in these countries. Bearing this reservation

in mind we present below selected descriptive statistics for the nominal interest rates in

these countries. The following tables also contain data on the level of long-term interest

rates and inflation rates for the EU 27 countries in order to illustrate the differences in

the cost of capital and inflation levels across the European countries.

2.3.4.2.1 Interest rates on credits for house purchases

As Table 42 shows, the median nominal interest rates in the EU 27 for housing loans of

variable (1-year initial fixation), 1-5 years initial fixation and over 5 years initial fixation

stood at 3.45%, 5.17% and 5.07% respectively. The corresponding figures for New

Member States were almost double these and were 7.59%, 10.43% and 9.23%.

Table 42: Interest rates on lending for house purchase across the EU 27; new business; December

2009; %

Lending for house purchase (by initial

period of fixation):

Country

Floating

rate up to 1

year

Over 1 and

up to 5

years

Over 5

years

Long-term

interest rates

(***)

Inflation

rate (****)

Austria 2.91 2.94 4.90 3.29 0.4

Belgium 2.92 3.61 0

186 iff/ZEW – Final Report on interest rate restrictions in the EU

Bulgaria* 13.24 10.67 9.93 6.61 2.5

Cyprus - - - 4.6 0.2

Czech Republic* 5.96 5.68 5.07 3.98 0.6

Germany 3.36 3.76 4.29 3.14 0.2

Denmark 3.53 1.1

Estonia* 7.59 16.00 - 0.2

Greece 3.08 4.60 4.06 5.49 1.3

Spain 7.17 3.81 -0.3

Finland 1.92 3.47 4.18 3.46 1.6

France 3.38 3.8 3.74 3.48 0.1

Hungary* 10.27 10.99 15.05 7.69 4

Ireland 2.61 2.68 4.88 -1.7

Italy 2.24 3.35 4.05 4.01 0.8

Lithuania 8.55 10.43 9.1 4.2

Luxembourg 2.03 3.8 0

Latvia* 13.19 7.18 - 13.75 3.3

Malta 3.52 4.41 1.8

Netherlands 3,84 4,87 5,26 3.44 1

Poland* 6.85 7.43 8.53 6.22 4

Portugal 2.22 3.91 -0.9

Romania* 12.97 11.60 6.65 8.66 5.6

Sweden 1.52 3.02 4.63 3.24 1.9

Slovenia 3.36 5.17 6.28 3.91 0.9

Slovakia 5.92 11.43 14.37 4.12 0.9

United Kingdom* 3.69 4.84 5.68 3.6 2.2

Median EU 27 3.45 5.17 5.07 3.95 0.90

St. Dev. 3.67 3.92 3.69 2.54 1.76

iff/ZEW – Final Report on interest rate restrictions in the EU 187

Median New Member

States

7.59 10.43 9.23 6.22 2.15

St. Dev. New Member

States

3.68 3.63 3.68 3.07 1.87

Sources: National central bank statistics, unless indicated otherwise. The reported rate is the Annual Agreed

Rate (AAR)/Narrowly Defined Effective Rate (NDER), unless indicated otherwise. Missing values indicate that

the data is not available. NMS stands for “New Member States”. * Interest rates for domestic-currencydenominated

loans. ** ECB Statistics: http://www.bundesbank.de/statistik/. *** ECB Statistics:

http://www.ecb.int/stats/money/. The rates are secondary market yields of government bonds with a

remaining maturity close to ten years. ****Source: ECB Statistics: http://epp.eurostat.ec.europa.eu/.

2.3.4.2.2 Interest rates on consumer credit for different maturities

Table 43 shows the median interest rates for consumer credit for variable rates/1-year

initial fixation, 1-5 years initial fixation and over 5 years initial fixation loans. The

corresponding interest rates in the EU were 7.54%, 8.28% and 7.69% respectively. The

corresponding figures for New Member States were almost double as high and were

13.85%, 13.94% and 14.09%.

Table 43: Interest rates consumer credit across the EU 27; new business; December 2009; %

Country Consumer credit

(by initial rate of fixation):

Consumer credit other

purposes

(by initial rate of fixation):

Longterm

interest

rates

(***)

Inflatio

n rate

(****)

Floating

rate up to

1 year

Over 1

and up

to 5

years

Over 5

years

Floating

rate up

to 1 year

Over 1

and up to

5 years

Over 5

years

Austria 4.26 4.44 3.74 3.29 0.4

Belgium 5.27 6.50 5.29 3.61 0.0

Bulgaria* 15.45 13.94 13.33 13.30 11.51 11.69 6.61 2.5

Cyprus - - - 4.60 0.2

Czech Republic* 15.05 13.58 14.09 6.69 5.97 5.36 3.98 0.6

Germany 6.38 4.83 7.57 3.14 0.2

Denmark 8.448 6.17 7.036 3.53 1.1

Estonia* 11.70 20.73 19.40 0.2

Greece 8.18 8.95 9.75 7.00 5.49 1.3

Spain 9.72 8.08 9.08 3.81 -0.3

Finland 3.04 4.76 4.73 3.46 1.6

France 6.91 6.15 5.74 3.48 0.1

188 iff/ZEW – Final Report on interest rate restrictions in the EU

Hungary* 17.77 26.15 30.57 16.50 10.67 7.69 4.0

Ireland 3.63 2.61 3.90 4.88 -1.7

Italy 9.85 8.28 6.96 4.01 4.91 5.28 4.01 0.8

Lithuania 13.85 17.65 8.99 11.11 9.10 4.2

Luxembourg 5.17 4.76 1.77 3.80 0.0

Latvia* 21.13 25.04 7.81 13.75 3.3

Malta 6.02 5.56 4.41 1.8

Netherlands 8.76 - - 3.44 1.0

Poland* 10.78 13.71 18.88 6.22 4.0

Portugal 5.53 12.17 6.08 3.91 -0.9

Romania* 17.21 18.28 14.53 14.08 27.40 10.58 8.66 5.6

Sweden 4.20 3.24 1.9

Slovenia 4.99 7.35 7.40 5.32 6.03 6.04 3.91 0.9

Slovakia 5.26 5.57 8.87 4.12 0.9

United Kingdom* 2.72 11.78 7.87 3.6 2.2

Median 7.54 8.28 7.69 6.77 10.67 6.04 3.95 0.9

St. Dev. 5.26 6.97 6.64 4.75 8.44 2.53 2.54 1.8

Median NMS 13.85 13.94 14.09 8.99 10.67 8.31 6.22 2.2

St. Dev. NMS 5.44 7.13 7.15 4.53 7.62 3.03 3.03 1.8

Source: National central bank statistics, unless indicated otherwise. The reported rate is the Annual Agreed

Rate (AAR)/Narrowly Defined Effective Rate (NDER), unless indicated otherwise. Missing values indicate that

the data is not available. NMS stands for “New Member States”. * Interest rates for domestic-currencydenominated

loans. ** Source: ECB Statistics: http://www.bundesbank.de/statistik/. *** Source: ECB

Statistics: http://www.ecb.int/stats/money/. The rates are secondary market yields of government bonds with

a remaining maturity close to ten years. ****Source: ECB Statistics: http://epp.eurostat.ec.europa.eu/

2.3.4.2.3 Interest rates on consumer credit by credit type

Table 44 shows that credit cards and overdrafts appear to be the most expensive type of

credit (in terms of interest charges), with the credit cards commanding 12.37% and bank

overdrafts commanding 10.51% median rates respectively. The corresponding interest

rates in the New Member States were again higher, with the median credit card rates

being as high as 16.07% and overdraft rates being 16.51% for these countries.

iff/ZEW – Final Report on interest rate restrictions in the EU 189

Table 44: Interest rates on overdrafts, credit cards and overrunning across the EU 27; new

business; December 2009; %

Country Overdrafts Credit

cards

Overrunning

the bank

account(**)

Long-term

interest rates

(***)

Inflation

rate

(****)

Austria 5.89 5.89 3.29 0.4

Belgium 9.81 6.95 3.61 0

Bulgaria* 6.61 2.5

Cyprus 7.25 4.6 0.2

Czech Republic* 3.98 0.6

Germany 10.38 3.14 0.2

Denmark 3.53 1.1

Estonia* 16.51 0.2

Greece 14.08 15.17 14.08 5.49 1.3

Spain 12.34 3.81 -0.3

Finland 7.83 7.83 3.46 1.6

France 10.28 10.28 3.48 0.1

Hungary* 27.52 7.69 4

Ireland 12.6 12.6 4.88 -1.7

Italy 6.6 4.01 0.8

Lithuania 17.02 9.1 4.2

Luxembourg 3.8 0

Latvia* 24.86 13.75 3.3

Malta 6.44 6.45 4.41 1.8

Netherlands 5.76 3.44 1

Poland* 11.59 16.07 6.22 4

Portugal 10.64 3.91 -0.9

Romania* 21.43 8.66 5.6

Sweden 3.72 3.24 1.9

190 iff/ZEW – Final Report on interest rate restrictions in the EU

Slovenia 8.64 8.64 3.91 0.9

Slovakia 14.35 4.12 0.9

United Kingdom* 7.74 17.76 3.6 2.2

Median EU 27 10.51 12.47 7.83 3.95 0.90

Standard

deviation

6.55 6.68 3.17 2.54 1.76

Median NMS 16.51 16.07 7.25 6.22 2.15

St. Dev. NMS 7.58 4.95 1.11 3.07 1.87

Source: National central bank statistics, unless indicated otherwise. The reported rate is the Annual Agreed

Rate (AAR)/Narrowly Defined Effective Rate (NDER), unless indicated otherwise. Missing values indicate that

the data is not available. NMS stands for “New Member States”. * Interest rates for domestic-currencydenominated

loans. ** ECB Statistics: http://www.bundesbank.de/statistik/. *** For convergence assessment

purposes, Source: ECB Statistics: http://www.ecb.int/stats/money/. The rates are secondary market yields of

government bonds with a remaining maturity close to ten years. ****Source: ECB Statistics:

http://epp.eurostat.ec.europa.eu/.

iff/ZEW – Final Report on interest rate restrictions in the EU 191

2.4 Case Studies

2.4.1 Introduction

The aim of this section is to provide a comprehensive overview and international

comparison of relevant aspects of the functioning of consumer credit markets in the six

selected European countries. It provides the necessary background for the discussion of

the hypotheses in the subsequent chapters of this study. We focus on the following six

countries which serve as case studies in this report: Germany, France, the Netherlands,

Poland, Sweden and the UK. The reasons behind this choice include diversity in terms of

economic characteristics, financial cultures, size and attributes of the consumer credit

markets, as well as the heterogeneity of the legal framework of interest rate restrictions

in these countries. It is this heterogeneity in the chosen countries which allows us to

draw conclusions with regard to the effect of interest rate regulation on consumer credit

markets and over-indebtedness.

The six countries selected represent more than half (54.3%) of the population of the

EU 27. Furthermore, they account for nearly two-third of the volume of total credit to

households and consumer credit to households in EU 27 countries at the end of 2008, as

can be seen from Table 45. Thus focusing on these countries allows us to perform a

comprehensive overview on the European consumer credit market by capturing an

important share of it in the first place. Heterogeneity among their national consumer

credit markets in all relevant aspects ensures further representativeness of the case

studies for the European consumer credit market. The six countries included in our study

differ considerably in market size and market structure: half of the selected countries –

the UK, Germany and France represent the countries with the largest national consumer

credit markets. They alone make up nearly sixty percent (57.1%) of the European

consumer credit market and account for more than half (50.5%) of the total credit to

households in EU 27 countries. The rest of the countries included in the study - Poland,

Sweden and the Netherlands make comparatively small contributions to the overall

volume of credit to households in Europe. As a typical representative of the New Member

States from CEE, Poland exhibits high growth of per capita credit to households, as

indicated in Chapter 2.3. The northern European countries the Netherlands and Sweden

typically exhibit large mortgage markets which make up almost ninety percent (89.2%)

and roughly seventy percent (65.2%), respectively, of the total volume of credit to

households in both countries, as indicated in Chapter 2.3.3.7. Its large mortgage market

makes the Netherlands the country with the fourth-highest rate of per capita housing

loans in Europe. In contrast, Germany and France are among the countries with the

lowest level of per capita housing loans, with exception of the New Member States.

192 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 45: Market size and structure of the consumer credit market in selected countries

Populatio

n

(millions

of people)

Total

Credit, EUR

billion

Total Credit

(% of

EU27)

Consumer

Credit,

EUR

billion

Consumer

Credit (%

of EU27)

Housing

Loans,

EUR

billion

Housing

loans (% of

EU27)

DE 82.12 1406.30 21 224.05 20 787.34 17

FR 62.28 921.17 14 155.73 14 687.22 15

NL 16.70 428.69 6 23.77 2 382.19 8

PL 38.10 88.90 1 32.80 3 46.37 1

SE 9.18 196.87 3 13.30 1 128.41 3

UK 61.07 1095.52 16 245.22 22 797.94 17

61 63.53 60.81

EU 27 495.92 6781.88 1093.85 4653.3

4

Source: ECRI (2009), “Lending to Households 1995-2008”.

Furthermore, the dimensions of diversity in the selected countries include different

demographic structures and characteristics of financial behaviour. The most populous

countries in Europe – Germany, France and United Kingdom are included in the survey

alongside with smaller countries.

As can be seen from Table 46, both Germany and France exhibit above-average

standards of living measured by GDP per capita (EUR 30.35 thousand and EUR 31.27

thousand in Germany and Frances respectively) and slightly more equally distributed

income and wealth (Gini index of 30 and 28 for Germany and France respectively) than

the EU 27 average. However, there are differences among both countries in terms of

their levels of indebtedness and over-indebtedness. In Germany the total credit to

households as a percentage of GDP lies slightly above average, a comparatively low

proportion of borrowers experience arrears in payments on mortgage or rent, utility bills

or hire purchases. Despite the notably lower total credit to GDP ratio (47%) France

experiences a notably higher proportion of borrowers experience arrears in payments as

compared to Germany.

The total credit to GDP ratio is higher in the northern European countries Netherlands,

Sweden and United Kingdom. Over-indebtedness measured by the percentage of people

experiencing arrears in payments is relatively low in all three countries compared to the

EU 27 average and compared to Germany and France. However, UK exhibits relatively

low standards of living indicated by comparatively low GDP per capita (EUR 29.67

thousands) and an at-risk-of-poverty ratio above average (19%) when compared to the

high GDP per capita in Netherlands (EUR 35.35 thousands) and Sweden (EUR 35.78

thousands) and their considerably low at-risk-of-poverty ratio (11% and 12%

respectively). Furthermore, the Netherlands and Sweden experience higher income

equalities (with a Gini index of 28 and 24, respectively) than the United Kingdom where

the income inequality (Gini index of 34) is above EU 27 average.

iff/ZEW – Final Report on interest rate restrictions in the EU 193

Different financial behaviour and demographic structure are also observed in CEE

Member States represented by the most populous country in this category – Poland.

Poland experiences significantly lower standards of living (GDP per capita of EUR 9.47

thousands) and a high at-risk-of-poverty ratio (17%). Consequently, despite the low

total-credit-to-GDP ratio (25%) a significant part of the population (11%) also

experiences arrears in payments.

Table 46: Indicators of demographic structure and financial behaviour in selected countries

GDP per

capita (000’s

EUR)

Total credit

(as % of GDP)

Gini

Index

At-risk of

poverty

ratio*

Proportion of population

facing arrears in

payments***

DE 30.35 56% 30 15 6%

FR 31.27 47% 28 13 10%

NL 35.35 73% 28 11 4%

PL 9.47 25% 32 17 11%

SE 35.78 60% 24 12 6%

UK 29.67 72% 34 19 5%

EU 27 25.21 54% 31 17 10%

Sources: ECRI (2009), “Lending to Households 1995-2008”, Eurostat, European Survey of Income and Living

Conditions (EU-SILC). Notes: * Measured as as a proportion of population whose income after social transfers

is below the poverty threshold. The annual national at-risk-of poverty threshold in the EU is set at 60% of the

national median income per equivalent adult (Eurostat, http://epp.eurostat.ec.europa.eu/). ** Measured in %

of total population experiencing heavy financial burden from housing costs; *** Measured by % of total

population facing arrears in payments on mortgage or rent, utility bills or hire purchase.

Figure 32 compares the selected countries with respect to their level of indebtedness and

proportion of the population living below the poverty threshold. As can be seen from the

figure, in most of the European countries less than one-fifth of the borrowers live below

the respective country’s poverty threshold. Thereby, the volume of total debt relative to

GDP varies among European countries from less than 20% to more than 120% (in

Luxembourg). Figure 32 further illustrates that the countries’ level of indebtedness is not

related to the share of the population being at-risk-of poverty. Furthermore, the figure

illustrates the heterogeneity of the case study countries with regard to both measures.

This heterogeneity allows for meaningful comparison of the impact of interest rate

restrictions while accounting for relevant economic aspects.

194 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 32: Heterogeneity among relevant economic factors

Finally, the selected countries have different consumer credit regulation and, in

particular, different levels and forms of IRR. Included in the study are Sweden and UK.

Both of these countries have neither general usury limits nor interest rate restrictions. In

contrast, in Germany and France regulations of consumer credit interest rates have been

in effect for a long time. In Germany a court-based jurisprudence limits the interest rate

charges on contractual interest for all consumer credits. Charging more than twice a

certain benchmark rate or a premium of more than 12 percentage points above the

benchmark rate (whichever is lower) is considered usurious. There is also a variable

ceiling on default rates which lies at 2.5 percentage points above a published base rate

for mortgage credits and 5 percentage points above the base rate for consumer credits.

In France the ceilings are fixed at 133% of the 12 respective reference rate, which are

differentiated by credit type and size. Despite their abovementioned similarities in

financial behaviour, the Netherlands and Sweden experience different level of regulation.

Whereas there are no IRR in Sweden, in the Netherlands a variable ceiling is set at 12

percentage points above a reference rate. Thereby a “legal interest rate” is implied which

reflects the maximum rate to be charged in case the debtors are in arrears. In Poland

interest rate caps were introduced in 2005. The ceilings are variable and are at the level

of four times the National Bank of Poland’s Lombard rate.

iff/ZEW – Final Report on interest rate restrictions in the EU 195

Figure 33: Consumer credit per GDP in a cross-country comparison

16.19%

4.58%

8.99%

8.00%

4.03%

10.76%

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

Vereinigtes

Königreich

Schweden Deutschland Frankreich United Niederlande Polen

Kingdom

Sweden Germany France Netherlands Poland

Source: ECRI (2009), “Lending to Households 1995-2008”.

As can be seen from Figure 33, there is no obvious correlation between the existence of

interest rate restrictions (red-coloured countries) and the aggregate volume of consumer

credit relative to the country’s GDP. Between both case study countries without existing

interest rate regulation the difference is most obvious as UK has the highest and Sweden

one of the lowest level of consumer credit indebtedness. That is why for the purposes of

gaining an insight on the impact of interest rate restrictions the following in-depth

analysis on the selected individual credit markets should be helpful.

2.4.2 Credit markets

2.4.2.1 Germany

At the end of 2009, the German credit institutions lent EUR 227.5 billion to consumers for

their consumption purposes.193 This represents an increase of 1.5% compared to the

previous year. In 2009 instalment credit grew for the first time since 2004. The volume

of instalment credit grew by 7.5% to EUR 142 billion. Revolving credit, which includes

overdrafts and credit cards, grew by 7.1% to EUR 85.4 billion.

193 Housing loans are excluded.

196 iff/ZEW – Final Report on interest rate restrictions in the EU

2.4.2.1.1 Overview

Figure 34: Consumer credit outstanding; Germany, EUR billion

189.487

198.783

204.309216.637

215.695

222.554

222.405

224.343

230.913

228.417

223.675

224.046

233.989

236.957

0.000

50.000

100.000

150.000

200.000

250.000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Between 1995 and 2008 consumer credit in Germany grew from EUR 189.5 billion to

EUR 224 billion (ECRI (2009), Figure 34), although the growth slowed down somewhat

during 2006-2008. In 2009 the German credit institutions lent to consumers EUR 227.5

billion for consumption purposes (Börsen-Zeitung, 2010), 1.4% more than in the

previous year. In 2008 consumer credit in Germany was almost 9% of GDP (Figure 35),

which is slightly above the European average of 8.75% (ECRI, 2009).

Figure 35: Consumer Credit as % of GDP, Germany

10.60%

10.67%

11.02%

10.72%

10.79%

10.52%

10.47%

10.67%

10.72%

10.43%

9.84%

9.23%

8.99%

10.25%

0%

2%

4%

6%

8%

10%

12%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Figure 36 shows nominal growth rates of consumer credit in Germany. The growth rates

vary from 6% in 1998 to -2.3% in 2006. Years 1999, 2001, and 2005-2007 saw negative

growth rates in the volume of consumer credit outstanding.

iff/ZEW – Final Report on interest rate restrictions in the EU 197

Figure 36: Growth of consumer credit, Germany; % p.a.

2.78%

6.03%

-0.43%

3.18%

-0.07%

0.87%

2.93%

2.62%

0.17%

-1.25%

-2.38% -2.08%

-3%

-2%

-1%

0%

1%

2%

3%

4%

5%

6%

7%

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Private consumption, which forms the basis for the consumer credit market, stood at

EUR 1,414.7 billion and grew by about 0.4%. This amounts to about 9.0% of GDP and

suggests significantly lower levels of consumer debt than in the USA and the UK where

this figure stood in 2008 at 18.2% and 16.2%, respectively. More than half of the

consumer credit is directed to the purchase of cars. In 2009, new and used cars

accounted for EUR 17.3 billion. In second place with 36% are cash loans in the form of

instalment or revolving credit intended for free use by the consumer. In third place with

9% are loans on furniture, home appliances, and consumer electronics. Other vehicles

such as mobile homes and motorcycles make up 2%.

2.4.2.1.2 Types and dynamics of consumer credit

Information in this section is drawn on the annual reports of Bankenfachverband (various

issues), unless indicated otherwise and relates to the segment of consumer credit market

covered by its participating institutions. The 58 credit banks of the Bankenfachverband

finance private consumption and commercial investments, above all, motor vehicles.194

The banks represented by Bankenfachverband account for about 48% of the overall

consumer instalment credit (Bankenfachverband, 2009).

Credit banks that are members of Bankenfachverband finance private consumption by

means of instalment credit (93%) and revolving credit (7%). While the volume of

instalment loans in 2009 grew by 14.4%, the volume of revolving credit declined by

19.0%. The number of new contracts has increased by 15.5% in 2009 to 9.1 million.

Contracts for instalment loans grew by 37.7%, driven largely by car-dealerships and

consumer-electronics stores. The number of revolving credit contracts declined by

16.1%.

According to the research conducted by the Bankenfachverband and GfK Financial Market

Research more than one in three households uses instalment loans, overdrafts

(Dispokredit) or other form of financing in order to acquire consumer goods such as cars,

furniture or household appliances. In September 2009, 38% of households took

advantage of at least one type of financing. The volume of consumer credit for purposes

such as furniture, electronic equipment, washing machines etc. constituted EUR 33.5

194 http://www.bfach.de.

198 iff/ZEW – Final Report on interest rate restrictions in the EU

billion and increased by 11.3% in comparison with the previous year. The point-of-sale

financing (Finanzierungen am Verkaufspunkt) constituted EUR 21.4 billion. This form of

consumer credit has shown the highest growth rate of 22.9% in comparison to the

previous year (Börsen-Zeitung, 2010).

Instalment credit. The most popular form of the consumer credit is instalment credit.

About 28% of consumers make use of an instalment credit to finance private

consumption.195 On average, German consumers pay monthly instalments of around

EUR 250 to repay their loans. Ten percent of consumers choose other forms of financing

such as the overdraft (Dispokredit), revolving credit (Rahmenkredit, eg. credit cards) or

leasing.

During the period between 2004 and 2008 instalment credit grew slowly (Figure 18). In

2009 instalment credit has grown for the first time since 2004. In 2009 the volume of

instalment credit grew by 7.5% to EUR 142 billion.

Figure 37: Instalment credit; Germany; EUR billion

93.094

98.428

103.141

106.201

105.667

108.649

110.664

114.338

118.584 129.208

130.087

130.580

129.264

132.085

0.000

20.000

40.000

60.000

80.000

100.000

120.000

140.000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Automotive credit. Nearly one in three private cars is financed entirely or partially by

credit. Accordingly to a survey conducted by GfK Financial Market Research more than

two-thirds of customers who purchase goods on credit would not be able to finance their

purchase otherwise. This suggests that about 20% of private cars would not be sold

without customers having access to credit. As car prices represent anything but

insignificant amounts (the average price of a new car is about EUR 21,000 and the

average price of a used car is about EUR 11,200), a growing number of car buyers

recourse to credit. Two-thirds of all consumers who plan to acquire a car until 2011 can

imagine purchase it on credit. Instalment credit is again the first choice of the financing

mode. 18% of consumers would use leasing or choose another form of third-party

financing. In the case of used cars instalment credit with a share of 84% is by far the

most important form of financing. The volume of automotive finance for acquisition of

private vehicles in 2009 stood at EUR 17.9 billion and grew by 19.2% in comparison with

the previous year (Börsen-Zeitung 2010).

195 The cost of a consumer good is divided into instalments that can be repaid monthly. The consumer can

secure credit either directly at a bank branch or online. Additionally, consumer can access credit in form of

financing deals from car dealers.

iff/ZEW – Final Report on interest rate restrictions in the EU 199

Point-of-Sale Financing. Point-of-sale (POS) financing represents an important segment

of consumer credit. Besides the financing vehicles, the POS transactions included goods

such as furniture and consumer electronics. The volume of POS transactions stood at

EUR 21.4 billion at the end of 2009 and grew by 22.9% in comparison to the previous

year. Two-thirds of the goods bought on credit at the point-of-sale are instalment credits,

the rest are cash loans from the banks. In 2009, the average amount of a cash loan was

EUR 8,350 (EUR 8,200 in 2008) with an average duration of 50 months (53 months in

2008).

Table 47: Consumer credit market; Germany; 2009

EUR billion Change to previous

year, %

New Business 33.5 11.3

Instalment Credit 31.3 14.4

Point-of-Sale Financing 21.4 22.9

Cash loans 9.9 0.6

Other Credit 2.2 -19

Revolving credit 1.8 -20.5

Credit cards 0.4 -20.2

New contracts (million) 9.1 15.5

Stock (at 31.12.2009) 73.9 11

Source: Bankenfachverband (2009). The data relates to the segment of consumer credit market covered by

institutions participating in Bankenfachverband.

As Table 47 shows, the crisis situation observed in the other credit market segments did

not appear to have significantly affected the instalment credit granted by the credit

banks participating in Bankenfachverband, which grew by 14.4% in 2009. Figures

presented in Table 43 reflect the broader trend observed in the German consumer credit

market. The instalment credit in Germany showed robust growth of 7.5% in 2009

(Börsen-Zeitung 2010).196

Figure 38 to Figure 40 below show information about the segment of consumer credit

market served by the credit banks which are members of the Bankenfachverband. The

data show the dynamics of this segment of the consumer credit market from 2006-2009

and a break-down into the instalment and revolving credit categories. As we can see from

Figure 38, the majority of consumer credit granted by banks takes the form of instalment

credit (93.4% of the overall new business volume at the end of 2009). Within instalment

credit, which at the end of 2009 stood at EUR 31.3 billion, EUR 21.4 billion or 68.3% is

accounted for by point-of-sale credit (Figure 39). The fraction of point-of-sale financing in

the overall volume of instalment consumer credit fluctuated between 72.4% (2006) and

64.4% (2008). Finally, the volume of other credit between 2006 and 2009 declined from

EUR 4.2 billion to EUR 2.2 billion (Figure 38). The other credit is dominated by the

revolving credit (Rahmenkredit), which in 2009 accounted for 81.8% of the overall other

credit.

196 Revolving credit includes overdrafts and credit cards.

200 iff/ZEW – Final Report on interest rate restrictions in the EU

It should be pointed out that the volume of new business for credit cards has been

declining steadily in the last four years. While in 2006 the credit card new business

volume stood at EUR 1.7 billion, in 2009 it declined to only EUR 0.4 billion.

Figure 38: Consumer credit in Germany; Dynamics of instalment credit vs. other credit; 2006-

2009; new business; EUR billion

29.7

25.4

26.6

31.3

4.2 4.4

2.8 2.2

0

5

10

15

20

25

30

35

2006 2007 2008 2009

Instalment Credit Other Credit

Source: Bankenfachverband Annual Reports, various issues, http://www.bfach.de.

iff/ZEW – Final Report on interest rate restrictions in the EU 201

Figure 39: Consumer credit: Dynamics of instalment credit; Germany; 2006-2009; new business

EUR billion

21.5

17.2 17.2

21.4

8.2 8.2

9.4

9.9

0

5

10

15

20

25

2006 2007 2008 2009

Point of Sale Financing Cash Loans

Source: Bankenfachverband Annual Reports, various issues, http://www.bfach.de.

Figure 40: Consumer credit: Dynamics of other credit; Germany; 2006-2009; new business,

EUR billion

2.5

2.7

2.3

1.8

1.7

1.8

0.5

0.4

0

0.5

1

1.5

2

2.5

3

2006 2007 2008 2009

Revolving Credit Credit Cards

Source: Bankenfachverband Annual Reports, various issues, http://www.bfach.de.

202 iff/ZEW – Final Report on interest rate restrictions in the EU

2.4.2.1.3 Pawn Broker Credit

Pawn broker credit in Germany is provided by around 200 entities, members of the

German Association of Pawn Brokers (ZDP, 2009). According to Dischinger and Mögel

(2004), in 2003 1.9 million contracts for a value of EUR 450 million were issued by the

230 pawn broking entities. The volume of pawn broking credit amounted to about only

0.2% of the total consumer credit in Germany in that year. The reported number of

customers in 2000 was 160,000 or about 2% of indebted households in that year

(Dischinger and Mögel, 2004).

Pawn broking credit has grown in popularity in the recent years due to a number of

factors, including quick and easy process of granting credit (only a personal ID or

passport is required); creditworthiness of the customer need not to be checked; values of

credit can range from quite small (EUR 50) to relatively large (EUR 10,000); the duration

of credit is flexible and is determined by the customer; the credit can be extended on the

wish of the customer upon payment of the due interest and fees; there are no penalties

for early repayment of credit.

After evaluation of the collateral by the pawn broker a credit that typically amounts to

about 80% of the resale value of the collateral is granted. In 2002 an average value of

the pawn broker credit was EUR 230 (Dischinger and Mögel, 2004). Interest rate

amounts to 1% per month. Additionally, a customer pays a fee for storage, insurance

and administration in accordance with the regulation. Dischinger and Mögel (2004) report

that in 2002 the interest and fees for an average credit of EUR 230 amounted to 36% per

annum.

2.4.2.1.4 “Kredite ohne SCHUFA-Auskunft”

“Kredite ohne SCHUFA(-Auskunft)” also known as “SCHUFA-frei” credit is unsecured

credit granted by non-bank credit providers without check of the borrower’ credit record

with SCHUFA.197,198 Consumers are often those, who were not able to obtain credit

from mainstream banks. Some credit providers in Switzerland focus specifically on clients

with already existing debt. High default risk associated with such credit is compensated

by higher costs. Since opportunity to charge extremely high level of interest is limited in

Germany due to legal provisions against usury interest rates, providers compensate for

their risks by charging additional fees, such as commission and administrative fees. Often

these fees are charged upfront, before the decision about granting credit is taken. In

case an agent visits customer at home, an additional fee is charged for this service.

Some companies resort to communicating the decision about granting of credit via

phone, using the high-cost numbers 0190/0900. This results in customer incurring

additional costs even if credit is not granted.199 While we do not have data on the

precise number of consumer using “SCHUFA-frei” credit, Der Spiegel (2007) reports that

the number of enquiries regarding granting a credit stands at 400,000 customers per

year.

2.4.2.1.5 Summary statistics

Summary statistics on the distribution of consumer credit in Germany by type are

presented in Table 48. The figures refer to the overall German consumer credit market.

However, provided figures in the table should be treated with caution, since in a number

197 SCHUFA stands for Schutzgemeinschaft für allgemeine Kreditsicherung. The company was founded in 1927

(www.schufa.de). Nowadays SCHUFA is one of the leading credit information agencies in Germany.

198 However, Grote (2007) states that despite credit providers advertise otherwise, credit checks with SCHUFA or other

assessment of consumer creditworthiness are made nevertheless.

199 Grote (2007), Der Spiegel (2007). In one case phone call cost a customer who was refused credit about EUR80.

iff/ZEW – Final Report on interest rate restrictions in the EU 203

of cases they are based on estimates, due to absence of official statistics. The aim of the

table is to provide information about the relative importance of the various types of

credit existing on German market.

Table 48: Distribution of consumer credit; Germany, 2008

Type of credit EUR

billions

% of Total

Credit

% of

Consumer

Credit

Total credit (including mortgages) 1406.30 100.00

Mortgages 787.30 55.99

Consumer credit, out of which… 224.05 22.15 100.00

...Installment credit, out of which… 132.09 13.06 58.95

… … Point-of-sale financing* 85.86 8.49 38.32

… Automotive credit** 15.02 1.48 6.70

...Debit balances on wage, salary and pension

accounts

17.06 1.69 7.61

…Other (residual)*** 74.90 7.41 33.43

…Revolving Credit**** 79.74 7.88 35.59

… ...Credit Cards (new business), 2009***** 0.40 0.03 0.13

Source: ECRI (2009) unless indicated otherwise.

Notes: Percentages for consumer credit do not add up to 100 due to difering sources of the data.

* Estimate based on Bankenfachverband (2009).

** Estimate based on data in Börsen-Zeitung (2010).

*** Estimate based on data in Bankenfachverband (2009).

**** Estimate based on Börsen-Zeitung (2010).

***** Estimate based on projected total credit to households for 2009.

2.4.2.2 France

2.4.2.2.1 Overview

The volume of consumer credit outstanding more than doubled during the last fourteen

years. It grew from slightly more than EUR 70 billion to EUR 156 billion in 2008 (Figure

41). In 2008 French consumer credit market was the third largest market in the EU in

terms of volume of consumer credit outstanding (see Table 6 above).

204 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 41: Consumer credit outstanding; France; EUR billion

76.057

83.756

93.797

103.586

112.634

117.477

120.343

127.695

133.482

141.322

148.060

155.555

155.733

70.052

0.000

20.000

40.000

60.000

80.000

100.000

120.000

140.000

160.000

180.000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

In relation to GDP volume of consumer loans grew from 5.9% of GDP to 8% of GDP

(Figure 41). Thus in 2008 consumer credit as a fraction of GDP in France was slightly

below the EU 27 average, which was 8.75% (ECRI, 2009).

Figure 42: Consumer credit as % of GDP; France

6.20%

6.61%

7.08%

7.58%

7.81%

7.85%

7.76%

8.00%

8.05%

8.20%

8.19%

8.22%

8.00%

5.86%

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Consumer credit grew rapidly in France between 1998 and 2000 (Figure 43). In 2001

growth slowed down sharply and continued showing lower rates until 2008 when it

virtually was at zero level. This is in line with the situation among the EU 15, which saw a

decline in consumer credit by about 5% in 2008.

iff/ZEW – Final Report on interest rate restrictions in the EU 205

Figure 43: Consumer credit growth rates; France; %

10.12% 11.99%

10.44%

8.73%

4.30%

2.44%

6.11%

4.53%

0.11%

5.87%

4.77%

5.06%

0%

2%

4%

6%

8%

10%

12%

14%

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

According to BIPE (2006), in 2004 outstanding consumer credit to households in France

accounted for around 18% of their total liabilities, which is expected to be close to the

European average. Between 1995 and 2000, there was an increase of the share of credit

consumption in total household liabilities of about 15% to 20%. Figure 44 shows the use

of consumer credit and bank overdrafts by French households. Between 1995 and 2004

the fraction of French households using consumer credit fluctuated between 27% and

33%. The fraction of consumers using bank overdrafts remained within a 20%-25%

range.

Figure 44: Evolution of the use of consumer credit and bank overdraft; France; % of total

households

0

5

10

15

20

25

30

35

40

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Consumer credits Bank overdrafts

Source: BIPE (2006).

206 iff/ZEW – Final Report on interest rate restrictions in the EU

2.4.2.2.2 Types and dynamics of consumer credit

As we can see from Figure 45, the share of personal loans is very important. In 2008

personal loans constituted 55% of the total volume of consumer credit. Personal loans

are followed by the credit accounts (20%) and instalment credit for financing of

purchases (13%). Leasing and advances on debit accounts remain much less important

forms of consumer credit in France (4% and 3% respectively). Estimates provided in

BIPE (2006) suggest that the average duration of consumer credit in France is between

18 and 24 months. The comparable estimate for the United Kingdom and Germany

are 12 months and over 5 years, respectively.

Figure 45: Consumer credit by type; France; 1995-2008; EUR billion

26.150

28.265

31.201

35.250

40.465

43.889

47.205

49.616

52.885

54.721

58.957

63.485

71.741

78.597

15.166

16.876

18.594

20.098

21.687

23.145

24.029

24.337

24.491

26.748

27.910

28.501

28.943

28.002

14.223

15.626

16.108

17.068

19.439

20.756

21.068

19.888

21.303

21.802

23.193

23.251

18.343

19.538

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

personal loans utilisation of opened permanent credit accounts

financing of purchases by installment credit advances on debit accounts

other (incl. differed payments on cash and debit cards) leasing and related

Source: ECRI (2009), “Lending to Households 1995-2008”.

Accordingly to BIPE (2006), despite the growth of household debt, the debt burden

(repayment of principal and interest relative to disposable income) during the past

decade has only moderately increased. The reason is that increasing repayments were

partially offset by lower interest rates (Table 49).

Table 49: Total debt burden of households; France

Type 1995 2004

Capital repayment 9.2 11.1

Housing loans 4.5 4.7

Consumer credit 4.7 6.4

Interest paid 4.2 3.2

Total debt burden 13.4 14.3

Source: BIPE (2006). Debt burden is measured as a repayment of principal and interest relative to disposable

income.

Most popular uses of consumer credit in France are car purchases, house equipment,

consumer bills and other expenses. At least 30% of purchases of new cars would be

financed via credit (this may be an underestimation since this figure does not include

iff/ZEW – Final Report on interest rate restrictions in the EU 207

personal loans used to finance new cars). The average length of credit on new cars would

be a little more than eighteen months, which is close to the average of all consumption

loans (BIPE, 2006).

2.4.2.2.3 Summary statistics

Summary statistics on the distribution by type of consumer credit in France is presented

in Table 50 which provides information about the relative importance of the various types

of credit existing on French market.

Table 50: Distribution of consumer credit; France, 2008

Type of credit EUR

billions

% of total

credit to

households

% of

consumer

credit

Total credit (including mortgages) 921.17 100.00

Mortgages 687.22 74.60

Consumer credit of which… 155.73 16.91 100.00

...financing of purchases by installment credit 18.34 1.99 11.78

…advances on debit accounts 5.99 0.65 3.84

…personal loans 78.60 8.53 50.47

...utilisation of opened permanent credit

accounts

28.00 3.04 17.98

...leasing and related 4.23 0.46 2.71

...other (incl. differed payments on cash and

debit cards)

6.57 0.71 4.22

Source: ECRI (2009).

Notes: Percentages for consumer credit do not add up to 100.

2.4.2.3 Netherlands

2.4.2.3.1 Overview

The volume of consumer credit outstanding in the Netherlands in 2008 stood at the level

of EUR 23.8 billion (Figure 46). This made Dutch consumer credit market the nineth

largest in the EU. In nominal terms, the market has grown from EUR 10.9 billion in 1997

at nearly 120%.

208 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 46: Consumer credit outstanding; Netherlands; EUR billion

12.188

12.758

13.728

13.805

18.509 20.157 23.204

24.317

25.065

23.440

23.772

10.864

0.000

5.000

10.000

15.000

20.000

25.000

30.000

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Loans to consumers in the Netherlands represented about 4% of the country’s GDP

(Figure 47). This is somewhat lower than the EU average of 8.75% in that year. It should

be born in mind that consumer credit in this country accounts for only 5.6% of the

overall lending to households. Loans to households are dominated by the housing loans

which account for 89% of the overall household borrowing (see Figure 31 above).

Accordingly, in 2008 Netherlands was the country with the highest proportion of housing

loans (89%) and the lowest proportion of consumer credit (5.6%) in total credit to

households in the EU.

Figure 47: Consumer credit; the Netherlands; % of GDP

3.36%

3.30%

3.28%

3.08%

3.98%

4.23%

4.13%

4.03%

4.64%

4.74% 4.72%

3.17%

0%

1%

2%

3%

4%

5%

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

iff/ZEW – Final Report on interest rate restrictions in the EU 209

Figure 48 illustrates growth of consumer credit in the Netherlands during 1999-2008. As

is the case of Germany and France, growth rates fluctuated during this period. Like in the

other two countries, growth has slowed down starting from 2005 and was negative in

2007 (-6.5%).

Figure 48: Consumer credit growth; the Netherlands; % of GDP

4.68%

7.60%

0.56%

34.07%

8.90% 15.12%

4.80%

3.08%

-6.48%

1.42%

-10%

-5%

0%

5%

10%

15%

20%

25%

30%

35%

40%

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

2.4.2.3.2 Flitskrediet

In 2007 a new form of consumer credit was introduced in the Netherlands, the so-called

Flitskrediet (Flash credit), which is represented by loans of less than one month

maturity.200 Due to its short maturity, Flitskrediet was until recently not legally covered

by any interest rate restrictions. A distinction of Flitskrediet is the short notice on which

money is lent: a new customer can receive credit in less than 24 hours, a known

customer even within 10 minutes.

Providers grant only 10% to 20% of all requests for loans. In 2007, the volume of

Flitskrediet granted is estimated at 25.000 contracts with a total sum of EUR 6.0 million.

This amounts to about 0.025% percent of the Dutch consumer credit market. The

average loan size is 230 Euro, wheras most contracts have a loan sum of 100 Euro. The

average maturity of the loan is 24 days.

According to the providers, customers are mostly employed persons with yearly or

permanent employment contracts, who face unexpected expenses such as traffic fines or

want to treat themselves to something special, who repay their debt after receiving their

next salary. In this respect, this loan type is most similar to payday loans in the UK

(Chapter 2.4.2.6) or SMS loans in Sweden (Chapter 2.4.2.5. According to branch

organisations, its high costs make Flitskrediet a down-market product for consumers who

have already depleted their other liquidity sources or have no other possibilities of credit

access.

Table 51 shows the cost structure for Flitskrediet loans.201 For example, the effective

annual interest rate of a loan of 400 Euro for a period of 30 days amounts to 1200%. In

200 The information in this subsection is taken from IOO (2009).

201 According to IOO (2009), this lower bound estimate, as some providers also charge eg. the costs of SMS

that are received from the credit provider.

210 iff/ZEW – Final Report on interest rate restrictions in the EU

case of arrears, the customer usually has to pay an average of 10 Euro per reminder and

all of the debt collection costs.

Until June 2009 five competitors entered the market. According to IOO (2009), providers

do not yet break even. One of the reasons is the high startup costs of the young

business, but mostly the high costs for employees´ salary, write-offs and advertising. At

the current demand for loans, the incurred fees per loan would have to be at 37% per

loaned Euro to break even. IOO (2009) points out that the extension of existing interest

rate restrictions to Flitskrediet would likely close down this market.202 The government

has announced that the maximum interest rates will apply also to flash credits.

Table 51: Cost structure of Flitskrediet

Fee in Euro Fee as percentage of loan

Maturity(days) 15 21 30 15 21 30

100 20 19 25 20 19 25

200 40 38 50 20 19 25

300 60 57 75 20 19 25

400 80 76 95 20 19 23.8

500 95 125 19 25

600 114 19

Loan, EUR

750 142.5 19

Source: IOO (2009).

2.4.2.3.3 Summary statistics

Summary statistics on the distribution of consumer credit in the Netherlands by type is

presented in Table 52. The aim of the table is to provide information about the relative

importance of the various types of credit existing on Dutch market.

Table 52: Distribution of consumer credit; the Netherlands, 2008

Type of credit EUR

billions

% of total

credit to

households

% of

consumer

credit

Total credit (including mortgages) 428.693 100.00

Mortgages 382.190 89.15

Consumer credit, out of which… 23.772 5.55

Flitskrediet 0.0004 0.001 0.002

Source: ECRI (2009), IOO (2009).

202 It points out that an interest rate cap of 18 % translates into a maximal fee of 1.37 euro for a 30-days loan

or 0.68 Euro for a 15-days loan, which can hardly cover the providers´ communication costs.

iff/ZEW – Final Report on interest rate restrictions in the EU 211

2.4.2.4 Poland

2.4.2.4.1 Overview

During the period from 1996 to 2008 volume of consumer credit in Poland increased sixfold

from EUR 5.5 billion to EUR 32.8 billion (Figure 49). This remarkable growth

propelled Poland to the sixth largest consumer credit market in the EU in 2008.

Figure 49: Consumer Credit Outstanding; Poland; EUR billion

7.542

8.735

11.680

15.544

18.689

12.036

13.118

16.518

20.225

28.487

32.803

10.490

5.464

0.000

5.000

10.000

15.000

20.000

25.000

30.000

35.000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Consumer credit has also been rapidly increasing in proportion to Poland’s GDP: from

about 4.7% in 1996 to about 10.8% in 2008.

Figure 50). The latter figure is above the EU average of 8.75%.

212 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 50: Consumer credit; Poland; % of GDP

5.68%

5.94% 6.80% 8.04%

8.38%

5.99%

7.31% 8.71%

10.76%

6.48% 5.80% 5.85%

4.66%

0%

2%

4%

6%

8%

10%

12%

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

With the exception of two years (2002 and 2003), annual growth of consumer credit in

Poland has been high, always above 15% with the exception of the aforementioned two

years when growth was negative (Figure 51). The growth rate in 2008 was well above

the EU 27 average and somewhat below the average of the New Member States, which

were -5% and 17.9%, respectively. Pruski and Zochowski (2006) cite increased

creditworthiness of individuals due to higher disposable income, upward adjustment of

income expectations and favourable interest rate environment as demand-side reasons

for the growth of credit to households in Poland. On the supply-side, easing of bank

lending policies due to increased competitive pressures, development of additional loan

distribution channels by banks and attractive profit margins contributed to the growth of

credit to households.

Figure 51: Consumer credit growth; Poland; % p.a.

20.23%

25.92%

25.05%

33.09%

15.81%

-12.85%

-35.60%

33.72%

22.44%

40.85%

15.15%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

iff/ZEW – Final Report on interest rate restrictions in the EU 213

2.4.2.4.2 Types and dynamics of consumer credit

Białowolski (2009) reports that the volumes of cash loans, credit card and installment

credit in Poland are PLN33 billion (EUR 8.4 billion), PLN14 billions (EUR 3.5 billion) and

PLN10.5 billions (EUR 2.7 billion) respectively. Figure 52 highlights the rapid increase in

the use of two forms of credit in Poland: loans on current accounts and credit card credit.

In particular, credit cards credit in Poland grew from virtually zero in 1996 to EUR 3

billion in 2008.

Figure 52: Types of consumer credit; Poland

0.676

1.084

1.894

2.902

3.582

3.421

2.852

3.416

3.768

4.928

4.991

0.006

0.026

0.078

0.184

0.359 0.403

0.441

0.716

1.100

1.521

2.481

5.154

6.860

7.624

9.708

12.458

14.749

8.212

7.197

8.986

11.650

14.733

21.072

24.763

0.307

3.970

0.003

3.048

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

loans on current accounts loans related to credit cards other (including hire purchases)

Source: ECRI (2009), “Lending to Households 1995-2008”.

2.4.2.4.3 Home Credit in Poland203

In contrast to the UK, where home credit is primarily targeted at credit impaired

customers, in Poland the most important customer base for home credit are people

without a bank account and hence no credit score. To evaluate the scope of home credit

in Poland, one may take 857,000 customers of the dominant provider (early 2010) as a

lower bound.

The average size of a home loan in Poland is EUR 385, which is higher than the average

loan size in other Central and Eastern European countries where Provident operates

(EUR 349). Loans can be granted for 26, 39 or 52 weeks (which correspond to 6, 9 and

12 months respectively). To qualify for a loan, customers must receive an income from

employment, which includes income from being self-employed, and every type of other

self-reported income. New customers may not obtain a loan exceeding EUR 150-200.

Customers may be granted a maximum of two loans at a time with the total amount

outstanding not exceeding EUR 500. The average interest rate is about 20%. The

customer may choose to either service the loan through their bank account or through

cash repayments returned to an agent visiting them at home. These two options entail

different pricing structures. In case of collecting repayments at home, a convenience fee

is charged for calling at customer’s home when collecting weekly payments. No default

charges are applied and the repayment schedule is usually flexible in that it is easy for a

customer to postpone a payment. In case a loan is serviced through the customer’s bank

203 The information in this section is drawn from an interview with Provident Polska, the main provider of home

credit in Poland (http://www.ipfin.co.uk/pages/key_milestones), unless indicated otherwise.

214 iff/ZEW – Final Report on interest rate restrictions in the EU

account, no convenience fee is charged, but default charges apply in case of arrears. The

home collection convenience fee and default charges in case of a loan serviced through

the bank account were introduced subsequently to the introduction of the interest rate

ceilings on borrowing rates in Poland in 2005. Most of customers (75%) are reported to

come back for the next loan within one year.

2.4.2.4.4 Summary statistics

Summary statistics on consumer credit distribution by type is presented in Table 53. The

aim of the table is to provide information about the relative importance of the various

types of credit existing on the Polish market.

Table 53: Distribution of consumer credit; Poland, 2008

Type of credit EUR

billions

% of total

credit to

households

% of

consumer

credit

Total credit (including mortgages) 88.90 100.00

Mortgages 46.37 52.16

Consumer credit of which… 32.80 36.90 100.00

...loans on current accounts 4.99 5.61 15.22

...loans related to credit cards 3.05 3.43 9.29

...other (including hire purchases) 24.76 27.85 75.49

Source: ECRI (2009).

2.4.2.5 Sweden

2.4.2.5.1 Overview

The volume of consumer credit in Sweden in 2008 stood at EUR 13.3 billion, making it a

medium-size market, ranked 15 among the EU 27 countries. Like in other EU countries,

the volume of consumer credit in Sweden grew during the last decade: in 2003 the

volume outstanding was EUR 9.6 billion.

iff/ZEW – Final Report on interest rate restrictions in the EU 215

Figure 53: Consumer credit outstanding; Sweden, EUR billion

10.493

11.287

13.426

14.267

13.298

9.621

0.000

2.000

4.000

6.000

8.000

10.000

12.000

14.000

16.000

2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Consumer credit in Sweden is less important for the economy than in other EU Member

States. The ration of the volume of consumer credit to GDP stood at 4.6% in 2008, which

is well below the EU average of 8.75% (Figure 54). Since 2003 the importance of

consumer credit remained roughly the same: in 2003 the ratio had a value of 3.5%.

Figure 54: Consumer credit; Sweden; % of GDP

3.61%

3.87%

4.18%

4.40%

4.58%

3.47%

0%

1%

2%

3%

4%

5%

2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Figure 55 shows growth rates in consumer credit between 2004 and 2008. Like in case of

other European countries, we observe a declining credit growth in 2007 (from almost

19% in 2006 to 6.3% in 2007) and even negative growth in 2008 (-6.8%).

216 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 55: Consumer credit growth; Sweden; % p.a.

6.27%

18.95%

9.06%

-6.80%

7.57%

-10%

-5%

0%

5%

10%

15%

20%

25%

2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

2.4.2.5.2 SMS loans

Spring of 2006 saw an introduction of a new form of consumer credit in Sweden, SMS

loans. Companies that offer SMS loans do not have to perform a credit check which

extends easy credit access to low-income individuals. Accordingly to some estimates,

many SMS loans are below EUR 200. Another source reports that the loans on average

amount to 3,000 kronor (EUR 313). The loans come with average fees of 500 kronor

(EUR 52) and interest payments of 50 kronor (EUR 5.2). The loans are usually repaid

during a rather short period of time, such as 30 days. SMS loans quickly became popular

among younger groups of population.204

2.4.2.5.3 Summary statistics

Summary statistics on the distribution of credit to households in Sweden by type is

presented in Table 54. The aim of the table is to provide information about the relative

importance of the various types of credit existing on the Swedish market.

Table 54: Distribution of consumer credit; Sweden, 2008

Type of credit EUR

billions

% of total

credit to

households

Total credit (including mortgages) 196.87 100.00

Mortgages 128.41 65.22

Consumer credit 13.30 6.75

Source: ECRI (2009).

204 http://www.thelocal.se/.

iff/ZEW – Final Report on interest rate restrictions in the EU 217

2.4.2.6 United Kingdom

2.4.2.6.1 Overview

The UK has experienced a very dynamic development of its consumer credit markets

over the last decades: the consumer credit in the UK has grown from EUR 82.4 billion in

1995 to EUR 245.2 billion in 2008, thus increasing nearly three-fold in the last fourteen

years. In terms of consumer credit volume, UK market was the largest market in the EU

in 2008. At the end of 2008 the volume of consumer credit outstanding was EUR 245.2

billion (Figure 56). Like in other European countries, market volume declined: it dropped

from its maximum of EUR 302.3 billion in 2006.

Figure 56: Consumer credit outstanding; UK; EUR billion

108.349

136.774

150.741

195.507

216.581

247.826

282.045

307.948

316.956

245.217

302.294

256.312

260.121

82.417

0.000

50.000

100.000

150.000

200.000

250.000

300.000

350.000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Consumer credit represents an important element of the UK economy. In 2008 consumer

credit outstanding was about 16.2% of the UK GDP (Figure 57). This is nearly double the

EU average of 8.75%. This ratio has grown from 13.8% in 2000, albeit there was a slight

decrease in 2006 and 2007.

218 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 57: Consumer credit; UK; % of GDP

10.22%

10.99%

12.10%

13.09%

13.84%

14.76%

16.19%

15.83%

16.10%

9.52%

15.73%

15.85%

16.56%

16.85%

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

Figure 58 illustrates growth rates of consumer credit volume. During the period 1996-

2000 the market saw double-digit growth rates ranging from 11% to 31%.

Figure 58: Consumer credit growth rates; UK, % change p.a.

29.70%

4.96%

-1.46%

10.21%

31.46%

14.43%

-18.88%

10.78%

26.23%

-4.63%

10.04%

9.18%

2.93%

-30%

-20%

-10%

0%

10%

20%

30%

40%

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

2.4.2.6.2 Types and dynamics of consumer credit

The consumer credit market in the UK comprises the following segments:

- Secured lending other than first charge mortgages;

- Credit cards;

- Loans, including short-term small-value loans (including home credit and

pawnbroking loans);

- Mail order, hire purchase and store cards;

- Credit Unions.

iff/ZEW – Final Report on interest rate restrictions in the EU 219

2.4.2.6.3 Secured credit

In September 2003 secured lending accounted for GBP737 billion or 81% of the total

consumer credit (DTI, 2003). Table 55 demonstrates the development of the different

categories of consumer credit as a proportion of household liabilities and also represents

typical amounts owned on specific forms of loans. As we can see from the table, by 2002

the importance of credit card debt in the overall household debt had increased to almost

20%, while the importance of other categories had somewhat declined. In 2008, credit

card credit accounted for 23% of total consumer credit. After credit card debt, the most

important form of credit is mail orders and loans. Overdraft accounted only for 9% of the

household debt in 2002.

Table 55: Composition of consumer debt over time; UK

Proportion of Households with

Current Commitments (%)

Average amount owed by

households

(GBP, adjusted to 2002 prices)

2002 1989 2002 1989

Credit cards 19 15 1570 565

Mail order 17 23 240 na

Loans 15 16 5000 2190

Hire purchase/Credit 13 17 3800 *

Overdraft 9 12 450 1676

Store cards/accounts 8 7 210 197

Source: DTI (2003).

2.4.2.6.4 Home credit

Home credit is represented by small-value, short-term loans. Home credit loans are

usually repaid over a period of around one year or less, in weekly instalments which are

collected from the customer by the agent calling for this purpose to the customer’s

home. According to estimates by the Department of Trade and Industry (DTI), this

segment of consumer credit market has about 3 million customers (DTI, 2003). Collard

and Kempson (2005), citing several sources, report a more modest figure between

2 million and 2.5 million people. There are around 30,000 agents working in the home

credit industry. The sums advanced are usually small, typically below GBP500, with a

repayment period in the range of 26 to 52 weeks. A later study by the Competition

Commission (2006) suggests that about 70% of home credit loans are for less than

GBP500. Around 90 per cent are for less than GBP1,000. The mean repayment period is

43 weeks. Compared to the costs of credit from the mainstream credit providers, costs of

home-collected credit are much higher, with APRs ranging between 100% and 400% or

more (Collard and Kempson, 2005). OFT (2010) report slightly higher APR between

150% and 500%. The charges are fixed and ‘all-in’. That is, they include costs of home

collection and the costs of late payment (DTI, 2003; Corr, 2007). It should be pointed

out, however, that the direct comparison of the costs of mainstream and high-cost credit

is hindered by the differences in the values of principal amount and risks characterising

these two types of lending.

Accordingly to the estimates provided in the Competition Commission (2006), in 2005

the outstanding volume of home credit amounted to GBP1.3 billion. The Competition

Commission also provides an alternative estimate for this market derived by

220 iff/ZEW – Final Report on interest rate restrictions in the EU

Datamonitor, which suggests that the size of the market was GBP2 billion in 2003. As

estimated by OFT (2010) the total value of loans in 2008 lies between GBP1.2 billion and

GBP1.3 billion. Among the reasons for taking out a home loan, customers cite both

urgent needs for money and special occasions, such as birthdays or Christmas

(Competition Commission, 2006). In 2005 the home credit market in the UK was

dominated by four companies (Provident, Cattles, London and Scottish Bank(LBS) and

S&U). Recently two of them, Provident Financial and S&U, have increased their customer

base and market share because the two others, LBS and Cattles, have reduced their

exposure.205 Provident, the largest of the four providers, controlled about 50% of the

market as reported by Collard and Kempson (2005). OFT (2010) estimates the recent

figure to be higher as Provident was able to increase its market share following the

difficulties faced by LBS and Cattles.

2.4.2.6.5 Pawnbroker loans

Pawnbroking is a form of secured lending in form of small cash loans secured by

property, usually jewelry. According to the estimates from the DTI, about 0.1% of

households admitted having a loan from a pawnbroker. This may be an underestimate

due to reluctance to admit use of credit in general and pawn-broking credit in particular

(Collard and Kempson, 2005). Corr (2007) provides an estimate of ca. 750 thousands

users of pawn loans.

The National Pawnbroking Association estimates that there are about 1,200 pawn broking

outlets in the UK. There exist two national chains: Albermarle and Bond and Harvey and

Thompson. Each has about fifty outlets. The rest of the industry consists of small

companies operating at local or regional level (Collard and Kempson, 2005).

According to the estimates by the National Pawnbroking Association, the volume of

pawnbroking loans in the UK is about GBP576 million (new business).206 However, this

estimate is based on a number of assumptions and should be treated with caution. The

average loan in the UK is believed to be about GBP120-GBP150 and the average term is

about 3.5 months. The average pledge book per outlet is around GBP160,000. According

to OFT (2010) the value of the annual loan book for 2008 was between GBP500-GBP600

million. With a reported number of 1200 pawn broking outlets this would suggest, that

the average annual book per outlet amounts to GBP417,000-GBP500,000. Collard and

Kempson (2005) report that APRs charged on pawn loans range from 70% to 200%. The

typical APR as reported by OFT (2010) is 100%. Like in case with home credit, the small

scale of lending, small values and short period for which pawn loans are granted make

the comparison of costs of mainstream and pawn loans less straightforward.

2.4.2.6.6 Payday Loans

Payday loans are a form of unsecured small-amount, short-term lending. To access this

form of credit, borrowers must have regular income, usually from paid employment;

have a bank account and a cheque book. Customers write a cheque to the company,

which in return for a fee agrees not to present the cheque for up to 30 days. They are

generally available for amounts between GBP100 and GBP1,000.207 The customer receives

a loan in the amount indicated on the cheque less the fee. OFT (2010) report that the

total charge of credit (TCC) for payday lending is varying between 10% and 20% for

high-street borrowing. TCC for borrowing online is generally higher. It is estimated that

205 LBS went into administration at the end of 2008 and Cattles stopped lending to customers shortly

afterwards (OFT 2010).

206 Based on the information obtained from the National Pawnbroking Association (2009).

207 See www.moneysupermarket.com.

iff/ZEW – Final Report on interest rate restrictions in the EU 221

about 1500 outlets in the UK provide payday loans (Collard and Kempson (2005), OFT

(2010)208). This type of lending is often associated with cheque cashing services which

advance funds on a post-dated cheque. Payday loans are not only used to cover

immediate personal needs by individuals who cannot obtain credit elsewhere but also

because of the ease of the transaction and because of the higher transparency in the

charges as compared to the bank charges levied for unarranged overdrafts (OFT 2010).

It is reported to be predominantly used by younger working male, unmarried with no

children, but also by low-income families struggling to make ends meet or in case of an

emergency (Collard and Kempson, 2005; Finlay, 2009; OFT, 2010). Dominy and

Kempson (2003) report that an average outlet issues about 50-60 payday loans a

months, with larger outlets issuing 150-200 payday advances a month. The estimated

number of customers of payday industry in 2003 was approximately 300,000 (Dominy

and Kempson, 2003). They estimate that slightly less than one million advances are

issued in the industry per year. OFT (2010) estimate that for 2008 the value of total

payday loans granted was between GBP700-GBP900 million.209 Considering the reported

typical loan size of GBP300 this would suggest that the number of advances issued in the

industry was between 2.3 and 3 million – well above the number estimated in Dominy

and Kempson (2003).

2.4.2.6.7 Credit cards

In 1971 there was only one type of credit card (Barclaycard) available in the UK. In 2003

there were around 1,300 types. The amount of money owed on credit cards has

increased substantially from GBP32m in 1971 (DTI, 2003) to over GBP25bn (BBA, 2009)

today. Accordingly to DTI (2003), the credit card market in the UK is the most developed

in Europe, accounting for about a third of all EU transactions, with many consumers now

using credit cards as their preferred payment medium. After the US, the UK has the

largest number of credit cards per capita with GBP63 million credit cards in circulation

(Department for Business Innovation and Skills, 2009). A major driver for the take-up

and usage of credit cards, in recent years, has been the rapid development in ecommerce.

Figure 59 presents development in the credit card credit and comparison with the use of

overdrafts from 1997 to 2009 based on the data provided by the British Bankers’

Association (BBA). Accordingly to BBA, the volume of credit card credit amounted to at

least GBP25.5 billion at the end of 2009. While debt on credit cards declined somewhat

during 2007-2009, it remained an important source of credit for consumers as other

forms of lending dried up due to funding constrains faced by financial institutions.

208 However, the number is believed to have increased to 2000 outlets in 2009 as reported in OFT (2010).

209 However, this estimate should be treated with caution as OFT (2010) is lacking publically available data to

underpin the value.

222 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 59: Consumer credit outstanding: Cards and overdrafts; UK; GBP million

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Credit Card Credit Overdraft Lending

Source: British Bankers Association (BBA), www.bba.org.uk

About 69% of consumers pay off their outstanding balances on credit cards every month

or most months while at the same time a significant number of customers carry high

level of unsustainable debt on one or sometimes more than one credit card. Customers

who sought advice from the consumer Credit Councelling Service (CCCS) in 2008 had

average credit card debts of nearly GBP15,000, with 90% of such customers having an

income less than GBP 30,000. For customers with income less than GBP 10,000 per year,

the average credit card debt was GBP 8,000 (Department for Business Innovation and

Skills, 2009).

2.4.2.6.8 Mail order, hire-purchase and store cards

Mail order credit provides consumers with an opportunity to make purchases that can be

paid off over an agreed period of time. The credit is paid off during regular periods of

time, such as on a weekly basis. DTI (2003) reports that mail order credit is used by

about 20.8 million users. Hire purchase agreements allow individuals to finance the

purchase of expensive assets, such as a car. The agreement is usually structured in a

way that the consumer pays a deposit, a number of monthly instalments and a final

payment to secure ownership of the asset. Store cards allow consumers a form of

running-account credit to purchase goods from a particular store. The cards are usually

store-branded, but the credit will usually be provided by a finance company. Often store

cards have incentives attached, such as participation in promotions and discounts (DTI,

2003). Department for Business Innovation and Skills (2009) estimates that as of

December 2008 the market for store cards had approximately 1.9 billion of outstanding

balances. This makes it a much smaller market than the credit card market. While the

use of store cards was on the rise until about 2006, since then it saw a decline, since

stores increased their offering of credit card products.

2.4.2.6.9 Summary statistics

Summary statistics on the distribution of consumer credit in the UK by type is presented

in Table 56. The figures refer to the overall UK consumer credit market. However, the

provided figures in the table should be treated with caution since, in the absence of

official statistics, in a number of cases they are based on estimates drawn from various

sources. The aim of the table is to provide information about the relative importance of

the various types of credit existing in the UK market.

As the table shows, high-cost credit, including money lending, payday loans and pawn

broker loans account for about 0.8% of the total consumer credit in the UK.

iff/ZEW – Final Report on interest rate restrictions in the EU 223

Table 56: Distribution of consumer credit; UK, 2008

Type of Credit EUR billion % of Total

Credit

% of Consumer

Credit

Total credit 1 095.52 100.00

Mortgages 797.94 72.84

Consumer credit out of which… 245.22 22.38 100.00

...credit cards 56.58 5.16 23.07

...other 188.64 17.22 76.93

Other credit of which…

…money lending* 0.89 0.08 0.36

…pawn broker loans** 0.52 0.05 0.21

…payday loans*** 0.5 0.05 0.20

Source: ECRI (2009) unless indicated otherwise.

Notes:

*Estimate for 2005 (Competition Commission (2006); ECB).

**Estimate for 2009 (National Pawnbroking Association (2010); ECB).

*** Estimate for 2003, based on Dominy and Kempson (2003).

Percentages for consumer credit do not add up to 100 due to difering sources of the data.

224 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5 Discussion of the hypotheses

This part of the report discusses the hypotheses we have derived from our theoretical

considerations in Chapter 2.1. In this discussion, we make use of a variety of sources.

We reconsider existing literature (partly already mentioned in Chapter 2.2) to discuss

previous evidence on these issues. Where deemed necessary, we point at shortcomings

of these reports and discuss their plausibility. We also make use of the data we have

already introduced in the overview of the credit markets in Chapter 2.3, and in the case

study descriptions provided in Chapter 2.4. Additional data sources such as the EU-SILC

and the Eurobarometer are also employed to obtain a more complete picture. One has to

note, however, that in many cases these macroeconomic data can only give rough

evidence on the hypotheses; a more exact identification of causal effects would require

micro data, which are not available at all or do not allow international comparisons to a

sufficient degree. To improve the evidence base, we also make use of particular

stakeholders’ experience in the consumer credit market, such as regulators, consumer

associations, providers and provider associations. This experience is predominantly taken

from the Stakeholder Questionnaire (SQ) attached in the Appendix. There is also a

stakeholder questionnaire sent exclusively to individual providers (PQ), which is

considered complementary to the SQ. This approach enables us to learn from

stakeholders’ experience with regard to the effect of interest rate restrictions (or lack of

thereof) on the credit markets in the respective countries. It also allows a comparison of

the answers from stakeholders in countries with IRR with the answers of those from

countries without IRR and a qualitative discussion of the differences. The inclusion of

heterogeneous stakeholders (regulators, consumer agencies, supplier associations) will

enable us to consider potentially divergent perspectives about issues in question in a

balanced way. Based on the entire set of information discussed above, we derive a final

judgement about the validity of each hypothesis. The brief evaluations “plausible”,

“inconclusive” or “unlikely” are given upfront in the overview parts.

As the quality of information from all sources differs across hypotheses (eg. for some

hypotheses we subsequently point to shortcomings of data to set the effect of interest

rate restrictions apart from other effects), a purely mechanistic weighting scheme of

potentially conflicting findings from different points of view (eg. 30 % literature review,

30% stakeholder assessment, 40% collected data) is not feasible. Rather, the evaluation

of each hypothesis requires a judgement of the credibility and usefulness of the different

information sources. Specifically, we base our evaluation on the following classification:

Plausible:A hypothesis is found to be “plausible” if there is sufficient evidence in favour

of it, and only little counterevidence to it. Such evidence may be based on convincing

analyses in previous studies, the data we have collected or the overall impression of

stakeholders in the market. We conjecture that a “plausible” hypothesis should contain at

least some relevance in different national settings, irrespective of the economic and

institutional circumstances. However, as market outcomes are a mixture of various

aspects of an economy, we will be careful to discuss the limitations of “plausible”

hypotheses.

Inconclusive: A hypothesis is evaluated to exhibit “inconclusive” results if there is

conflicting evidence from various information sources (without a dominance of either

favouring or opposing evidence), or if the available data are insufficient for an evaluation.

We will point to such data limitations in our discussion. Conflicting evidence suggests that

interest rate restrictions may have an effect as postulated by the theoretical

considerations, but that this effect does not necessarily materialise in an economically

significant way under all circumstances in all countries. For example, observed

developments might be caused by interest rate restrictions under the institutions (or, eg.

consumer preferences) of one country, while this need not hold in other countries.

Unlikely: A hypothesis is found to be “unlikely” if there is no evidence (or even

substantial counterevidence) for it. A judgement of “unlikely” to a hypothesis suggests

iff/ZEW – Final Report on interest rate restrictions in the EU 225

that even theoretically convincing effects of interest rate restrictions are not expected to

materialise in an economically meaningful way, as the scope of the effect is too little or it

is washed out by other simultaneous developments in the market.

226 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5.1 H1: IRR reduce credit access, in particular for low-income borrowers

2.5.1.1 An overview

Based on findings from previous literature, the Eurobarometer survey and the

Stakeholder Questionnaire, we find H1 to be plausible. This result has to be seen in the

context of the following remarks which underline its limitations:

H1 is unanimously confirmed by the existing literature which analyses or

comments interest rate restrictions. While the identification of the effects of

interest rate restrictions appears to be solid, one has to keep in mind that the

analysed interest rate caps (eg. 12%) are rather low compared to the typical

levels in some countries of the European Union. It is therefore questionable

whether the interest rate caps in the Member States have similar effects on, eg.

auto loans, as it has been found in the US studies.

In contrast, comparing data from the Eurobarometer survey on households

holding neither a credit card nor having an overdraft facility in five of the case

study countries, H1 cannot be confirmed. Conflicting findings from the

literature and survey data might stem from the different influences interest rate

restrictions have on the access to mainstream and high-cost credit. Due to the

relatively high levels of interest rate caps operated in most EU countries, H1 may

thus solely hold in the context of high-cost credit, while access to mainstream

credit products is rather unaffected.

One has to be careful, however, interpreting the incidence of credit to be

equivalent to the supply of credit, assuming the demand for credit to be constant

across countries. The attitude to credit differs substantially, explaining both the

existence of interest rate restrictions and differences in the prevalence of different

credit facilities.

Even if H1 holds with regard to high-cost credit, it does not imply that

consequences of reduced credit access are purely negative. The majority of

stakeholders does not see a need for increasing credit access to low-income

borrowers.

The stakeholders in countries with IRR see a slightly stronger need in increasing

credit access to low-income borrowers. This is an (albeit weak) confirmation of

H1.

The stakeholders unanimously express their view that H1 is likely to hold.

2.5.1.2 Literature

Chapter 2.2.6 has summarised a consensus of existing studies that interest rate

restrictions may limit credit access to some groups of customers. Villegas (1982) finds

that lower interest rate caps increase the rejection probabilities in automotive credit

markets. Zinman (2008) investigates the effect of the interest rate cap on payday loans

in Oregon and documents that payday credit access is strongly reduced compared to

Washington State, where no such cap exists. The view that credit access is reduced in

the presence of interest rate restrictions is also taken by Policis (2004), Białowolski

(2009), and OFT (2010) in their studies on high cost credit or interest rate restrictions in

Member States of the European Union. IGF/IGAS (2009) confirm this view, but point out

that the reduction of credit access may be seen as an objective of interest rate

restrictions.

iff/ZEW – Final Report on interest rate restrictions in the EU 227

Some argue that if providers cannot charge the levels of interest rates necessary to cover

higher risk, they will restrict credit access accordingly. This argument has also been

made by Fernando (2006) in the context of microcredit for developing Asian countries:

he points out that microcredit is a personal, intensive, costly (on a per-unit basis)

business, such that high effective interest rates are required to maintain the lenders’

activity to serve poor people. Helms/Reil (2004) argue that interest ceilings discourage

commercial banks from expanding into higher-cost rural or microcredit markets in

developing countries.

Overall, the literature contains strong arguments that credit access to specific groups of

the population may be reduced, which is broadly in line with H1. However, as detailed in

Chapter 2.2.6, the US studies consider the case of much lower interest rate caps (eg.

12%), and are thus only partly transferable to European countries.

2.5.1.3 Cross-Country Comparison

Issue 1: Perception of difficulty of credit access

Facts and Figures

If H1 holds, one can expect credit access to be more complicated in countries with

interest rate restrictions, in particular for low-income borrowers.

Figure 60 depicts the fractions of households agreeing on the question from the

Eurobarometer survey (60.2, 2003)210 asking: “Do you tend to agree or tend to disagree

on the statement: You can borrow as much as you like, there are no real checks?”.

Overall, these responses indicate that households can borrow as much as they like in

only a minority of all countries. Interestingly, the highest fractions of households

agreeing on the statement are found in Germany and Sweden, whereas the lowest

fractions agreeing on the statement are from the Netherlands and the UK.

It seems plausible that Swedish households feel to have easy access to credit because

the level of credit exclusion211 is low among the total population (Figure 61). However, it

is surprising that households in Germany, where the level of credit exclusion is highest,

feel to have easy and unlimited credit access, while households in the Netherlands,

where the level of financial exclusion is lowest and overdraft facilities and credit cards are

widely spread do not feel to have easy access to credit.212

210 The Eurobarometer survey is collected on behalf of the European Commission. Relying on this cross-national

survey avoids incomparability emanating from lacking harmonisation. For more information about the

survey and the special module analysed in this part of the study, see the Appendix. Since information on

household income is not included in the more recent Eurobarometer 63.2. (2005), the analysis on H1 has to

rely on the data from 2003.

211 Credit exclusion is defined on the basis of Eurobarometer data. Those households are regarded as excluded

from short-term credit that do no have access to neither credit cards, nor other cards or overdraft facilities.

212 The incidence of different short-term credit facilities (overdraft facility, credit card, other card) has been

analysed on the basis of the Eurobarometer question “Do you personally have…?”. Venn diagrams showing

the prevalence of multiple and single holdings of the different credit instruments are included in the

Appendix.

228 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 60: Perception of easy and unlimited credit access

Source: Eurobarometer 60.2, Variables Q20.3, D29.

Discussion

Credit access is not judged to be easier in countries without IRR, which might be seen

as evidence against H1.

Issue 2: Access to short-term credit (overdraft, credit cards, other cards)

Facts and Figures

We conduct a cross-country comparison of the case study countries to investigate

whether there are differences in access to credit cards and overdraft facilities which may

be due to interest rate restrictions.

The proportions of households excluded from short-term credit are depicted in Figure 61

for the entire population and for the group of low-income households.213 Figure 61 reveals

that there are countries with high levels of credit exclusion from short-term credit such

as Germany (where there are IRR in place) and the UK (where there are no significant

interest rate restrictions). The case of the Netherlands is interesting, as credit exclusion

from short-term credit is very low in this country and levels of exclusion from short-term

credit are equal between low-income households214 and the overall population. This could

be a sign that households are not being excluded but simply do not demand short term

credit.

213 Since it cannot be ruled out that credit cards and other cards are pure debit cards, considering those

households to have access to at least one of the three facilities rather overestimates the fraction of

households that have access to short-term credit. Looking at the opposite is therefore a tentative estimate

of the percentage of households which are excluded from short-term mainstream credit.

214 Households are considered to live on a low income if their monthly total income lies below 60% of the

median income.

iff/ZEW – Final Report on interest rate restrictions in the EU 229

Figure 61: Households without access to short-term credit

Source: Eurobarometer 60.2.

Discussion

First, as expected, low-income households have less access to short-term credit facilities.

However, it is not possible to draw the conclusion that differences in credit

access (irrespective of households’ income) are due to interest rate restrictions,

as H1 suggests. One explanation is that these figures of the non-existence of particular

credit sources do not reflect credit constraints, but simply the lack of willingness to

obtain credit. Also, one has to note that the Eurobarometer survey only captures

mainstram credit facilities, which seem to be influenced less by interest rate restrictions

than is high-cost credit.

Issue 3: Is credit useful?

Following from the literature finding that interest rate restrictions seem to reduce access

to high-cost credit it is worthwhile examining whether reduced credit-access is perceived

as a shortcoming at all.To shed more light on this latter point, we discuss different

attitudes in the case study countries. These attitudes are derived from the

Eurobarometer 60.2 question: “Do you tend to agree or tend to disagree on the

statement: Buying on credit is more useful than dangerous?”.

Facts and Figures

The fractions of households which tend to agree on the statement that credit is more

useful than dangerous are depicted in Figure 62. It can be seen that, among the case

study countries, the statement is only shared by a majority in the UK. Sweden is second,

while, in contrast, there are low support rates for this statement in Germany, France and

the Netherlands215. Figure 62 also shows the prevalence of consumption loans216 and

savings ratios in the different countries. There is a clear relationship in a sense that, in

countries where credit is rather perceived useful than dangerous, households borrow

more often and save less.

215 Poland is not included in the Eurobarometer survey. However, a major home credit provider operating in the

country shares experience of general “debt reluctance” of the society compared to the borrowers in UK.

216 Loans with a maturity of more than 12 month which are not used for car or house purchases.

230 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 62: Households considering credit to be useful

Source: Eurobarometer 60.2, Variables Q20.2, Q11, D29; BIS, Savings ratio.

Note: The percentage of households indicating to hold a consumption loan and the savings ratio relate to the

right hand axis.

Discussion

It appears that the public opinion about the nature of credit is able to explain the

existence or non-existence of interest rate restrictions. More than that: there is higher

incidence of credit in countries where credit is more widely accepted than in countries

where public support for credit is lower. It therefore seems that it is not only credit

access but also to a substantial degree credit demand which influences the number of

households relying on different sources of credit.

2.5.1.4 Views of Stakeholders

Issue 1: Is there a need to increase credit access for low-income borrowers?

H1 suggests that there is a lack of credit access for low-income borrowers in countries

with IRR. This paragraph therefore discusses whether or not stakeholders in countries

with IRR see a higher need for increasing the level of credit access (compared to the

present level) than stakeholders in countries without IRR.

Facts and Figures

As Figure 63 shows, more Stakeholders are against an increased access to consumer

credit for low-income borrowers than in favour of it. In written comments, consumer

organisations and government officials point out that credit access is a necessary

condition for over-indebtedness. Interestingly, the share of proponents of increased

credit access for low-income borrowers is similar for provider and consumer associations.

Yet, in repeated feedback credit card providers and home credit providers217 from UK and

Poland regard credit exclusion as a macroeconomically and politically undesired effect of

interest rate regulation.

217 eg. Credit card providers from UK as well as a home credit lender from Poland

iff/ZEW – Final Report on interest rate restrictions in the EU 231

Figure 63: Desirability of Credit Access for low-income borrowers, by Stakeholder Type

0% 20% 40% 60% 80% 100%

Consumer

Organisations

Provider

Associations

Other

Yes

No

Don't know

SQ Question: In your view, should low-income borrowers be given any greater access to credit than they are

given at present?

Figure 64 reveals that there is no strong difference in the responses by stakeholders from

Member States with IRR and Member States without IRR. It appears that the share of

proponents of an increased credit access of low-income borrowers is slightly larger in

countries with IRR. However, the number of its opponents is equally large in both types

of countries.

Figure 64: Desirability of Credit Access for low-income borrowers, by Type of Country of Origin

0% 20% 40% 60% 80% 100%

MS without IRR

MS with IRR Yes

No

Don't know

SQ Question: In your view, should low-income borrowers be given any greater access to credit than they are

given at present?

Discussion

If H1 holds, one would expect respondents to report insufficient credit access for lowincome

borrowers particularly in countries with IRR, while credit access should pose no

problem in countries without IRR. Overall, the need for further credit access for lowincome

borrowers is found to be at comparable levels for both types of countries. As the

results above document, there is only a slight tendency in favour of H1, if any.

However, one has to note that these answers do not reveal anything about the absolute

level of credit access in these two types of countries: it may well be that the level of

credit access is higher in countries without IRR, but that stakeholders in both regulatory

regimes consider their respective status quo level of credit access appropriate to the

same extent. This would imply that the politically desired level of credit access should be

heterogeneous across Member States, depending on the respective financial culture.

Issue 2: Differences in Credit Options for low-income borrowers

H1 suggests that, in countries with IRR, low-income borrowers have fewer options to

obtain credit than in countries without IRR. The following paragraphs collect the

stakeholders’ views on this issue.

232 iff/ZEW – Final Report on interest rate restrictions in the EU

Facts and Figures

Figure 65 shows that low-income borrowers are perceived to have access to rather few or

very few credit options, as indicated by the majority of respondents (66%). The evidence

is consistent for both Member States with IRR (with 64% of respondents in this category

indicating the latter) and Member States without IRR (with 68% of respondents in this

category indicating the latter).

Figure 65: Level of Credit Options to low-income borrowers

0% 20% 40% 60% 80% 100%

MS without IRR

MS with IRR Very low

Rather low

Rather high

Very high

SQ Question: How would you describe the level of credit options available to low-income consumers to choose

from? (Very low/rather low/rather high/very high)

Discussion

Unlike implied by H1, the stakeholders’ responses about the levels of credit options for

low-income borrowers do not differ substantially between countries with and without

interest rate restrictions. These finding contradicts the validity of H1.

Issue 3: Effect of the introduction of interest rate restrictions on credit access

The following paragraph discusses the views of stakeholders responding to the

Stakeholder Questionnaire on the introduction of interest rate restrictions in a

hypothetical country which has had no IRR before.218 For purposes of comparison, we

suggested two alternative ways of interest rate restrictions: a relative one, defined as

twice the average rate prevailing in the market, as well as a fixed interest rate cap of

30% p.a.

Facts and Figures

Figure 66 provides separate graphs for the different stakeholder types analysed before

(provider associations, consumer associations and a third category “other activity” which

includes financial regulators, banking authorities, government officials and others). It

also distinguishes between these groups in countries with IRR and those without.

Figure 66 illustrates that 70% of stakeholders expect the introduction of a floating

interest rate restriction (ie. a relative rate ceiling) to lead to reduced credit access of lowincome

borrowers. Thereby, provider associations from both Member States with IRR and

Member States without IRR are unanimous in their expectations. There is a slight

disagreement among respondent from consumer organisations, as 11% of them would

even expect an increase in credit access following an introduction of IRR. The percentage

218 However, we cannot be sure at this point that the stakeholder did not misunderstand the question and give

an answer which is appropriate for their own country instead of the suggested hypothetical country.

iff/ZEW – Final Report on interest rate restrictions in the EU 233

of respondents expecting an increase (20%) is at largest among “other” stakeholders

such as financial regulators and banking authorities from countries without IRR.

Figure 66: Effects of IRR (2x average) on credit access, for low-income borrowers

0 5 10 0 5 10

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market access to

credit of the low-income consumers would… (decrease(-1)/stay same(0)/increase(1))?

Figure 67 shows that the results for a hypothetical introduction of a fixed interest rate

cap of 30% APR differs from the results above. Part of the difference (especially in

Member States without IRR) is based on the different initial average interest rates in the

distinctive countries which makes the fixed absolute ceiling of 30% differently restrictive

as compared to relative interest rate restrictions.219 Interestingly, (despite of repeated

feedback that the level of 30% APR is rather high and unlikely to be binding) a large

majority of stakeholders (78%) claim that a fixed ceiling introduction would lead to a

decrease in credit access of low-income borrowers. Unlike for the relative ceiling

discussed above, ”other” stakeholders from Member States without IRR tend to agree on

this.

219 All respondents to the Provider Questionnaire from the Netherlands, Sweden, Germany and some

respondents from the UK indicate that a fixed ceiling set at 30% would not be binding for their institution

(PQ Question: In your view, what minimum level of interest rate ceiling (as calculated by the annual

percentage rate of charge (APRC)) would have a significant impact on the volume of consumer credit

granted BY YOUR INSTITUTION?). It should further be noted that the level of 30% would not be binding for

these institutions. In UK the level would only be binding for pawnbrokers, payday lenders and

moneylenders. On the other hand, when analogically asked about the level of a relative interest rate ceiling,

some respondents from the UK (among others a major commercial bank, an important building society as

well as a credit card provider) also indicated that a level of two times the market average would even be

binding for a broad range of loans, particularly general instalment loans, credit cards, overdrafts and even

mortgage loans.

234 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 67: Effects of IRR (max. of 30%) on credit access, for low-income borrowers

0 5 10 0 5 10

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

fixed at 30% APR for all credits, access to credit of the low-income consumers would… …

(decrease/no change/increase))?

Figure 68: Effects of IRR (2x average) on credit access, for average borrowers

0 5 10 0 5 10

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market access to credit

of average consumers would… (decrease/no change/increase)?

So far, we have discussed how, according to the views of different stakeholders, interest

rate restrictions are likely to affect credit access for low-income borrowers. In contrast,

there are diverging assessments with respect to credit access for average borrowers, as

iff/ZEW – Final Report on interest rate restrictions in the EU 235

Figure 68 documents that the provider associations from countries without IRR expect

negative effects not only for low-income, but also average-income consumers. This view

is not shared by the majority (63%) of the shareholders in countries with IRR, which

rather foresees an unchanged situation for average consumers.

In the Provider Questionnaire, loan providers were asked about their expectations on the

impact of a hypothetical introduction of a more restrictive, relative interest rate ceiling on

the credit access of high-risk borrowers.220 Most of the respondents would expect an IRR

to be followed by a higher credit exclusion to high-risk borrowers. Given that low-income

borrowers are often associated with high-risk, the evidence is in line with the results from

the Stakeholder Questionnaire. However, it should be noted that some providers

expressed their view that the terms “low-income” and “high-risk” should be kept apart as

they consider risk to be a function of the loan size rather than of the borrower´s

income.221

Discussion

These graphs document that H1 is attributed a substantial degree of plausibility by

virtually all groups of stakeholders: when asked for a hypothetical introduction of interest

rate restrictions, the respondents point to decreasing levels of credit access for lowincome

borrowers. An equivalent effect even for average consumers is only expected by

provider associations (and providers) who are primarily operating in countries without

IRR.

The finding that low-income customers are assumed to be more affected than average

consumers also supports the thesis that the influence of interest rate restrictions differs

for mainstream and high-cost credit.

Issue 4: What is the effect of different regulatory measures on credit access of lowincome

borrowers?

If H1 holds, interest rate restrictions would be an obstacle to credit access for lowincome

borrowers. If regulators have a choice between different measures (including

IRR) for a specific policy goal (eg. reduce over-indebtedness, see Chapter 2.5.6, page

274), they may take this potentially adverse effect into consideration. The following

paragraphs discuss the importance of this problem compared to adverse effects of other

measures on credit access.

Facts and Figures

As revealed in Figure 69 and Figure 70, almost all respondents from provider associations

(93%) and half of the respondents from consumer organisations assert that interest rate

restrictions would reduce credit access.222 Compared to other regulatory measures,

interest rate restrictions are expected to have the most pronounced effect in this respect.

In contrast, consumer organisations even expect policy measures which limit rolling over

of existing credit and which set tighter responsible lending standards are to increase the

level of credit (Figure 70). Provider associations, in contrast, do not expect any of the

listed policy measures to be effective in reducing credit access (Figure 69).

220 PQ Question: In 2009 PORTUGAL introduced rate ceilings capping the authorised interest rate on consumer

credit. Since the second quarter of 2010, the ceiling has been set as the average market APR (annual

percentage rate) charged in the previous quarter times 1.33. How, in your view, will this measure affect the

proportion of highrisk borrowers able to obtain credit (… ) Assume that the same interest rate ceilings, as

those introduced in Portugal in 2010, were introduced IN YOUR COUNTRY. What do you think will be the

effects of this measure on the proportion of high risk borrowers able to obtain credit IN YOUR COUNTRY?

221 Eg. home credit providers from UK and Poland as well as a provider association from UK.

222 This effect of reduced credit supply in the presence of interest rate restrictions is also highlighted by

Eurofinas (2010).

236 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 69: Effectiveness of different measures to increase credit access for low-income borrowers,

views of provider associations

13

1

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

IRR

6

5

1

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Regulations to limit rolling over

10

4

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Regulation on personal bankruptcy

2

9

1

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Tighter responsible lending requirements

9

3

1

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Strengthening disclosure obligations

SQ Question: Which of the following regulatory activities would have the most pronounced effects on improved

credit access? [the opposite effect/no effect/ very little effect / some effect /strong effect/ very

strong effect]

Figure 70: Effectiveness of different measures to increase credit access for low-income borrowers,

views of consumer organisations

8

2

1

5

0 2 4 6 8

opposite effect

no effect

very little effect

at least some effect

IRR

6

8

0 2 4 6 8

opposite effect

no effect

very little effect

at least some effect

Regulations to limit rolling over

4

2

4 4

0 2 4 6 8

opposite effect

no effect

very little effect

at least some effect

Regulation on personal bankruptcy

4

2

7 0

2

4

6

8

opposite effect

no effect

very little effect

at least some effect

Tighter responsible lending requirements

3

4

2

4

0 2 4 6 8

opposite effect

no effect

very little effect

at least some effect

Strengthening disclosure obligations

Discussion

It appears that, among other potential measures, interest rate restrictions are seen as a

particular obstacle to credit access for low-income borrowers. According to the

iff/ZEW – Final Report on interest rate restrictions in the EU 237

stakeholder´s views, H1 is thus plausible. It should be noted, however, that the

improvement of credit access is not a sensible policy target in the opinion of the majority

of stakeholders (see Figure 63).

238 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5.2 H2: IRR lead to a decline in the volumes of consumer credit granted.

2.5.2.1 Overview

Overall, it appears unlikely that the presence of IRR has the effect of substantially

shrinking the consumer credit market volume.

This hypothesis is potentially important because it is directed towards

macroeconomic implications: if there is evidence that the entire credit market is

impaired by interest rate restrictions, this will have consequences for the

economic development.

Earlier literature has compared credit market sizes in France, Germany and the

UK, concluding that this difference arises, among other factors, from different

interest rate regulation. This view is in favour of H2.

Comparing a larger number of countries with and without interest rate restrictons,

the details above document a heterogenous pattern of the level as well as the

growth rates of consumer credit across countries. These findings are against

H2.

After the introduction of interest rate restrictions in Poland and Germany, these

countries have experienced higher subsequent credit market growth, which is

against H2. In contrast, in the Dutch case, credit volume declined after a

tightening of interest rate caps. With respect to H2, these are mixed results.

However, since there are many other influencing factors in all cases, these results

have to be interpreted with caution.

As revealed in Chapter 2.4.2, the high cost credit market segments (which are

supposed to be the most sensitive to interest rate restrictons, such as home credit

providers, payday lenders and pwan brokers) only represent a minor share of the

total credit volume. An example is the UK, where non-mainstream lending

represents about 1% of the total consumer credit volume.

2.5.2.2 Introduction

The claim made in H2 is bold in the sense that it addresses the entire market for (at

least non-mortgage) consumer credit. It is not concerned with specific market segments

which might be particularly sensitive to interest rate restrictions. As a consequence, this

hypothesis is not so much directed towards microeconomic questions (eg. credit access

of a particular household), but rather to macroeconomic questions (eg. the role of credit

for economic growth): as described in Chapter 2.2, it has been argued that the level of

consumer credit has an impact on domestic spending and, ultimately, on economic

growth. This “multiplier effect”-type argument is also made by practitioners in the credit

market industry. The following paragraphs will therefore discuss the plausibility of H2.

2.5.2.3 Existing literature

For the Member States of the European Union, it is primarily the study by Policis (2006a)

which claims that interest rate restrictions are associated with lower levels of consumer

credit in the respective countries. To illustrate this point, Policis (2006a) compares the

size of the consumer credit markets in the UK, France and Germany and finds that the

UK (without IRR) has the largest, whereas Germany (with stricter rules) has the smallest

consumer credit market among these three countries. They consider these findings

evidence in favour of H2. Eurofinas (2010) claims that the low levels of consumer credit

per GDP in the Netherlands were caused by its regime of interest rate restrictions.

iff/ZEW – Final Report on interest rate restrictions in the EU 239

The expectations of declining volumes of consumer credit following an introduction of an

interest rate ceiling are largely shared by credit providers participating in our provider

survey.223 With only a few exceptions, participants expect the total volume of lending to

decrease.

Policis (2006a) even goes one step further and attributes the higher UK growth rates in

GDP per capita between 1995 and 2005 to the stronger growth in consumer credit

(compared to Germany and France). We find it doubtful that the main distinctive driver

between GDP growth in the UK and in Germany can be found in the prevalence of

consumer credit. Rather, it is a multitude of influencing factors differing across these

countries, in particular the sectoral composition of their economy, labour market

institutions, and the timing of business cycles. Even the housing price bubble in the UK

until the financial market crisis (while there was non in Germany) has impacted the

consumer spending behaviour. It also needs to be noted that in the years after the ones

considered by Policis (2006a), real GDP growth rates tended to be more favourable in

Germany than in the UK, although no substantial changes in consumer credit regulations

occurred.

As described in the following sections, the picture of the consumer credit market is much

more multi-faceted than Policis (2006a) suggest. This becomes apparent when

considering several countries beyond the ones considered by Policis (2006a), as is done

in Chapter 2.5.2.4.

2.5.2.4 Cross-Country comparison

Figure 71 indicates the size of the consumer credit markets (without mortgages) of the

selected six countries: Germany, France, the Netherlands, Poland, Sweden and the UK. It

becomes apparent that the largest market in this group is the UK market with a

consumer credit volume of EUR 245.2 billion. It is followed by Germany and France with

EUR 224 billion and EUR 155.7 billion. Notably, Poland’s consumer market already

surpassed those of the Netherlands and Sweden in absolute terms in 2008.

Figure 71: Consumer Credit in EUR billion, current prices

245.217

82.417

189.487

70.052

9.365

NA

NA

224.046

155.733

13.298

32.803 23.772

0

50

100

150

200

250

300

UK DE FR PL NL SE

1995 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

223 Among the participants are commercial banks, credit card providers, home credit providers, pawnbrokers

and a building society from the Netherlands, the UK, Germany, Sweden, Poland and the Czech Republic.

240 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 72 shows the trends in consumer credit in the period 1996-2008, with the year

1995 as base year. In all countries, with the exception of Germany, we observe growth

rates in the stock of consumer credit during this period. With the exception of Poland,

growth in all six countries slowed during the period of financial turmoil of 2007-2008.

In the context of interest rate restrictions, it is interesting to note that the country with

the fastest growth is Poland, where interest rate restrictions have been in place since

2005. Poland experienced the highest annual real growth rate of 13.2% during the period

from 1997-2008. After the introduction of IRR in Poland in 2005, real growth rates stood

at impressive 25.2%. Of course, we do not know what growth rates would have been

observed in this country had interest rate restrictions not been introduced. France and

the Netherlands, where interest rate restrictions are also in place, have enjoyed a stable

growth in the volume of consumer credit, albeit not as strong as in the UK.

Figure 72: Trends in Consumer Credit, End-of-Period Stocks, 1995=100%

0

100

200

300

400

500

600

700

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

DE FR NL PL SE UK

Source: ECRI (2009), “Lending to Households 1995-2008”.

Regarding the market share of non-mainstream lending in the six case study

countries, it seems that its importance compared to conventional credit is limited: For

Provident Polska, the main home credit provider in Poland, we estimate a market share

of less than 1% of the whole consumer credit market. The Dutch market for flash credits

(Flitskrediet) is tiny with an annual volume of EUR 6 million224 estimated, which accounts

for less than 0.03% of the consumer credit market. Even in UK, where the volume of

non-mainstream credit such as pawn broking, payday lending and home lending are

comparably high, their market share is slightly more than 1% of total consumer credit.

Discussion

The details above document a heterogenous pattern of the level as well as the growth

rates of consumer credit across countries. When looking at several countries with

diverging forms of interest rate regulation (instead of only UK, France and Germany), it

is apparent that this pattern cannot solely be explained by interest rate restrictions. As

other factors dominate relative size and growth of the consumer credit markets, it

appears unlikely that H2 holds.

224 With an annual volume of EUR 6,000,000 and an average maturity of 24 days, the average outstanding

credit would amount to only EUR 400,000.

iff/ZEW – Final Report on interest rate restrictions in the EU 241

2.5.2.5 Evidence from the past

In this section we provide an illustration of the development of the volumes of consumer

credit after the introduction of interest rate restrictions in Germany and Poland in 1981

and 2005, respectively, and a change in interest rate regulations in the Netherlands in

2006. A brief summary of the results is given in Table 57.

Table 57: Evidence from the past – volume of credit (H2)

Country Year Event Observation Tendency Alternative

Explanation

Poland 2005 Introduction

of IRR

Strong increase

in consumer

credit volume

Evidence

against H2

Strong

consumer

credit growth

in all New

Member

States;

convergence

to EU 15 levels

Netherlands 2006 Legislative

Change:

Decrease in

interest rate

cap

Decrease in

consumer credit

volume

Evidence

in favour

of H2

Financial

market crisis;

similar

development

in, eg.

Germany;

cyclical

downturn

Germany 1981 Introduction

of IRR

Increase in total

consumer credit

volume;

increase

particularly for

longer

maturities,

decrease for

shorter

maturities

Evidence

against H2

Secular

upward trend

for consumer

credit in the

1980s in

Europe

Example 1: Poland

Background

As described in Chapter 1.2.1.3.8, Poland introduced interest rate restrictions in 2005.

These restrictions refer to the borrowing rate and not to the total cost of credit (ie. APR),

and are defined as four times the central Lombard rate. In addition, restrictions were

imposed on the fees and additional charges related to granting a loan, which must not

exceed 5% of the amount of credit.

Facts and Figures

Figure 73 shows the dynamics of the volume of consumer credit in Poland in absolute

terms and relative to GDP for the period from 1996 through 2008: the volume of

consumer credit has been steadily increasing since 2003 and we observe no decline in

the volume of consumer credit in the three years after 2005. Notably, the average annual

growth rate of consumer credit during 2005-2008 was 22.5%. This is higher than the

242 iff/ZEW – Final Report on interest rate restrictions in the EU

average annual growth rate for the period 1996-2008, which constituted 18.2%. As a

percentage of GDP, consumer credit has grown from 6.5% in 2005 to 10.8% in 2008.

Figure 73: Consumer Credit in Poland, 1996-2008

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

0

5

10

15

20

25

30

35

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

% of GDP

billions

of Euros, constant 2005 prices

Consumer credit, in billions of EURO, constant 2005 prices (left axis) Consumer credit, % of GDP (right axis)

before IRR after IRR

Source: ECRI (2009), “Lending to Households 1995-2008”.

Figure 74 presents consumer credit per capita. Likewise, after 2005 consumer credit per

capita has grown from EUR 433 to EUR 790, an increase of 83%.

Figure 74: Consumer Credit per Capita in Poland, 1996-2008, in EUR.

0

100

200

300

400

500

600

700

800

900

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Consumer credit per capita, in thousands of EURO, constant 2005 prices

before IRR after IRR

Source: ECRI (2009), “Lending to Households 1995-2008”.

Discussion

These observations do not support the view that the introduction of interest rate

restrictions has led to a decline in the overall volume of credit granted, as suggested by

H2. Instead, the volumes continue to increase even after the introduction of interest rate

caps in 2005. However, one has to be cautious to draw general conclusions from these

data: it is impossible to tell whether or not the increase of consumer credit volume would

iff/ZEW – Final Report on interest rate restrictions in the EU 243

had been even more pronounced in the absence of such regulation. As Figure 25 through

Figure 28 demonstrate, the growth in the Polish consumer credit market (in the

considered time span) is rather typical for new EU Member States (which face

heterogeneous forms of interest rate regulations), and reflects a convergence in living

standards and GDP per capita in comparison to the EU 15 countries as well as the

development of financial services provision to households (consumer credit practically did

not exist in the country until ca. 1996 (ECRI, 2009). These figures in the Polish case

demonstrate that the introduction of IRR does not appear to have been a severe obstacle

in this process of convergence to EU 15 standards, as it does not result in lower total

consumer credit volumes.

Example 2: The Netherlands

Background

Unlike Poland, where interest rate restrictions were introduced in the recent past, the

Netherlands have decreased the level of existing interest rate caps from an allowed

spread of 17% to 12% above the legal rate in 2006. See Chapter 1.2.1.3.7 for details on

the interest rate ceiling in the Netherlands.

Facts and Figures

Figure 75 shows the development of consumer credit in absolute terms and as a

percentage of GDP in the Netherlands from 1995 to 2008. Consumer credit declined by

about 8.9% between 2006 and 2008. Consumer credit per capita declined from

EUR 1,508 to EUR 1,348, which represents a 10.6% decline (Figure 76).

Figure 75: Consumer Credit in the Netherlands, 1995-2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

before

change

after

change

0.000

5.000

10.000

15.000

20.000

25.000

30.000

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

billions

of Euros, constant 2005 prices

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

3.50%

4.00%

4.50%

5.00%

% of GDP

Consumer credit, in billions of EURO, constant 2005 prices (left axis) Consumer credit, % of GDP (right axis)

before

change

after

change

244 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 76: Consumer Credit per Capita in the Netherlands, 1995-2008; in EUR 000s

Source: ECRI (2009), “Lending to Households 1995-2008”.

Discussion

These observations suggest that the Netherlands experienced a decline in consumer

credit after the tightening of interest rate regulations, which would be in line with H2.

However, it appears advisable not to overstress these findings for several reasons: first, it

has to be noted that the years 2007 and 2008 were overshadowed by the financial

market crisis, which is likely to have impaired the ability of the financial institutions in the

Netherlands to provide credit to households.225 Second, it is unclear whether or not the

decline would also have occurred had there been no changes in regulation. In this

context, it has to be noted that in Germany (which had not seen a change in its

regulation at this point in time), the years 2007 and 2008 were also those with the lowest

consumer credit volume since 2002 (see Figure 34). Third, the time span of only two

years after the event is too short to allow firm conclusions: as the consumer credit

market has experienced relatively high growth rates in the Netherlands in the years prior

to 2006, the decline can also be seen as a cyclical downturn.

Example 3: Germany

Background

In Germany, interest rate restrictions have been introduced by courts since 1981.

According to these, it is not allowed to charge more than double the average interest

rate. (For more detail, see Chapter 1.2.1.3.1.) While the process of the introduction

required several court decisions over a time span of several years, we consider the year

1981 the starting point of the regulation for the purpose of this analysis. However, one

should also have a closer look at the subsequent years, in which interest rate restrictions

became increasingly effective.

Facts and Figures

The analysis of this less recent event is more cumbersome than the others due to the

lack of data availability. As the Bundesbank reports, the categorisation of consumer loan

contracts in different subgroups was entirely different before 1980 and is thus not

comparable to the figures from the post-1980 period. As a consequence, Figure 77 is

only able to capture the final stage of the pre-regulated period in Germany.

Figure 77 reveals the following: Germany experienced an increase in consumer credit

during the 1980s. Just before the implementation of interest rate restrictions in 1981, the

225 See Netherlands Bureau for Economic Policy Analysis, 2009. The Credit Crisis and the Dutch Economy 2009-

2009, (http://www.cpb.nl/eng/pub/cepmev/cep/2009/pdf/summary_uk.pdf.).

0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

1.600

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Consumer credit per capita, in thousands of EURO, constant 2005 prices

before change after change

iff/ZEW – Final Report on interest rate restrictions in the EU 245

levels of volumes for medium-term and long-term credit were similar. In the following

years during the 1980s, however, Germany experienced a steady increase of long-term

instalment credits, while there was a modest decline in medium term instalment credit

contracts in the first half of the 1980s (and a rebound afterwards). Also, the volume of

short-term instalment credit (which is on a much lower level) declined during this

decade.

Figure 77: Volume of Instalment Consumer Credit in Germany, different maturity groups,

in bn. Euros

0

20

40

60

80

100

120

19 80

19 82

19 84

19 86

19 88

19 90

19 92

19 94

19 96

19 98

20 00

20 02

20 04

20 06

20 08

0

0.5

1

1.5

2

2.5

3

3.5

Medium term Long term Short term

after

Source: www.bundesbank.de, own calculations.

Note: Interest rates used are effective annual interest rates (new business). “Short term” captures <1yr.,

“Medium term” captures 1-5 yrs., “Long term” captures >5 yrs.; the right y-axis displays the monetary

values for short term contracts only, while the left y-axis captures the values for the medium term and

long-term contracts.

Discussion

The overall instalment credit volume does not decrease, as H2 would suggest. It is

likely that there is a is a strong secular trend for consumers to finance part of their

expenditures by credit underlying this development. Again, it is impossible to quantify

whether or not this trend would have been even stronger without IRR.

We see a differentiated pattern for different maturity types of consumer credit: the

growth of consumer credit is most pronounced for long-term maturities, while there is a

decline in short-term and medium-term credit in the first half of the 1980s. This might be

a reaction to the introduction of interest rate restrictions: long-term contracts might give

lenders the opportunity to cover operational costs, as well as to lower the APR while

maintaining a given level of upfront administrative fees. These factors could help

maintain interest rates below the interest rate cap.

246 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5.3 H2a: Without IRR, more product types exist in the market

2.5.3.1 An Overview

Comparing the incidence of credit and especially high-cost credit in countries with and

without IRR, we conclude that H2a is plausible.

Countries without interest rate restrictions tend to have a higher prevalence of

personal loans/auto loans than countries with interest rate restrictions. This

observation supports H2a. However, the relationship between interest rate

restrictions and the level of credit is unlikely to be purely mechanistic. Rather,

both the lower levels of consumer credit and the tougher regulation of credit

(including interest rate restrictions) are consequences of a country’s attitude

towards credit.

The prevalence of mortgages can be expected to depend on a variety of factors,

such as the demand for owner-occupied housing or lending standards of banks. In

contrast, they are likely to be rather unaffected by typical interest rate caps, as

they have usually relatively low interest rates.

It is likely that interest rate restrictions prevent market entry of one or

some forms of high-cost credit, as H2a suggests. On the other hand, it is

probably also common that (in countries with IRR) providers tailor the products

such that existing interest rate legislation does not apply.

Higher-cost small-volume credits tend to be more important in countries without

IRR. This is revealed by the stakeholders, and confirms H2a.

Evidence from the stakeholder questionnaire reveals that countries without IRR

tend to have higher levels of coverage with (true) credit cards; there is mixed

evidence for overdraft facilities. It is likely that the acceptance of credit card credit

is again closely related to the country’s credit culture.

According to the views of the provider associations responding to the

Stakeholder Questionnaire, it is likely that the introduction of interest rate

restrictions would lead to a reduction of available credit types. This would

support H2a. The critical credit types are those which are most suitable for lowincome

customers.

2.5.3.2 Evidence from the Past / Perception of Stakeholders

Issue 1: How common are different credit types?

Credit markets are heterogeneous across EU Member States. To obtain a broad picture

about the product variety in a large set of countries, we attempt to characterise the

importance by consulting data from different sources: the Eurobarometer Survey from

2005, data from Oliver Wyman referring to the year 2006 and, where appropriate, we

return to data from ECRI as summarised in the Chapters 2.3.3 and 2.4. Furthermore we

consider the more recent perceived relative importance of different credit types as

reported by the surveyed stakeholders in the EU 27 countries. Along the line of these

findings, we discuss hypothesis H2a with respect to the individual types of credit.

a) Mortgages

Facts and Figures

iff/ZEW – Final Report on interest rate restrictions in the EU 247

Figure 78: Distribution of loans in EU 27 Member States, 2008

Source: ECRI (2009), “Lending to Households 1995-2008”.

As Figure 78 shows, mortgage loans amount for the highest shares of credit volumes

compared to other loans in most countries.226 Still, this does not directly tell how

widespread mortgage loans are in the different countries. Therefore, Figure 79 shows the

prevalence of mortgages which can be deducted from Eurobarometer data. The survey

asked households whether they hold a mortgage. The figure reveals, that even in

countries in which mortgage credit amounts for high shares of total credit, only small

fractions of households actually hold a mortgage loan (for example, this is the case for

Estonia, where 80% of total credit are mortgage loans, but only 4% of households hold a

mortgage). Altogether, data from ECRI and Eurobarometer match in a sense that most

countries, in which mortgage credit accounts for a high fraction of total credit, the

proportion of households holding a mortgage is high as well (examples are the

Netherlands, Denmark and the UK). The opposite shows true as well, since Slovenia,

Poland and Hungary (with low percentages of households holding a mortgage loan) are

among the countries in which mortgage credit accounts for less than (or around) 50% of

total credit.

Figure 79: Prevalence of mortgages among EU-25 households, 2005

Source: Eurobarometer 63.2, Variable: QB 11.

Note: Countries in which mortgage credit amounts to less or equal 50% are highlighted.

226 There are some exceptions of some south and east European countries.

248 iff/ZEW – Final Report on interest rate restrictions in the EU

These findings largely coincide with the results from the Stakeholder Questionnaire. To

characterise the importance of credit types in the different countries, we asked

stakeholders to evaluate the importance of specific products in their countries according

to the scale (0: not existent, 1: uncommon, 5: widespread). The median stakeholder

responses are displayed in Table 58. Mortgage loans are also subjectively perceived to

have the highest prevalence compared to other loans in most countries. An exception is,

eg. Germany, which exhibits the second lowest figure for mortgage loans after Slovakia

(the result for Slovakia is plausible comparing them to the figures from ECRI and

Eurobarmoter). The relatively low value for Germany is consistent with the low owneroccupation

rate for housing in Germany (Figure 80). It is also in line with the relatively

low (compared to other EU 15 countries) level of housing loans per capita, as

documented in Figure 17 and the low prevalence of mortgage debt which is documented

by Eurostat. However, one has to note that even for Germany, the volume of housing

credit amounts to more than half (55.99%) of the entire volume of credit to households

(Table 48, p.203). This is lower than the share of mortgages to total credit in France

(74.6 percent) or the UK (72.84 percent).

Figure 80: Owner-occupation rates for the EU 27 Member States, 2008

Source: Hypostat (2008). Countries in which mortgage credit amounts to less or equal 50% are highlighted.

Figure 80 demonstrates that the highest shares of ownership rates are found among the

New Member States of the European Union. This observation is the outcome of a massive

privatisation process in the transition to market economies in those countries.227 Still, as

Figure 17) reveals, the typical level of mortgage volume per capita in the New Member

States is rather low, and mortgages only account for 52.2 percent of total credit to

households. Nevertheless, with the exception of Slovakia, the stakeholders assess

mortgages to be relatively widespread in the New Member States. Note that the

Netherlands has the largest share of mortgages to total credit (89.15%), and

stakeholders accordingly perceive it to have a very strong importance.

227 For more details on the housing markets in Poland, see Kierzenkowski (2008).

iff/ZEW – Final Report on interest rate restrictions in the EU 249

Table 58: Prevalence of different product types across countries

Mortgage loan

Auto / vehicle

loan

Other personal

loans

Point-of-sale /

Mail-order loans

Overdraft

facilities

Credit-card

credit

Pawnbroker

loans

Home loans

SMS loans

Payday loans

Austria 4 2 4 3 5 3 2 1 1 1

Belgium 4.5 0 2.5 2 2.5 2.5 0 0 0 0

Bulgaria 4.5 1.5 4 1.5 2.5 3 2 1 1 1

Cyprus 5 5 4 4 5

Czech Republic 4 5 4 3 4 4 3 2 0.5 0

Denmark 5 5 5 5 5 5 2 0 3 0

Estonia 2.5 2.5 2.5 1 2 2.5 2.5 0.5 2.5 0.5

Finland 5 5 5 5 4 5 4 2 5 0.5

France 4 4 4 3.5 4 4 2 0.5 1 1

Germany 3 4 4 3 4 3 1 1 0 0.5

Greece 5 5 5 5 5 1

Hungary 5 4 4 0 2 3 2 3 0 0

Italy 5 5 5 4 3 4 1 2 1 1

Latvia 5 5 4 0.5 4 4.5 2.5 2 4.5 4

Lithuania 0 0 0 0 0 0 0 0 0 0

Luxembourg 5 5 4.5 3.5 3.5 4.5 1 1

Malta 5 5 5 3 5 1

Netherlands 4.5 4 3.5 3 5 3.5 2 2 2 2.5

Poland 4 3 4 1 4 4 1 2 0 0

Portugal 5 5 5 3 4 5 2 1.5 1 2

Romania 5 5 2.5 3.5 3.5 3.5 4.5 2

Slovakia 4 4 4.5 1 2.5 3 0.5 1.5 0.5 0.5

Slovenia 4 4 4.5 2.5 4 2.5 1.5 2 2 2

Spain 5 5 5 3 3 5 2 1 1 2

Sweden 5 3 3 3 1 3 1 1.5 2 0.5

United Kingdom 5 5 5 4.5 5 5 3 3 1 2.5

Source: Stakeholder Questionnaire, median values where multiple answers.

Scale for selection was 0: not existent, 1: uncommon to 5: widespread.

250 iff/ZEW – Final Report on interest rate restrictions in the EU

Discussion

Mortgages are unlikely to be affected by a general interest rate cap applying to all

different forms of consumer credit:228 as demonstrated by Table 42 (p. 185), the typical

levels of mortgage interest rates (depending on the fixation period) are lower than other

forms of consumer credit,229 ie. only few percentage points above long-term government

bond yields in those countries. In this context, note that the demand for owner-occupied

housing (and as a result for mortgages) depends on a variety of factors, such as the

availability of high quality rental dwellings as an attractive substitute to owner-occupied

housing, possibilities of tax deduction of mortgage interest rates, as well as the lenders’

collateral standards.

b) Overview on Consumer Credit

Facts and Figures

Consumer credit can itself be split in several sub-groups such as personal loans,

auto/vehicle loans, point-of-sale credit and revolving credit. An overview of the relative

importance of different forms of consumer credit, measured by each forms volume

relative to the national GDP is given in Figure 81. The results are taken from a report of

Oliver Wyman (2008) which relies on different institutional sources. It reports findings for

2006. The analysis will then focus on specific forms of consumer credit: personal loans

and auto / vehicle loans, revolving credit and non-conventional forms of credit.

Figure 81 shows an overview of the outstanding volumes of different credit forms in

relation to national GDP (labelled “market penetration”). These vary widely across

Europe. Concerning different regulatory regimes on interest rates (countries with IRR are

given in bold letters), there are higher volumes of the different credit forms outstanding

in countries without IRR.

228 Effects on mortgage markets may be more relevant in countries with different ceilings for different forms of

credit with comparably low levels for caps on mortgage interest rates. However, one should note in this

context that even for the French case (with separate IRR on mortgage credit), it has still been argued that

interest rate restrictions have a neglible effect on the mortgage markets (IGF/IGAS 2009, p.4/17).

229 As described in Section 4.5.4.2.1., the EU 27 median for 2009 is 3.52%- 5.38% for mortgage rates, 7.54%-

7.69% for consumer credit rates and 18.97% for overdrafts.

iff/ZEW – Final Report on interest rate restrictions in the EU 251

Figure 81: Market penetration for different consumer credit products, 2006

UK DK GR IE NO DE ES AT FR SE HU PT FI IT PL BE CZ SK

Source: Oliver Wyman (2008). Countries with IRR in place are given in bold, italic letters.

Discussion

There is a tendency that countries without IRR display higher volumes of different forms

of consumer credit than countries with IRR. This would support H2a. However, as

specified in the report by Oliver Wyman (2008), it is likely that penetration levels in

practice frequently coincide with countries’ attitudes towards credit in general and

different forms of consumer credit in particular. For example, it is specified, that Belgium

has always been culturally adverse to consumer credit which leads to lower levels of

penetration with consumer credit in total compared to other equally developed markets

(eg. Austria). On the other hand low market penetration in Slovakia and the Czech

Republic can be explained by the fact, that the markets are still very young. An example

for the dominance of a specific form of consumer credit can be given by Nordic markets

which are highly concentrated on personal loans as a consequence of their predominant

banking distribution channel.

c) Personal loans or auto loans

Facts and Figures

Personal loans and auto/vehicle loans are types of credit to finance major household

investments. We first refer to the Eurobarometer (2005) survey responses of interviewed

households. These objective figures are then cross-validated with the subjective

prevalence of the credit forms as reported in the Stakeholder Questionnaire.

252 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 82: Prevalence of personal loans across EU 25 Member States, 2005

Source: Eurobarometer 63.2, Variable QB 11.

Note: Since regulations on interest rates in Poland and Portugal have been introduced in 2005 and 2009

respectively, they are not included in the group of Member States with IRR.

Figure 83: Prevalence of auto / vehicle loans across EU-25 Member States, 2005

Source: Eurobarometer 63.2, Variable QB 11.

As Figure 82 and Figure 83 demonstrate, the prevalence of these types of credit is

different across countries. It appears that the households from countries with IRR (grey

colour) tend to report lower importance of this type of credit as the respondents from

countries without IRR (blue colour): there is a clustering of “blue” colour on the right

hand side of Figure 82 and Figure 83.

In line with this, Denmark, one of the countries without IRR, belongs to the countries in

which personal loans and auto/vehicle loans appear to be very common. This picture

corresponds with the description in Figure 21 (p. 176), which details that Denmark is

among the countries with the highest consumer credit per GDP. Same holds true for

Ireland, Sweden and the UK, although auto loans are less prevalent in the latter. As

opposed to these, the case study countries with IRR the Netherlands, France and

Germany appear to be below the average in terms of importance of personal or auto

loans.

The findings from the Eurobarometer survey can be validated with the up to date

perceptions of the respondents to the Stakeholder Questionnaire. Their subjective

MS with IRR MS without IRR

MS with IRR MS without IRR

iff/ZEW – Final Report on interest rate restrictions in the EU 253

perception how important personal loans and auto / vehicle loans are in their home

countries are displayed in the following figures. Again, the results for Member States with

IRR in place are shown in grey colour.

Figure 84: Prevalence of personal loans across countries

(0: not existent, 1: uncommon - 5: widespread)

SQ Question: Please indicate which of the following forms of consumer credit exist and to what extent they are

used.

Figure 85: Prevalence of auto/vehicle loans across countries

(0: not existent, 1: uncommon - 5: widespread)

SQ Question: Please indicate which of the following forms of consumer credit exist and to what extent they are

used.

Note: Cross categorical mean of the median response within a category.

Although Figure 84 and Figure 85 differ in some respects from the statements made

above for individual countries, it can be seen that countries without IRR tend to be

among those with the highest share of credit. This finding is slightly more pronounced in

the case of loans which are used to finance a vehicle.

Discussion

It appears that countries without IRR tend to have higher prevalence of personal

loans/auto loans than countries with IRR. One reason could be that low-income

borrowers (who have to accept differing conditions to compensate the lender for risk)

cannot be offered credit at prices below the cap. This would support H2a. However,

typical interest rates of personal loans (EU median 7.54%-7.69%) are usually well below

the restricted levels. Therefore, it is likely that tougher interest rate restrictions in

practice frequently coincide with other types of regulation which all have the purpose of

discouraging lending to impaired households. Hence, it is plausible to assume that both

lower levels of consumer credit and tougher regulation of credit (including interest rate

restrictions) both are consequences from the preferences in the society of the respective

country (see also the discussion about Figure 62 on the attitude to credit).

d) Revolving Credit

MS with IRR MS without IRR

MS with IRR MS without IRR

254 iff/ZEW – Final Report on interest rate restrictions in the EU

Facts and Figures

Overdraft credit and credit card credit are revolving credit products. Overdraft facilities

are linked to a current bank account, and permit to borrow money spontaneously when

the account’s credit balance turns negative. Credit cards grant credit up to a specified

limit. For both types, the amount borrowed and the date to pay off the debt can be

chosen flexibly. Compared to personal loans, however, credits from overdraft facilities

and credit cards are thought to be of smaller amounts and shorter durations. One has to

note that the mere existence of an overdraft facility does not imply that the consumer

actually makes use of it; there are also credit cards which are only used as a payment

device rather than as a device of credit.

To judge the prevalence of overdraft facilities it is useful to consider first the share of

households in the Eurobarometer (2005) survey who indicate that they have such a

facility.

Figure 86: Prevalence of overdraft facilities across EU-25 Member States, 2005

Source: Eurobarometer63.2, Variable: QB 11.

Figure 86 indicates these values for the year 2005 for the then 25 Member States. Again,

countries with IRR are coloured in grey, while those without significant restrictions are

shaded in blue. The UK is reconfirmed to have a highly developed credit market (while

having no IRR), with high importance also of overdraft facilities. In contrast, however,

Sweden (another case study country without IRR) is found to have only little prevalence

of overdraft facilities. Broadly there is no clear clustering of countries with or without IRR

on any end of the scale.

Comparing the importance of overdraft facilities as perceived by the respondents to the

Stakeholder Questionnaire across countries, Figure 87 demonstrates that countries with

IRR (grey colour) tend to be above average. Although the picture differs in some respects

from the statements made above for individual countries, the overall impression is

unchanged: No clear statement can be made about whether overdraft credit is more

prevalent in countries with or without IRR.

MS with IRR MS without IRR

iff/ZEW – Final Report on interest rate restrictions in the EU 255

Figure 87: Prevalence of overdraft facilities across countries

(0: not existent, 1: uncommon - 5: widespread)

SQ Question: Please indicate which of the following forms of consumer credit exist and to what extent they are

used.

Note: Median of the reponses within a category

However, these figures should be interpreted with caution. When stakeholders indicate

that overdraft facilities are “widespread”, it is not clear whether they judge a coverage

rate of 20%, 60% or 90% of the total population. In some sense, respondents may

consider all these figures to be characterised as “widespread”, as they are not asked to

compare them with corresponding figures from other countries (as we do in this text),

but with other types of credit in their own country.

Figure 88: Prevalence of credit cards across EU-25 Member States

Source: Eurobarometer 63.2, Variable QB 11.

It can be seen in Figure 88 that credit cards are least frequent in young credit markets

and in Portugal. It is striking that Eurobarometer respondents report high levels of credit

card usage in Sweden whereas only limited importance of credit card credit is found

when asking the stakeholders (see Figure 89 below). This can be explained as follows:

while stakeholders assess the importance of credit card credit, Eurobarometer asks

whether respondents hold a credit card. In Sweden, however, a pure debit function is

about twice as frequent as a delayed debit or credit function, which means that there is

no true credit involved in credit card payments. Another example is Germany where

there are relatively few pure credit cards (3 million)230 compared to delayed debit cards

(18.1 million), which are both counted as credit cards in the Eurobarometer survey.

Therefore, we consider the Eurobarometer (2003) data on credit cards to be poorly

suitable to measure the occurrence of credit card credit.

230 Figures are from BIS (2009).

MS with IRR MS without IRR

MS with IRR MS without IRR

256 iff/ZEW – Final Report on interest rate restrictions in the EU

Judging from stakeholders’ perceptions of the importance of credit card credit (Figure

89), it appears that there is only a limited number of countries with IRR in which credit

card credit is perceived to be important. Only in Portugal (which has introduced the

restrictions only recently) credit card credit is perceived to be very important. Again, the

UK belongs to the countries with high levels of importance, whereas Sweden shows low

levels.

Figure 89: Prevalence of credit card credit across countries

(0: not existent, 1: uncommon - 5: widespread)

SQ Question: Please indicate which of the following forms of consumer credit exist and to what extent they are

used.

Note: Median of the responses within a category

Discussion

It appears that credit card credit is more frequently used in countries with IRR. For

overdraft facilities, we obtain rather mixed evidence. We expect that parts of these

differences across countries are originating from a different credit culture and different

attitudes with respect to credit or the maturity of credit markets.

e) Pawnbroking loans, Home loans, SMS loans, Payday loans

This paragraph discusses different non-conventional forms of credit. They have in

common that they are granted for very short or short time spans, and that the credit

amounts are typically very small. As can be seen from the relatively low numbers

indicated in Table 58, these types of credit are perceived to be less important than other

forms of credit. This assessment corresponds to the relatively small share of these credit

types compared to the overall consumer market. Eg. as discussed in Chapter 2.4.2.6 the

volume of payday lending in the UK in 2008 is estimated at £ 700-900 million. This

amounts to only less than 1% of total consumer credit in the UK.

In general, it is obvious from the legal part of the study and the relatively low interest

rate ceilings in operation across the EU (see Table 8 on p63) that high priced doorstep

credit according to the terms as observable in the UK, would not be possible in these

countries where IRR exist. However, this does not answer the question of whether such

credit would not be equally available at much lower cost in these countries. The existence

or non-existence of each individual non-conventional credit product appears to be rather

idiosyncratic for each country: Home loans are perceived to be relatively widespread in

the UK and in the New Member States Czech Republic, Hungary, Latvia, Romania,

and Slovenia. As discussed in more detail in the case study section (p. 213), home loans

also exist in Poland. It appears that these loan types are particularly successful in

serving unbanked customers in Eastern European countries.231 In contrast, Payday loans

are found to be most important in the UK, in the Netherlands and in Latvia. SMS

231 In the UK, however, home credit serves primarily customers with a bad credit score.

MS with IRR MS without IRR

iff/ZEW – Final Report on interest rate restrictions in the EU 257

loans are reported to play a role in the Northern countries of Latvia, Finland, Estonia

and Denmark. In the case studies, we have also discussed the existence of SMS loans in

Sweden (p. 216) and that of the related Flitskrediet in the Netherlands (p. 209).232

Pawnbroking is a non-conventional source for credit which appears to be existent in a

majority of countries. It is not apparent that the degree of its importance is significantly

related to the regime of interest rate restrictions. However, the importance of

pawnbroking is perceived to be particularly high in the Finland, Romania, the UK and the

Czech Republic, which are countries without interest rate restrictons. However, one has

to keep in mind that pawnbrokers are frequently subject to regulatory regimes differing

from those applying to other types of lenders.

Discussion

Overall, it appears that the existence of these non-conventional credit types is

idiosyncratic to a specific country and the economic and the prevailing institutional

circumstances. Nevertheless, it is perceivable that interest rate restrictions do

prevent market entry of one or the other form of high cost credit, as H3

suggests. On the other hand, it is probably also not uncommon that providers tailor the

products to the regulatory needs, such as the (very short-term) Flietskrediet in the

Netherlands, which has made use of a loophole in the legislation which is currently being

closed.

2.5.3.3 Perception of Stakeholders

Issue 2: Prevelance of high cost small volume credit

If H2a holds, one would expect countries with IRR to have a lower prevalence of, eg.

higher-cost small-volume credit. The following paragraphs therefore compare the

stakeholders´ answers from Members States with and without interest rate restrictions.

Facts and Figures

Figure 90 shows that higher-cost small-volume credit is regarded as an unimportant or

only moderately important credit source by 80% of the responses from Member States

with IRR. In contrast, more than two-fifths of the respondents from Member States

without IRR indicate that high-cost small-volume credit is an important or very important

source of credit in their countries. The share of respondents indicating that high-cost

small-volume credit holds a significant or very significant importance is twice as high in

countries without IRR.

232 For the Netherlands, it is perceivable that the Stakeholders were referring to Flitskrediet when answering in

the category of “Payday loans”.

258 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 90: Importance of alternative lending (higher cost small volume credit)

as % of total lending activities

0% 20% 40% 60% 80% 100%

MS without

IRR

MS with IRR

Non-existent/

Insignificant

Of little significance/

Of some significance

Significant/

Very significant

SQ Question: What is the importance of alternative lending (higher-cost small credit)? (in EUR/local currency

and % of total activities if possible) [non-existent/insignificant/of little significance/of some

significance/significant/very significant]

Discussion

This confirms H2a: The stakeholders report lower levels of high-cost small-volume

credit in Member States with IRR.

Issue 3: Effect of the introduction of interest rate restrictions

The following paragraph discusses the views of stakeholders on the introduction of

interest rate restrictions in a hypothetical country which has had no IRR beforehand. For

purposes of comparison, we suggested two alternative ways of interest rate restrictions:

a relative one, defined as double the average rate prevailing in the market, as well as a

fixed interest rate cap of 30% p.a.

Facts and Figures

Figure 91 shows that stakeholders’ opinions clearly diverge with regard to the impact of

interest rate restrictions on product variety available to low-income borrowers. A large

majority of provider associations (94%) expects an introduction of a ceiling to decrease

the range of products available to low-income borrowers. It should be noted that

respondents from provider associations from Member States with IRR agree on the

matter with provider associations from countries without restrictions. In a separate

question regarding their assessment of the access of low-income borrowers to credit

options, stakeholders from countries without IRR indicated less credit options available to

low-income borrower than in countries with an existing ceiling (see Chapter 2.5.1.4).

Among consumer organisations and other stakeholders, such as banking authorities and

financial regulators, opinions on the impact of IRR on product variety seem to diverge.

iff/ZEW – Final Report on interest rate restrictions in the EU 259

Figure 91: Impact of a relative/floating IRR (2x market average) on the product variety available

to low-income borrowers

0 5 10 0 5 10

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market, for the lowincome

consumer variety of products would…[decrease/not change/increase]?

Figure 92: Impact of a relative/floating IRR (2x market average) on the product variety available

to an average borrower.

0 2 4 6 8 0 2 4 6 8

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market, for average

consumers variety of products would…[decrease/not change/increase]?

260 iff/ZEW – Final Report on interest rate restrictions in the EU

Regarding the product variety available to average-income borrowers, provider

associations from Member States without IRR appear to be also sceptical (Figure 92). In

contrast, among provider associations from Member States with existing interest rate

restrictions there is a higher tendency (43%) to thik that a zero-impact is possible. A

tendency towards expectations of a zero-impact on average borrowers can also be

observed in the majority (54%) of the responses from other stakeholder categories.233

Discussion

According to the views of the provider associations, it is likely that the introduction of

interest rate restrictions leads to a reduction of available credit types. This would support

H2a. The credit types affected by this reduction are those which are most suitable for

low-income customers.

Issue 4: Which regulatory activities could increase the product variety?

If H2a holds, interest rate restrictions would reduce product variety. If regulators have a

choice between different measures (including interest rate restrictions) for a specific

policy goal (eg. reduce over-indebtedness, see Chapter 2.5.6, page 274), they might

take this potentially adverse effect into consideration. The following paragraphs discuss

the importance of this problem compared to adverse effects of other measures with

respect to product variety.

Facts and Figures

Figure 93 displays the results from the provider associations only: between 92% and

100% of the respondents expect either an adverse effect or no effect of all listed

regulatory measures. This effect appears most pronounced for interest rate restrictions.

In the case of an introduction of an interest rate ceiling, most of the provider associations

(86%) even assert that the effect would be strictly adverse.

233 These results are similar to stakeholders’ responses with regard to the impact of a fixed interest rate cap at

30%.

iff/ZEW – Final Report on interest rate restrictions in the EU 261

Figure 93: Effectiveness of different regulation measures with respect to wider product variety

(responses from provider associations)

12

2

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

IRR

6

5

1

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Regulations to limit rolling over

8

6

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Regulation on personal bankruptcy

2

9

1

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Tighter responsible lending requirements

7

6

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Strengthening disclosure obligations

In a separate questionnaire sent to individual providers almost all respondents indicate

that the “offering of consumer credit products” would decrease if an interest rate

restriction was introduced in their country (respectively increase if an existing interest

rate ceiling was abolished).234 From a written feedback from home credit providers in the

UK235 it appears that an interest rate restriction based on the APR would lead to a

decrease in the offering of short-term products (or even more so – exclusion from very

short-term products) and increase in the offering of long-term products (loan maturity

extension).

234 PQ Question: Assume that interest rate ceilings capping the authorised interest rate on consumer credit set

as the average market APR charged in the previous quarter times 1.33 were introduced/were abolished IN

YOUR COUNTRY. What do you think will be the effects of this measure on the various aspects of consumer

credit markets IN YOUR COUNTRY?

235 See eg. home credit lenders from UK

262 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 94: Effectiveness of different regulation measures with respect to wider product variety

(responses from consumer organisations)

7

2 2

5

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

IRR

1

4

9

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Regulations to limit rolling over

2

5

3

4

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Regulation on personal bankruptcy

4

3

6

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Tighter responsible lending requirements

1

6

2

4

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Strengthening disclosure obligations

Figure 94 shows that the consumer organisations come to similar conclusions with

respect to the effect of IRR: (44%) agree with the opinion of provider associations that

an interest rate ceiling would lead to a decrease in product variety. However, as noted in

a written comment by consumer organisations, in a longer term there still may be a

positive impact on product variety because a ceiling introduction may give rise to a

“positive product innovation”. Unlike provider associations, consumer organisations

recognise a non-negligible positive effect of regulations limiting rolling over of existing

credit.

Discussion

The answers reveal that respondents take the plausibility of H2 for granted.

iff/ZEW – Final Report on interest rate restrictions in the EU 263

2.5.4 H3: IRR lead to credit from non-bank sources, such as paying bills late

2.5.4.1 An overview

With respect to H3, our results remain inconclusive.

Looking at arrears to utility providers, we do not find evidence that households in

countries with IRR show stronger tendencies to take credit from these sources;

the data rather suggest otherwise. This contradicts H3.

Historical examples deliver mixed evidence on H3: Poland has seen a decrease in

arrears on utility bills after the introduction of interest rates, whereas there are

increases for low-income households in the Netherlands. Overall, one has to be

careful with the interpretation because of cyclical macroeconomic trends around

the event.

The Stakeholders do not report significant differences between countries with and

without IRR with respect to the level of informal lending; this again contradicts

H3.

2.5.4.2 Cross-country comparison

It is frequently argued that households need access to credit to enable them to pay their

bills even if they face negative income shocks; if they are denied credit (eg. due to

interest rate restrictions), they are assumed to make use of alternative ways of “credit”,

such as paying bills late. If H3 holds, we expect arrears on bills to be higher in countries

with IRR.

In Figure 95 countries are compared by the incidence of households who are in arrears

on utility bills. The data is taken from the EU-SILC , which is a pan-European panel

survey that collects data on living conditions, including social exclusion and poverty on an

annual basis.236 Countries with IRR are depicted in red colours, countries with no

significant restrictions are coloured in blue shades. Light colours are used to show the

overall prevalence of households in arrears, while darker shades indicate the fractions of

low-income households.237 Countries are sorted by descending prevalence of arrears in

the overall population.

236 More information on the EU-SILC is included in the Appendix.

237 Households with a monthly income below 60% of the median income are considered low-income.

264 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 95: Cross-country comparison of households in arrears on utility bills (2008)

Source: EU-SILC; Variable: “Arrears on utility bills”

It first appears that southern and eastern European countries are those in which 10% or

more of the households are late on their payment obligations. Among these countries,

there are three countries with IRR and five countries without significant restrictions in

place. There are slightly fewer countries with IRR where the percentages of households

with payment arrears are particularly high and slightly more of them at the lower end.

Therefore, the average percentage of households in arrears in countries with IRR is a

little lower than in their counterparts without significant restrictions (7% compared to 9%

for the overall population).

Discusssion

There might be many other factors influencing households’ evaluation of the importance

to pay in time in the different countries. For example there can be pre-payment meters

in use, which render late payments impossible. Also, private utility providers might

collect debt more quickly compared to public utilities. The evidence contradicts H3.

Still, this seems to be too weak evidence to refute H3.

2.5.4.3 Evidence from the Past

To study the effect of the introduction of interest rate restrictions, it is possible to

analyse the case of Poland and the Netherlands, which have already been discussed for

the evaluation of H2.238 The results are summarised in Table 61.

238 Unfortunately, data has only been available since 2005. Therefore the introduction of an interest rate

restriction in Germany cannot be studied at all.

iff/ZEW – Final Report on interest rate restrictions in the EU 265

Table 59: Past experience with arrears on utility bills

Example 1: Poland

Background

The case of Poland is described in closer detail in Chapter 1.2.1.3.8 and has been

consulted before to draw conclusions on H2. The interest rate restrictions were

introduced in 2005 and refer to the borrowing rate and not to the total cost of credit (ie.

APR). Additionally, restrictions were imposed on the fees and additional charges related

to granting a loan.

Facts and Figures

The fractions of Polish households in arrears on utility bills for the years after the

introduction of the rate ceiling 2005 to 2008 are depicted in Figure 96. The number of

households in arrears on utiliy bills follows a steady downward trend for the overall

population as well as for low-income households.

Country Year Event Observation Tendency Alternative

Explanation

Poland 2005 Introduction

IRR

Declining

fraction of

households in

arrears on

utility bills

Evidence

contradicting

H3

Improvements

in households’

financial

standing

Netherlands 2006 Legislative

Change:

Decrease of

the interest

rate cap

Constant

fraction of

households in

arrears for the

overall

population,

slight increase

among lowincome

households in

arrears

Mixed

evidence

266 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 96: Polish households in arrears on utility bills, development 2005-2008

Source: EU-SILC; Variable: “Arrears on utility bills”

Discussion

This downward trend clearly contradicts H3, which would predict the percentage of

households that are forced to rely on alternative sources of credit – such as paying utility

bills late – would increase after the introduction of an interest rate cap. Still, this

counter-evidence has to be taken with caution since this development might be due to

improvements in households’ financial standing, which can be read from high and stable

growth rates of the real GDP for the same time horizon.

Example 2: The Netherlands

Background

In the Netherlands the exisiting interest rate caps were lowered in 2006. Interest rate

restrictions therefore decreased. More detailed information can be found in Chapter

1.2.1.3.7.

Facts and Figures

Respective data for the Netherlands is depicted in Figure 97. While the fraction of the

overall population in arrears on utility bills stayed largely constant (decrease of one

percentage point) for the full time horizon, the fraction among low-income households

changed. First, the lowering of the maxium rate was followed by a decrease in the

proportion of low-income households in arrears, but this decrease was followed by an

increase in the following year which resulted in a higher level of households in arrears in

the year 2008 compared to all preceding years.

iff/ZEW – Final Report on interest rate restrictions in the EU 267

Figure 97: Dutch households in arrears on utility bills, development 2005-2008

Source: EU-SILC; Variable: “Arrears on utility bills”

Discussion

Although the unchanged level of the overall population in arrears on utility bills speaks

against H3, the increased fraction of low-income households falling into arrears might be

in favour of H3, since especially low-income borrowers might only have access to higher

cost credit, which will be affected first by the lowering of the interest rate ceiling.239

2.5.4.4 Views of Stakeholders

If H3 holds, we expect that non-bank sources of lending are more prevelant in countries

with IRR, since households have less access to official credit sources. We therefore ask

the stakeholders how important they perceive the existence of informal/community

lending activities to be.

Facts and Figures

Figure 98 documents that informal lending sources are generally regarded as credit

sources of little to some importance by the majority of the respondents. This holds true

in both Member States with IRR and Member States without IRR. In both groups of

countries, a significant proportion of responses indicate that lending from informal

sources is insignificant or non-existent, while there is only less than one-tenth of the

stakeholders assigning significant importance to informal lending.

239 This may be enfeebled by the argument that interest rate caps do not apply to some forms of high-cost

credit (eg. very short-term loans like Flitskrediet). Since these forms of credit are usually granted by nonbanks,

this is rather a further point in favour of H3.

268 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 98: Importance of community/informal lending as % of total lending activities

0% 20% 40% 60% 80% 100%

MS without

IRR

MS with IRR

Non-existent/

Insignificant

Of little significance/

Of some significance

Significant/

Very significant

SQ Question:What is the importance of community/informal lending (family, social banks, welfare)? (in

EUR/local currency and &% of total activities if possible) [non-existent/insignificant/of little

significance/of some significance/significant/very sign]

Discussion

As there are no systematic differences across countries with and countries without IRR,

the findings above are in contradiction to H3.

iff/ZEW – Final Report on interest rate restrictions in the EU 269

2.5.5 H4: IRR lead to a substantial illegal market in lending

2.5.5.1 An Overview

The results on H4 are inconclusive, in particular as there is –unlike previous studies

suggest- no reliable data to evaluate this hypothesis.

There is an influential report claiming that the level of illegal lending is higher in

countries with interest rate restrictions, which is favour of H4. However, the

empirical base of these results is controversial.

The Stakeholders expect levels of illegal lending to increase after a hypothetical

introduction of interest rate restrictions. This is in line with H4.

In contrast to this, the Stakeholders report even higher levels of importance of

illegal lending from countries without IRR, which contradicts H4.

To our knowledge, the question if and to what extent customers enter the illegal

market due to interest rate restrictions has not yet been empirically answered.

2.5.5.2 Existing literature

To describe potential differences between EU countries with and without IRR, Policis

(2004) compares the UK (as a country without interest rate caps) with France and

Germany (as countries with interest rate caps). The report claims that the rate of lowincome

households in Germany and France being in contact with illegal lenders is twice or

three times as high as in the UK. Policis (2004) interprets this as a confirmation of H4.

While Policis’ (2004) report has been frequently cited in policy discussions to illustrate

drawbacks of interest rate caps, the validity of its empirical base is controversial: the

empirical results in this study are based on a panel of poor households conducted by TNS

exclusively on behalf of Policis. nef (2009) criticises that Policis does not indicate how the

sample of about 2,700 respondents was chosen and how many people replied in the

individual countries. Furthermore, nef (2009) points out that neither the questionnaire

nor information about how the questions were asked is provided by the authors of this

study, which raises concerns about the external validity of the survey results.

Policis (2004) assumes that credit demand is uniform in all countries, irrespective of the

economic and social circumstances. nef (2009) points out that this assumption is unlikely

to hold since countries differ in terms of social inequality as well as credit institutions. In

Chapter 2.5.1.4, we demonstrate different levels of credit acceptance across countries;

to our understanding, this should also be a reason for different levels of credit demand in

different countries.

Policis (2004) also understands credit demand as entirely inelastic to interest rates. While

this might hold true for a subsection of low-income households, there may be others who

are sensitive to interest rates when formulating their credit demand. Perhaps even more

important in the context of H4, it is also questionable whether or not the demand for

credit remains constant when official credit access is denied. As with other prohibited

goods, there are two ways of responding to denied access to official credit: the illegal

market or coping without credit at all. It is perceivable that the anticipated non-existence

of official credit sources alters the behaviour of consumers in the first place: people

might help each other more within the family and the neighborhoods, or simply save

more for a rainy day.

270 iff/ZEW – Final Report on interest rate restrictions in the EU

We conclude that, from a theoretical point of view, it is possible that a reduced (official)

credit supply increases illegal lending; however, the scope of this phenomenon is an open

question and (due to the weaknesses of the empirical grounds of that study) not yet

reliably quantified by the considerations in Policis (2004).

2.5.5.3 Views of Stakeholders

Issue 1: Presence and importance of an illegal market

If H4 holds, one would expect illegal lending practices to be a more severe problem in

countries with IRR. We therefore ask the stakeholders in all countries how they evaluate

the presence of illegal lending to low-income households.

Facts and Figures

Figure 99 demonstrates that a majority of respondents acknowledges that illegal lending

may happen in their countries. Interestingly, this view is even more pronounced in

countries without IRR. In those countries, less than 30% report that illegal lending is

(quasi) non-existent, while this is reported for almost 40% in countries with IRR.

Similarly, there is a larger fraction of stakeholders from countries without interest rate

restrictons indicating that the presence of illegal lending is more significant than in

countries with IRR.

Figure 99: Presence of illegal lending

0% 20% 40% 60% 80% 100%

MS without

IRR

MS with IRR

Quasi non-existent/Nonexistent

Likely to be small/Very

small

Significant/Maybe

significant

SQ Question: How would you describe the presence of an illegal market in lending money to low-income

households? (Significant/ May be substantial/ Likely to be small/ Very small/ Quasi nonexistent/

non-existent)

Discussion

The findings above suggest that the level of illegal lending is even more

pronounced in countries without IRR. This is evidence against H4. However, the

evidence should be interpreted with caution because a comparison of Member States with

and without IRR does not depict the isolated impact of interest rate restrictions alone on

the extent of illegal lending. Countries with effective interest rate restrictions may also be

assumed to have effective institutions prohibiting illegal lending.

iff/ZEW – Final Report on interest rate restrictions in the EU 271

Issue 2: Impact of IRR on illegal money-lending

The following paragraph discusses the views of stakeholders on the introduction of

interest rate restrictions in a hypothetical country which has had no IRR beforehand.240

Facts and Figures

Figure 100: Impact of relative/floating IRR (2x market average) on illegal lending to low-income

borrowers

0 5 10 0 5 10

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market, for low-income

consumers the number of individuals resorting to the illegal credit market would…[decrease/not

change/increase]?

A majority of the respondents (80%) fears that an introduction of an interest rate ceiling

would lead to an increase in the level of illegal lending to low-income borrowers. This

opinion is held even unanimously by all responding provider associations.241

240 While we discuss the answers with respect to interest rate restrictions of twice the average interest rate, the

answers for a hypothetical introduction of interest rate caps of 30% are very similar.

241 It should be noted that several Provider Associations from UK give the report from Policis (2004) as a

reason for their assertion in written comments. Furthermore individual home credit providers and a

commercial bank from UK repeatedly refer to this report when claiming that the level of illegal lending will

rise as a result of an interest rate restriction.

272 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 101: Impact of relative/floating IRR (2x market average) on illegal lending to low-income

borrowers

0 5 10 0 5 10

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market,for low-income

consumers the number of individuals resorting to the illegal credit market would…[decrease/not

change/increase]?

While provider associations in Member States without IRR claim that illegal lending will

increase for average borrowers as well, the majority of other stakeholder categories

assert that an introduction of interest rate restrictions will not change the level of illegal

lending to the average borrower (Figure 102). The evidence does not change significantly

when considering fixed interest rate restrictions.

iff/ZEW – Final Report on interest rate restrictions in the EU 273

Figure 102: Impact of relative/floating IRR (2x market average) on illegal lending to an average

borrower

0 5 10 15 0 5 10 15

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market, for average

consumers the number of individuals resorting to the illegal credit market would…[decrease/not

change/increase]?

Discussion

According to the stakeholders’ views, H4 is plausible with respect to illegal

lending to low-income borrowers.

274 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5.6 H5: The lack of IRR leads to a higher level of over-indebtedness

2.5.6.1 An overview

Comparing evidence from different countries and taking into account the answers from

the Stakeholder Questionnaire, H5 is unlikely to hold:

Comparing different countries with respect to the prevailing level of overindebtedness

according to two different measures, it does not appear that overindebtedness

is predominantly related to the regime of interest rate restrictions.

These findings do not confirm H5.

Looking at the time after the introduction of IRR in the Netherlands and Poland,

we find weak evidence in favour of H5, ie. the indtroduction of IRR seems to

lead to a lower level of over-indebtedness. However, we argue that one should

not over-stress these findings as other cyclical factors may drive the results.

The problem of over-indebtedness is perceived almost equally strong among

stakeholders from the same category in countries with and without IRR, which

does not support H5 in general.

The Stakeholders’ views on the expected effects of a hypothetical introduction of

IRR are heterogeneous and suggest that H5 is unlikely to hold.

Stakeholders from countries without IRR predominantly expect increasing levels of

over-indebtedness after the introduction of IRR, while this is not the case for

stakeholders from countries with this kind of regulation. It appears that

stakeholders who do not experience IRR in their daily business attribute adverse

consequences to them, which are ultimately not observed in countries in which

IRR are in place.

2.5.6.2 Cross-Country comparison

If H5 holds true, the absence of IRR should reflected in statistics on over-indebtedness.

The following paragraphs therefore discuss the importance of over-indebtedness in

various countries.

Despite the growing awareness on over-indebtedness,242 there is no common definition of

the term which is accepted and applied throughout the European Union. Therefore

national data sources are not applicable for cross-country comparisons. Furthermore the

problem of heterogeneous definitions is regarded as one of the reasons why there are no

comprehensive comparable statistics that provide insights about the current state, extent

and past development of over-indebtedness in the individual EU Member States.243 To

verify hypotheses concerning over-indebtedness, a working-definition of the term was

established: Generally over-indebtedness is a state in which consumer’s current assets

and prospective income streams no longer suffice to cover the present value of his or her

payment obligations.244 The consumer is forced to cut back on his accustomed standard of

242 Compare for example the recent report from the British Department for Business, Innvoations and Skills,

BIS (2010) or the report EUC (2008) seeking for a common definition of over-indebtedness throughout the

European Union.

243 See EUC (2008).

244 Compare Betti et al. (2007), p. 140.

iff/ZEW – Final Report on interest rate restrictions in the EU 275

living in order to meet his obligations.245 In such a situation, the previously chosen

consumption path is no longer sustainable.

Based on this definition, the usability of alternative data sources was judged. In previous

studies on over-indebtedness several measures of over-indebtedness have been adopted

amongst which payment arrears and self-assessed indicators seem to be most promising

identifying over-indebted households.246

It therefore seems all the more reasonable to rely on EU-SILC (Survey on Income and

Living Conditions), which provides comparable data on all EU countries. In the following

we use a subjective indicator of over-indebtedness (households’ ability to make ends

meet) as well as an objective measure (households in arrears).

Facts and Figures: Total Arrears

Table 60 displays households in arrears in different Member States. Countries with

significant interest rate regulation in place are marked in grey shades.247 The highest

fractions of over-indebted households are found in Bulgaria, Romania and Greece. There

seems to be some kind of geographical clustering, as percentages of households in

arrears are similar for some geographically close countries (such as Greece and Romania;

Spain and Portugal; Hungary and Slovenia; Estonia and Lithuania).

Table 60: Percentages of households in arrears

(on mortgage or rent, utility bills or hire purchase), 2008

level of over-indebtedness among ...

Country all households low-income households

significant IRR

in place?

AT 7% 17% no

BE 7% 18% yes

BG 35% 49% no

CY 13% 15% no

CZ 4% 16% no

DE 6% 14% yes

DK 4% 9% no

EE 9% 13% no

ES 7% 11% no

FI 10% 20% no

FR 10% 25% yes

GR 24% 42% no

HU 16% 37% no

IE 11% 20% no

IT 16% 29% yes

LT 8% 10% no

245 Indicators of over-indebtedness based on administrative measures such as defaults rates or legally declared

bankruptcy omit people forced onto a lower consumption path.

246 For a discussion on the possible measures of over-indebtedness and their suitability to measure overindebtedness

in general and in particular for the purpouse of veryfing hypotheses by a cross-country

comparison please refer to the Appendix.

247 Since Portugal introduced regulation only recently (in 2009) we consider it to be unregulated when

analysing data from 2008, the period before the IRR became effective.

276 iff/ZEW – Final Report on interest rate restrictions in the EU

LU 2% 8% no

LV 14% 19% no

MT 8% 14% no

NL 4% 14% yes

PL 11% 23% yes

PT 6% 13% no

RO 25% 38% no

SE 6% 15% no

SI 16% 29% yes

SK 5% 12% yes

UK 5% 11% no

Source: EU-SILC. Objective indicator: “Arrears on mortgage or rent, utility bills or hire purchase”.

Figure 103 depicts this potential geographical relation. Its pattern is surprising since

there are also neighbouring countries with similar over-indebtedness problems and

different regimes concerning IRR: while France (10% over-indebted households) has a

strict regulation on interest rates, there is only weak regulation in Spain (7%). While

France and Italy both have strict regulatory regimes, over-indebtedness is a more severe

problem in Italy (16% over-indebtedness) than in France.

Figure 103: Over-indebtedness in the EU in 2008

Source: EU-SILC. Objective indicator: “Arrears on mortgage or rent, utility bills or hire purchase”. Graph: ZEW.

Discussion

Evidence stemming from households in any kind of arrears is mixed and does not support

H5. Among the strictly regulated countries are populations with high fractions of over-

5%

6% - 10%

11% - 15%

16% - 20%

> 20%

+ signifcant IRR

- Less significant IRR

iff/ZEW – Final Report on interest rate restrictions in the EU 277

indebted households (Italy) as well as very low fractions (the Netherlands). The same is

true for weakly regulated countries with the Czech Republik as an interesting example of

a lax regulatory regime and few over-indebted households as opposed to Hungary, where

regulation is weak as well but over-indebtedness poses a serious problem.

Facts and Figures: Specific types of arrears, different population groups

Instead of considering the entire set of potetential arrears of a household (credit

obligations, utility bills, mortgage or rent payments), one can also look at specific types

of arrears, which are most likely to be linked to consumer credit. However, the arrears on

credit obligations (hire purchases or loans) of EU Member States do not provide a clearer

picture (see Figure 104): the most frequent incidences of arrears on hire purchases or

loans appear in Greece (12% of all households) and Hungary (5%). But again, different

types of interest rate regimes are found in various categories of over-indebtedness due

to arrears on credit obligations.

Figure 104: Prevalence of arrears on hire purchases or loans, comparison of overall population and

low-income households in 2008

Source: EU-SILC. Objective indicator: “Arrears on hire purchases or loans”

Compared to the case of arrears on all obligations, the levels of arrears on financial

commitments differ less across countries and are generally less prevalent among the

population. Except for Greece no more than 10% of households in each Member State

report arrears on their obligation from hire purchases or loans.

It is also impossible to obtain a more clear-cut picture by regarding only low-income

households instead of the entire population (in Figure 104 fractions of low-income

households are depicted by dark blue bars).

Facts and Figures: Subjective over-indebtedness

Figure 105 and Figure 106 present a subjective measure of over-indebtedness (the

“ability to make ends meet”.248 They, again, group countries according to the prevailing

248 Although it cannot be controlled whether these households are indebted at all (a household not holding debt

that indicates its inability to make ends meet could rather be considered to “struggle on a low income”

278 iff/ZEW – Final Report on interest rate restrictions in the EU

degree of interest rate regulation. As Figure 105 indicates, the highest levels of

subjective over-indebtedness of the population appear in Greece, Romania, Portugal and

Hungary. In Greece more than half of all households experience difficulties or great

difficulties to make ends meet. In Portugal nearly one in four households has great

difficulties making ends meet. The lowest percentages of households feeling overindebted

are found in Finland, Germany, Denmark and Sweden. It appears that overindebtedness

is a less severe problem in northern- and central- European countries and

more severe in southern European countries and the New Member States.

Figure 105: Households' ability to make ends meet in 2008

Source: EU-SILC; Subjective indicator: “Ability to make ends meet”

Figure 106 focuses on low-income households in the different countries. These

households might be especially affected if interest rate regulations lead to credit

exclusion of certain customer groups. Figure 106 supports the assumption that overindebtedness

is influenced by other factors than interest rate restrictions. First of all,

percentages of households experiencing problems to make ends meet are notably higher

across all countries, while low-income households in Greece, Hungary, Portugal and

Romania still face the greatest problems making ends meet.

(Kepson and Atikons, 2006) or to be “underindebted” (Betti et al., 2007) recent research however suggest

to rely more on subjective indicators of over-indebtedness as explained in the Appendix.

iff/ZEW – Final Report on interest rate restrictions in the EU 279

Figure 106: Low-income households' ability to make ends meet in 2008

Source: EU-SILC; Subjective indicator: “Ability to make ends meet”

Discussion

It does not appear that the level of over-indebtedness (as measured by two different

proxies) is predominantly related to the regime of interest rate restrictions. This does not

reveal anything about whether a particular country would exhibit higher or lower overindebtedness

if it had a different regime of interest rate restrictions. Overall, however,

these findings do not confirm H5.

2.5.6.3 Evidence from the Past

To study the effect of the introduction of interest rate restrictions it is possible to analyse

the case of Poland and the Netherlands, which have already been discussed for the

evaluation of H2.249 The results are summarised in Table 61.

249 Again, data is only available from 2005 onwards. So the introduction of an interest rate restriction in

Germany cannot be studied.

280 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 61: Evidence from the past - over-indebtedness (H5)

Example 1: Poland

Background

The case of Poland is described in closer detail in Chapter 1.2.1.3.8 and has been

consulted to draw conclusions on H2. Interest rate restrictions were introduced in 2005

and refer to the borrowing rate and not to the total cost of credit (ie. APR). Additionally,

restrictions were imposed on the fees and additional charges related to granting a loan.

Facts and Figures

Figure 107: Polish households in arrears,

development for entire population 2005-2008

Figure 108: Polish households in arrears,

development for low-income households 2005-2008

Source: EU-SILC; Objective indicator: “Arrears on payment obligations”

Figure 107 and Figure 108 depict the development of the fraction of Polish households in

arrears on different commitments. The left hand panel shows the development for the

250 See Pruski, Żochowski (2006).

Country Year Event Observation Tendency

Alternative

Explanation

Poland 2005 Introduction

IRR

Declining

fraction of

over-indebted

households

Evidence

in favour

H5

Increased

competition amongst

credit providers,

decreasing interest

rates, improvements

in households’

financial standing250

Netherlands 2006 Legislative

Change:

Decrease of

the interest

rate cap

Declining

fraction of

over-indebted

households

Evidence

in favour

of H5

Long-term

downward trend

iff/ZEW – Final Report on interest rate restrictions in the EU 281

entire Polish population, while the right hand panel sheds light on the situation for lowincome

households. Albeit on different levels (the y-axis of Figure 108 spans double the

scale of Figure 107) the developments are similar: arrears of low-income households on

all commitments decreased by 22 percentage points (-50%) from 2005 to 2008. The

respective figure for the overall population amounts to 13 percentage points (this equals

an even higher relative decrease of -59%). Arrears on hire purchases and loans were

seldom, compared to arrears on utility bills, and have become less prevalent. The

decrease is 7 percentage points for low-income households and 5 percentage points for

the overall population. The development for arrears on mortgages or rents251 is

comparable.

As in the earlier argumentation it is worth considering households´ subjective

assessment of over-indebtedness measured by households’ ability to make ends meet.

Figure 109: Development of over-indebtedness amongst Polish households 2005-2008

Source: EU-SILC; Subjective indicator: “Ability to make ends meet”

Discussion

Although the findings from data on subjective and objective over-indebtedness support

H5, there could be other reasons for an ongoing downward trend in objective overindebtedness

and therefore a decreasing level of over-indebtedness. In Poland credit

supply has eased since 2003 due to different reasons:252 Inflation stabilised and interest

rates decreased. Furthermore, competition increased due to stagnating lending to

corporate and international banks that entered the Polish market after Poland’s accession

to the EU. The increase in lending was also influenced by demand-side factors such as

the expectations of property price increases following Poland’s accession to the EU and a

251 The European Mortgage Federation reports an owner-occupation rate of 75% for Poland for the year 2004.

The numbers stated by eurostat lie considerably higher at 95% in 2005 and 97% in 2008. Still, no clear

statement is possible on whether arrears on mortgage repayments or rents are more frequent. Arrears on

mortgage repayments remain very scarce.

252 The argumentation follows Pruski, Żochowski (2006).

282 iff/ZEW – Final Report on interest rate restrictions in the EU

more open attitude towards credit in general253. Objective and subjective overindebtedness

might therefore have decreased.

Example 2: The Netherlands

Background

In the Netherlands the exisiting interest rate caps were lowered in 2006. Interest rate

restrictions therefore decreased. More detailed information can be found in Chapter

1.2.1.3.7.

Facts and Figures

Percentages of Dutch households in arrears are displayed in Figure 110 and Figure 111.

The development does not exhibit a constant trend as in the Polish case described.

Regarding arrears on all obligations, the fraction of households in arrears is low and

decreased slightly from 4% in the years 2005-2006 to 3% in 2007-2008. The fraction of

households falling behind on their obligations on hire purchases or loans stayed constant

since 2005.

Figure 110: Dutch households in arrears,

development for entire population 2005-2008

Figure 111: Dutch households in arrears, development

for low-income households 2005-2008

Source: EU-SILC; Objective indicator: “Arrears on payment obligations”

Only considering the group of low-income households changes the picture, showing a

slight increase in the prevalence of arrears when comparing 2008 to 2005. While

percentages of households in arrears on all three groups of obligation decreased from

2005 to 2006 they became more prevalent in the consecutive years. Regarding different

obligations the prevalence of arrears on hire purchases and loans run in co-movement

with the development of arrears in mortgage repayments or rent with the difference that

percentages stabilised at a higher level compared to the year 2005 (up 1 percentage

point). The fraction of households in arrears on utility bills reached a slightly higher level,

too, (up 2 percentage points respectively) when comparing 2008 to the year 2005. Most

interesting are the numbers of low-income households in arrears on mortgages or rent.

Their fraction dropped markedly by 6 percentage points (a 50% decrease in relative

terms) in 2006, it increased in 2007 and stabilised at the level of 9%. This development

might254 be explained by mortgage market characteristics (see Figure 112).

253 See Żochowski, Zajączkowsk (2006).

254 No clear statement is possible on whether arrears on mortgage repayments or arrears on rents are more

frequently reported in this subgroup.The European Mortgage Federation reports an owner-occupation rate of

iff/ZEW – Final Report on interest rate restrictions in the EU 283

Figure 112: Development of the Dutch mortgage market and mortgage repayment arrears 2005-

2008

Datasource: Bank of the Netherlands, EU-SILC, Graph: ZEW

From 2001-2008, the Netherlands experienced an increase in housing prices of 4 to 5%

per year.255 Since 2005, net residential lending has slowed down; this decline continued

into 2008, from 36,539 loans approved in 2007 to 30,550 in 2008. As shown in Figure

112 average interest rates on mortgages were lowest in 2005 and experienced a steady

rise till 2008 (overall increase of more than 1.5 percentage points). In the same time

volumes of floating rate or short term initial fixed rate contracts went down. Interestingly

volumes of outstanding mortgages developed inversely compared to the fraction of lowincome

households in arrears on mortgage repayments or rents, peaking in 2006. This

might be an indicator that low-income households experience less repayment problems in

times of eased credit access.

Furthermore, the subjective indicator for over-indebtedness is taken into account and is

depicted in Figure 113. Once more it is not possible to preclude an ongoing downward

trend because data from earlier years is lacking. Still, the decline of perceived difficulties

in 2005-2008 was accompanied by a favourable employment environment.

Unemployment rates declined from 4.8% in 2005 to 2.8% in 2008256, with slow GDP

growth257. With the employment rates already very low in EU comparison, they are hardly

the main source of better perceptions on financial distress. The improvement does not

stem, however, from financial markets, with interest rates for household credit staying in

the range of 2005 and volumes decreasing.258

57% for the Netherlands for the year 2008. The numbers stated by Eurostat for the same year are again

considerably higher at the 68% level.

255 The European Mortgage Federation reports an owner- occupation rate of 57% for the Netherlands for 2008.

The numbers stated by Eurostat for the same year are again considerably higher at the 68% level. Still, no

clear statement is possible on whether arrears on mortgage repayments or rents are more frequent.

256 Unemployment rates according to Eurostat.

257 ECRI Statistical Package 2009.

258 De Nederlandsche Bank; ECRI Statistical Package 2009.

284 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 113: Development of over-indebtedness amongst Dutch households 2005-2008

Source: EU-SILC; Subjective indicator: “Ability to make ends meet”

Discussion

Regarding objective over-indebtedness indicators it seems most reasonable to consult

the overall (non-)development with regard to arrears on hire purchases to judge H5.

Changes in the prevalence of arrears cannot clearly be attributed to changes in the

interest rate regime.

Concerning the subjective over-indebtedness indicator, however, tells a different story:

As for the other case study countries (France, Germany, Sweden and the UK), the picture

of the development of subjective over-indebtedness is heterogeneous: subjective overindebtedness

has stayed fairly constant in France and the UK, it decreased in Germany

and Sweden from the years 2005-2007 but increased in 2008. The downward trend in

the Dutch data should therefore not be attributed to an overall Europan trend.

While there have been major changes to the credit market in Poland, the same is not

true for the Dutch credit market, so that the case of the Netherlands might constitute a

better basis to argue in favour of H5, since the level of over-indebtedness decreased

after the lowering of the interest rate cap in the Netherlands.

2.5.6.4 Views of Stakeholders

Issue 1: Is over-indebtedness a problem?

Facts and Figures

Whether or not one should consider the role of IRR for over-indebtedness obviously

depends on the relevance of over-indebtedness in a society. In our questionnaire, we

asked stakeholders if over-indebtedness is a severe problem in their respective countries.

Figure 114 reveals that the stakeholders’ views on this issue are mixed: half of the

respondents (52%) indicate that over-indebtedness is either a “severe” or a “very

iff/ZEW – Final Report on interest rate restrictions in the EU 285

severe” problem in their country. Consumer organisations attribute stronger importance

to this problem (81%) than other stakeholder categories. In a written comment, some

respondents point out that in certain countries over-indebtedness might be a small

problem on an aggregate level but a relatively severe problem on an individual level.

Figure 114: Severity of over-indebtedness, by type of respondent.

0% 20% 40% 60% 80% 100%

Provider

Associations

Consumer

Organisations

Other activity very severe

severe

small

very small

not at all

SQ Question : Do you think that private over-indebtedness is a problem in your country?

The fraction of respondents answering that over-indebtedness is a “very severe” or a

“severe” problem in their countries is approximately equally large for countries both with

and without IRR. This finding is documented in Figure 115: there is a slightly higher

proportion of respondents from MS with IRR considering over-indebtedness a severe

problem, which is related to the differences in the answers of subgroups of respondents

from consumer organisations. The majority of consumer organisations (63%) from MS

with IRR indicate a “severe” problem of over-indebtedness. In contrast, the majority of

consumer organisations in countries without IRR (60%) indicates a “very severe”

problem of over-indebtedness (Figure 116).

Figure 115: Severity of over-indebtedness, by country of origin with respect to interest rate

regulation

0% 20% 40% 60% 80% 100%

MS with IRR

MS without IRR

very severe

severe

small

very small

not at all

SQ Question: Do you think that private over-indebtedness is a problem in your country?

286 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 116: Severity of over-indebtedness, responses from consumer organisations

0% 20% 40% 60% 80% 100%

MS with IRR

MS without IRR very severe

severe

small

very small

not at all

Discussion

It appears that, overall, the problem of over-indebtedness is perceived almost equally

strong among stakeholders from the same category in countries with and without IRR,

which does not support H5 in general. However, there is a tendency that, if there are

problems of over-indebtedness, these are perceived in a more pronounced way by

consumer organisations in countries without IRR.

Issue 2: Which regulatory activities could reduce over-indebtedness?

Governments may pursue multiple strategies to try to prevent over-indebtedness. While

the evaluation of these different approaches is beyond the scope of this report, it is

crucial to determine how important various stakeholders perceive interest rate

restrictions to be compared to other potential measures.

Facts and Figures

Figure 117 and Figure 118 discuss different potential government actions directed at

reducing the level of over-indebtedness. These include interest rate restrictions (upper

left, “IRR”), regulations to limit the rolling-over of credit contracts, regulations on

personal bankruptcy, tighter responsible lending requirements as well as strengthened

disclosure obligations for lenders. As regulation affects providers and consumers

differently, we take into consideration that both groups may provide strategic answers to

these questions by looking at the answers of both subgroups separately. The charts in

Figure 117 and Figure 118 can be interpreted as follows: the bar charts represent the

number of respondents who chose a specific category. We consider one category for

those who respond that a specific regulatory measure would have some effect, a strong

effect, or a very strong effect on the reduction of over-indebtedness, another category

for those who see little effect, a third category for those who see no effect and a fourth

category for those who would even expect an adverse effect on the reduction of overindebtedness.

With respect to provider associations, Figure 117 documents controversial views on the

effectiveness of interest rate restrictions to reduce over-indebtedness: while most

respondents show support for the idea that IRR would reduce over-indebtedness to some

extent, there are two-third who claim that IRR either have no effect or even lead to an

increase in over-indebtedness. Interestingly, the majority of respondents from this

stakeholder category even assert that from the listed regulation measure, the only

alternative which might be effectively applied against over-indebtedness would be a

regulation towards limiting the rolling-over of existing credit.

iff/ZEW – Final Report on interest rate restrictions in the EU 287

Figure 117: Effectiveness of different measures to reduce over-indebtedness

(view of providers)

7

1

3 3

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

IRR

1 1 1

10

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Regulations to limit rolling over

4

8

1 1

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Regulation on personal bankruptcy

1

8

1

2

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Tighter responsible lending requirements

8

5

0 2 4 6 8 10

opposite effect

no effect

very little effect

at least some effect

Strengthening disclosure obligations

SQ Question: Which of the following regulatory activities would have the most pronounced effects on reduced

level of over-indebtedness?

Figure 118: Effectiveness of different measures to reduce over-indebtedness

(view of consumer organisations)

1

2

13

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

IRR

1

13

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Regulations to limit rolling over

2

1

11

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Regulation on personal bankruptcy

1

12 0

5

10 15

opposite effect

no effect

very little effect

at least some effect

Tighter responsible lending requirements

13

0 5 10 15

opposite effect

no effect

very little effect

at least some effect

Strengthening disclosure obligations

SQ Question: Which of the following regulatory activities would have the most pronounced effects on reduced

level of over-indebtedness?

From the consumer associations’ perspective, however, the role of interest rate

restrictions to reduce over-indebtedness is only seen in positive light: they almost

288 iff/ZEW – Final Report on interest rate restrictions in the EU

unanimously (94%) assert that IRR have a positive effect on the reduction of overindebtedness,

whereas 81% of the respondents further expect more than just a little

positive effect. Compared to provider associations, the majority of consumer

organisations even assess all listed policy measures as particularly effective instruments

against over-indebtedness.

Discussion

According to the views of consumer organisations, H5 appears plausible. The views of

provider associations on this differ, with a tendency to disagree with this hypothesis.

Provider associations see other regulatory measures to be more effective to reduce the

level of over-indebtedness.

Issue 3: Hypothetical case: effect of the introduction of IRR

The following paragraph discusses the stakeholders’ views on the introduction of IRR in a

hypothetical country which has had no IRR beforehand. For purposes of comparison, we

suggested two alternative ways of interest rate restrictions: a relative one, defined as

double the average rate prevailing in the market, and a fixed interest rate cap of 30%

p.a.

Facts and Figures

Figure 119 provides separate graphs for different respondent types (provider

associations, consumer associations and a third category, “other activity”, which includes

financial regulators, banking authorities, government officials and others). It also

distinguishes between these groups in countries with IRR and those without.

Interestingly, the expected effects of the introduction of an interest rate restriction of

twice the average rate differ substantially across countries with and without IRR. In

countries without IRR, there is very little support for the view that interest rate

restrictions could reduce the level of over-indebtedness for low-income consumers. This

view is equally pronounced for provider associations, the “other activity” category and

consumer associations. Instead, the prevailing opinion even suggests that an introduction

of interest rate restrictions would increase the level of over-indebtedness.

The responses from countries with IRR on these issues are different:259 here, a majority

of the consumer associations (44%) expect the level of over-indebtedness to decrease

after the introduction of interest rate restrictions. Also, the position of the provider

associations and the “other activity” category is more heterogenous than that of the

countries with IRR.

259 In some cases respondents from MS with IRR refer to the case of their own country instead of a

hypothetical country.

iff/ZEW – Final Report on interest rate restrictions in the EU 289

Figure 119: Effects of IRR (2x average) on over-indebtedness,

for low-income borrowers

0 5 10 0 5 10

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

was double the average of the interest rates offered on the entire credit market (ie. set at 200%

of the average calculated price for all credit products in a previous period) over-indebtedness of

the low-income consumers would… (decrease/stay same/increase)?

Figure 120 shows that the results for a hypothetical introduction of a fixed interest rate

cap of 30% APR differs from the results above. Nearly 80% of respondents argue that

the interest rate ceiling should have a nonzero impact on the level of over-indebtedness

of low-income borrowers. However, opinions are almost equally split between an increase

(41%) and a decrease (39%). The same is also true for the responses of consumer

organisations. In contrast to responses from Member States with IRR, 88% of the

responding provider associations from Member States without IRR claim that a fixed IRR

introduction would lead to an increase in the level of over-indebtedness of low-income

borrowers.

The findings in Figure 119 and Figure 120 appear to be slightly contradictory. However,

verbal comments of stakeholders in the questionnaires reveal that at the given levels the

relative (floating) ceiling and the absolute (fixed) ceiling are to a different extent binding

in the different countries. Stakeholders who perceive an interest rate restriction of “twice

the average rate” to be more binding than an interest rate cap at 30% slightly prevail.

We therefore expect the results in Figure 119 to be more informative with respect to the

views of the stakeholders on effectively binding interest rate restrictions.

290 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 120: Effects of IRR (max. of 30%) on over-indebtedness,

for low-income borrowers

0 2 4 6 8 0 2 4 6 8

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

fixed at 30% APR for all credits, over-indebtedness of the low-income consumers would…

(decrease/stay same/increase)?

Figure 121: Effects of IRR (2x average) on over-indebtedness,

for average-income borrowers

0 2 4 6 8 0 2 4 6 8

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

SQ Question: Imagine a country with no IRR at all. If IRR were introduced and effective and set at a level that

fixed at 30% APR for all credits, over-indebtedness of average consumers would… (decrease/stay

same/increase)?

iff/ZEW – Final Report on interest rate restrictions in the EU 291

While the results in Figure 119 and Figure 120 concern the role of interest restrictions on

over-indebtedness of low-income consumers, the heterogeneity of survey responses

documented in Figure 121 suggest that there is no clear effect on the over-indebtedness

of average consumers: the majority of provider associations (50%) and consumer

organisations (61%) do not expect any impact on the over-indebtedness of an average

borrower. The same is true for 48% of “other” stakeholders dominated by banking

authorities and financial regulators.

Discussion

Overall, the responses on the expected effects of a hypothetical introduction of interest

rate restrictions are heterogenous and suggest that H5 is unlikely to hold. Strikingly,

respondents from countries without IRR predominantly expect even increasing levels of

over-indebtedness after the introduction of interest rate restrictions, while this is not the

case for respondents from countries with this kind of regulation. It appears that

respondents who do not experience interest rate restrictions in their daily business

attribute adverse consequences to it, which are ultimately not observed in countries in

which interest rate restrictions are in place.

292 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5.7 H5a: The lack of IRR has particularly adverse effects on default

rates/over-indebtedness in the presence of negative shocks (eg.

recessions) to the economy

2.5.7.1 An overview

Since there are not yet enough data available covering the financial crisis no concluding

judgment can be made about the validity of H5a. Evidence is therefore inconclusive:

For the financial market crisis until 2008, cross-country comparisons do not reveal

that countries without interest rate restrictions experience higher growth rates of

over-indebtedness than countries with interest rate restictions This observation

does not support H5a. However, one has to keep in mind that the time period

for the years 2009 and 2010 is likely to be more informative, as the crisis has

only gradually affected private consumers in those countries.

To partly fill the gap of lacking external data on over-indebtedness for the most

recent years, we also consider the views of the stakeholders on recent

developments in their countries. These views reveal that there has been an

upward tendency of defaults by credit households in the time since 2008. This

effect appears to be slightly more pronounced in countries without IRR.

These findings might be seen as a confirmation of the hypothesis, although this

evidence base might be too weak to draw final conclusions. We therefore

recommend verifying these impressions at a later stage when objective (and

internationally comparable) data on over-indebtedness for those years are

available.

There is no direct evidence from the literature, since the development of overindebteness

after a negative shock has not been examined with regard to

prevailing interest rate regimes. Still, other significant factors have been found,

which might have a stronger impact on over-indebtedness than restrictions on

interest rates.

2.5.7.2 Evidence from the past / Views of Stakeholders

There is reason to assume that the financial market crisis has had an impact on the

financial situation of households and ultimately on over-indebtedness. Creditreform

(2010) report a double digit number of increases of private insolvencies in Europe

between 2008 and 2009, with particularly pronounced increases in the UK and France. As

private insolvencies are not comparable on a cross-country basis,260 it appears promising

to study whether there has been an increase in percentages of households in payment

arrears after the financial market crisis.

Unfortunately, data covering the year 2009, when the impact of the crises may be fully

mirrored in payment difficulties of households is not available. With respect to the EUSILC

data, we are therefore only able to discuss the development until 2008. Responses

from the Stakeholder Questionnaire261 will be reported to get an idea of possible recent

developments (which are not yet reflected in the data).

260 For a more detailed reasoning on that point, see the Appendix discussing measures for over-indebtedness

(in particular administrative measures).

261 In a separate questionnaire, providers were also asked: “Since the start of the financial market crisis in

2007, how would you characterize the changes to customer defaults in your institutions”. However, given

the fact that mostly respondents from the UK have answered the question, no comparison can be made and

iff/ZEW – Final Report on interest rate restrictions in the EU 293

Facts and Figures: The financial market crisis

Figure 122 presents the objective indicator “Arrears on mortgage or rent, utility bills or

hire purchase”. It depicts the percentages of households (all households) in arrears in the

six case study countries. In all countries but France the share of over-indebted

households stayed constant or decreased. It is impossible to differentiate the

development of over-indebtedness in the year 2008 on the basis of different interest rate

regimes.

Figure 122: 6-Country comparison of households in arrears 2005-2008

Source: EU-SILC; Objective Indicator: “Arrears (utility bills, mortgages, loans)”

Note: All countries ,with the exception of Poland, relate to the left hand y-axis. Countries without or with only

insignificant IRR are depicted with green shaded dotted lines (Sweden and UK) and countries with IRR

are depicted in blue shaded continuous lines except for the development of Poland which is displayed by

a dashed line.

Figure 123 takes a closer look at low-income households, as they can be presumed to be

the first to be affected by negative shocks.262 It shows the development of fractions of

low-income households falling in arrears in different countries. Not very surprisingly,

payment difficulties on different obligations (utility bills, mortgages or rent, hire

purchases or loans) are much more prevalent among low-income households compared

to the overall population. Interestingly, percentages of over-indebted households

increase in countries with IRR (with the exception of Poland) and decrease in Sweden

and the UK.

hence no conclusion can be drawn regarding the differences between Member States with and without

interest rate restrictions.

262 Eg. low-income households might have to rely on two incomes to cover their obligations. Therefore, an

increase in unemployment rate might quickly lead such households into payment difficulties as opposed to

households that can meet their obligations by the earnings of just one of the household’s members.

294 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 123: 6-Country comparison of low-income households in arrears 2005-2008

Source: EU-SILC; Objective Indicator: “Arrears (utility bills, mortgages, loans)”

As subjective burden indicators might mirror difficulties more promptly, corresponding

data from EU-SILC (indicator: “Households experiencing great difficulties to make ends

meet”) is analysed for possible effects of the financial market crises on overindebtedness.

As before, we consider all households and low-income households

separately. But again, no statements can be made on the connectedness of overindebtedness

and interest rate restrictions in the case of a negative shock comparing the

entire populations of the countries: as Figure 124 shows, there have not been notable

developments of over-indebtedness in any countries (except Poland) in the period from

2005 to 2008. There is more movement in the data for low-income households (Figure

125). In four countries low-income households more often experience great difficulties

making ends meet, which could be interpreted as an effect of the crisis: however, two of

them have interest rate restrictions while in the remaining two there are no significant

regulations on interest rates. There is thus, again, a mixed pattern.

iff/ZEW – Final Report on interest rate restrictions in the EU 295

Figure 124: 6-Country comparison of subjective over-indebtedness 2005-2008

Source: EU-SILC; Subjective Indicator: “Experiencing great difficulties to make ends meet”

Figure 125: 6-Country comparison of subjective over-indebtedness among low-income households

2005-2008

Source: EU-SILC; Subjective Indicator: “Experiencing great difficulties to make ends meet”

Discussion

The evidence from the objective indicators provided above is not in line with

H5a, as there is no observable effect (for all households) or an opposing effect (for lowincome

borrowers) for 2007/2008. Likewise, the information from the subjective

indicators points to the same direction. Summing up, the results from subjective and

objective indicators of over-indebtedness rather disprove H5a, as IRR do not seem to

have any impact on over-indebtedness in the event of a negative shock or a rather

296 iff/ZEW – Final Report on interest rate restrictions in the EU

negative effect in the sense that more low-income households reported problems keeping

up with their obligations in countries with IRR.

However, as data for 2009 and 2010 are lacking, these results are not yet conclusive. It

is therefore recommended to re-evaluate the impact of the financial market crisis and its

impacts on the real economy upon availability of EU-SILC data.

Facts and Figures: The burst of the dotcom bubble

To shed further light on H5a, the ECHP (European Community Household Panel)263 offers

data from 1994-2001 that will be used to analyse developments after the burst of the

dotcom bubble. Unfortunately, data is again not available for a longer period of time after

the negative shock.

Figure 126 shows the fraction of households holding a mortgage and behind on their

payment obligations. It is possible to grasp a slight increase in percentages of

households with repayment difficulties for the Netherlands, Spain and the UK in the year

2001. It has to be noted, though, that interest rate regimes differ among these

countries: while we consider the UK and Spain to have no effective interest rate

restrictions in place, the regulatory regime in the Netherlands is rather strict.

Interestingly, there are also countries in which the rate of households in payment arrears

on their mortgages decreased in 2001 (Denmark and France).

Figure 126: Cross-country comparison of households falling behind on their mortgage repayments

1994-2001

Source: ECHP; Objective Indicator “Households with mortgages in payment arrears”

Again, we will examine the development of arrears on mortgage repayments among lowincome

households separately (Figure 127). First, and somewhat surprisingly, arrears on

mortgage repayments are much less prevalent amongst low-income households

compared to the overall population. Concerning the analysis of low-income households’

mortgage repayment difficulties, the majority of countries exhibiting a rise in the

percentage of over-indebted low-income households are countries with IRR. However,

263 The ECHP can be considered the predecessor of EU-SILC. The discontinuity between ECHP and EU-SILC not

only concerns the years 2002-2004, which are not covered by the surveys, but also variable definitions.

iff/ZEW – Final Report on interest rate restrictions in the EU 297

development of arrears on mortgage repayments in Belgium and Spain rather seem to

follow cyclical movements with up and down-swings strengthening the aforementioned

reservation that other factors may have greater impact on repayment problems than

interest rate restrictions.

Figure 127: Cross-country comparison of low-income households falling behind on their mortgage

repayments 1994-2001

Source: ECHP; Objective Indicator “Households with mortgages in payment arrears”

¹ For Denmark there is no data for the year 2000. The development is therefore depicted only exemplarly in

tiny grey dots.

Discussion

There is again no clear pattern in the results above which argues in favour of the

validity of H5a. However, one has to be cautious using this finding as an argument in

favour of H5a: it is probable that differences in developments of arrears on mortgage

repayments are related to mortgage market characteristics and not to interest rate

regulatory regimes.

Facts and Figures: Stakeholders’ responses

As described above, the most recent data from SILC does not yet enable us to effectively

evaluate the (potentially) different impact of the financial market crisis. To fill this gap,

we consider responses from the Stakeholder Questionnaire about current developments

in the respective countries. In particular, we ask stakeholders whether there has been a

trend in consumer credit business with respect to credit defaults by private households

after the financial crisis (since 2008).

The fraction of respondents answering that over-indebtedness “increased strongly” or

“increased somewhat” in the course of the financial crisis is nearly equally large for

countries both with and without IRR. This can be taken as a sign that, in fact, the

financial market crisis has had an impact on over-indebtedness which became apparent

only in the time after 2008. It is also important to notice that there is a larger share of

respondents reporting “strong increases” in over-indebtedness from countries without

IRR. In line with this, the share of those responding that there has even been a decrease

in the level of over-indebtedness is slightly higher in countries with IRR.

298 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 128: Trend in credit defaults by private households after the financial crisis

0% 20% 40% 60% 80% 100%

MS without

IRR

MS with IRR Increased strongly

Increased somewhat

No change

Decreased somewhat

SQ Question: Did you notice a trend in consumer credit business with respect to credit defaults by private

households after the financial crisis (since 2008)

Discussion

The findings highlight the importance of looking at the financial market crisis with data

on the years 2009 and 2010, which are not yet available. We therefore recommend to

revise this analysis once the data are accessible. Although we interpret the stakeholders’

responses only as a first indication of a tendency that needs to be verified later with

objective data, we see that there appears to be a more pronounced increase in overindebtedness

from countries without IRR. This observation supports H5a.

2.5.7.3 Evidence from the literature

There are some studies and reports examining the vulnerability of households in case of

economic shocks such as interest rate increases, fuel price increases or exchange rate

changes. The findings from literature can shed light on other factors, besides IRR,

impacting the level of over-indebtedness in case of negative shocks to the economy.

Furthermore there is research examining the risk of over-indebtedness on the level of the

individual hoeshold into which an insight will be given at the end of the paragraph.

Negative shocks to the economy

For the Netherlands, Brouwer (2008) expresses the concern that, since the loan-to-value

ratios are very high,264 a downturn in the economy may have serious consequences for

borrowers’ ability to meet their payment obligations. While the average LTV ratio on

outstanding loans amounts to over 40%, for new contracts it is often higher than

110%.265 This might be a particular concern, as many new borrowers are first-time house

buyers, with relatively low incomes, low job security and in many cases dependence on

two incomes. They are therefore particularly vulnerable.

264 Strictly speaking loan to foreclosure values (LTFV) are high. The foreclosure values refer to the estimated

sum that can be realised in a forced sale and usually amount to 85%-95% of the market value of the

property. As LTFV can reach up to 125%, the LTV can be calculated to range between 85-115%. Compare

Standard & Poors (2005). High loan-to-value ratios are motivated by tax deductibility of mortgage interest

payments.

265 Furthermore, tax deductibility leads borrowers to maximise the outstanding principal balance on their

mortgages for as long as possible, thereby increasing the tax benefits over the duration of the loan.

Compare Standard & Poors (2005).

iff/ZEW – Final Report on interest rate restrictions in the EU 299

Although the level of aggregate indebtedness in Poland is low compared to other EU

countries, Zajączkowski and Żochowski (2006) express the concern that Polish

households may still be more vulnerable to shocks to the economy because of the higher

share of basic living costs in total consumption expenditure. In Poland, new loans, and

especially mortgages, are often denoted in Swiss francs.266 Furthermore, mortgage debtservice

ratios in lower-income groups are increasing. Zajączkowski and Żochowski (2006)

therefore analyse the effects of a foreign exchange or interest rate shock. They hold the

opinion that the debt-service ratio level, at which one of the risks may prove crucial for

the financial system´s stability, is lower than in EU 15 as the proportion of fixed

expenditure in the budgets of Polish households is higher. The safety income buffer

within which the debt may freely rise as a result of different shocks without a risk to

stopping debt repayment is therefore lower in EU 15 countries.

Johansson and Persson (2006) study the possible consequences of increases in interest

rates or unemployment rates households´ability to pay in Sweden. According to their

analysis of 5 different income quintiles, debt is backed by sufficient collateral in all

quintiles. They find the most vulnerable households in the lowest income group. Although

this income quintile holds only 2% of total household debt, more than half of the losses

incurring in case of default would stem from this category.267 They find the effect caused

by an interest rate hike to be stronger compared to an only very moderate effect of rising

unemployment. They explain their findings with the fact that large amounts of debt are

concentrated in the highest income quintile where households are presumed to have dual

incomes and high unemployment benefits.

For the UK Kempson and Atkinson (2006) investigate the impact of two areas of

expenditure increase – fuel bills and mortgages - on households’ ability to make ends

meet. For the expenditure on fuel bills they find that price increases have the greatest

impact on poor people and those aged over 70, but that they do not have a major effect

on most households’ ability to make ends meet. Regarding mortgages, concerns have

been raised questioning the ability of some mortgagors to keep up with repayments

should interest rates rise against the background of increases in the sums of money that

have been borrowed. In contrast, to fuel bills, mortgages absorb a far higher proportion

of household income (on average 20% of net monthly household income, at the time of

the baseline survey of financial capability in the year 2005, while fuel bills absorb only an

average of 4.1% of net household income). Kempson and Atkinson (2006) assess the

impact of interest rate increases considering a subjective indicator as well as a statistical

analysis. They show that far fewer people would actually experience payment difficulties

than expected. While more than 20% of mortgagors anticipated financial difficulties

following a rise in monthly payment obligations by 10%, according to the statistical

analysis only one in a hundred seemed likely to experience these difficulties.

Their findings are consistent with previous research which has shown that, across the

population as a whole, increases in expenditure are not a major cause of financial

difficulties. However, price increases have been found to be a contributory factor (Ford,

Kempson and Wilson, 1995; Herbert and Kempson, 1995).

266 According to Zajączkowski and Żochowski (2006) the majority of new loans (70% - 90%) are loans

denominated in foreign currencies, usually in Swiss francs. Housing loans are usually extended at a floating

interest rate, where fixed interest rate is generally used for a short promotional period – up to two years

from the date of loan contract. Brown and al. (2009) find that while foreign currency residential housing

loans accounted for up to 60% of all foreign currency loans they are much less widespread in consumer

lending and corporate finance. In 2007, less than 6% of CHF lending was ascribed to consumer credit and

corporate credit.

267 Johansson and Persson (2006) express the suspicion, that households in this income category may have

incomes and assets not reported to the tax authority providing the analysed data.

300 iff/ZEW – Final Report on interest rate restrictions in the EU

Creditreform (2009) judge the risk of over-indebtedness to be lowest for German

households, as compared to their US or UK counterparts. They base their analysis on

different economic indicators such as unemployment rate, GDP growth and the lower

ratio of consumer credit to GDP. Creditreform (2009) underline that the default rate on

consumer credit in Germany amounts to only half of the rate in the US or the UK – 3%

versus 6% respectively. Schufa (2010) find the default rate in Germany to be fairly

constant for the years 2007-2009. Furthermore, they publish an index assessing the risk

of over-indebtedness and predicting its future development. This index (called PVI –

Private Indebtedness Index) showed a slight decrease from 2008 to 2009, signaling a

decrease in critical signs of over-indebtedness268 but for 2010, an increase is expected.

The development of PVI growth rates is depicted in Figure 129.

Figure 129: Growth rates of PVI (Germany) 2004-2009

Source: Schufa.

Negative Shocks on the level of the individual household

Theoretically, it is major events which force households into over-indebtedness. Such

circumstances may lead households into an economically destabilised situation, either

because of lower household income or higher need for household expenses. However, in

practice there are other reasons for over-indebtedness such as over-commitment or bad

management of households’ finances. A great part of the analysis of potential causes

relies on self-reported reasons and may therefore be biased.

For Germany there is data reported by debt counselling agencies. Although this data is

not representative, it can give some insights into which triggers might have lead

households into over-indebtedness, according to the assessment of counsellors. In the

view of counsellors, the main triggers are unemployment or an income reduction of

another type (30%), family breakdowns (13%), illness (7%), avoidable behaviour such

as unsustainable consumption behaviour or bad housekeeping (17%), and failed

attempts of self-employment (12%).269

According to Kempson (2002), a loss of income is the most frequently stated reason

(45% of households) by people who have fallen into arrears in the previous 12 months or

268 Such as personal insolvency (Privatinsolvenz).

269 See Statistische Bundesamt (2009).

iff/ZEW – Final Report on interest rate restrictions in the EU 301

who are experiencing financial difficulties in the UK. This loss of income was often a

result of dismissal. The second and third reasons were relationship breakdowns and longterm

sickness. This result is confirmed by a later analysis which finds drops in income to

be strongly associated with the risk of arrears.270 Low income as such was reported by

14% of households in the study of Kempson (2002). Over-commitment, overlooking or

withholding payments were only mentioned in around one tenth of the cases. As some

people prefer to give reasons which relate to external factors, rather than admitting that

they are personally responsible for their indebtedness,271 Kempson et al. (2004) expect

that over-commitment and overlooked or withheld payments account for a larger fraction

of actual reasons for debt problems. Trying to evaluate which events significantly affect

self-assessed over-indebtedness, Disney et al. (2008) identify divorce and a decrease in

income as the major changes in personal circumstances which induce a household’s

perception to feel over-indebted. In another analysis they try to determine changes

causing households to fall into arrears. Among them are previous relationship

breakdowns and an increase in the number of credit commitments. Taking a closer look

at the effect of family dynamics, Kempson et al. (2004) find that family changes have a

greater impact on the likelihood of arrears on unsecured credit commitments than they

do on household bills. Interestingly, changes in current income and loss of employment

are not found to have a direct impact on a change in arrears. Disney et al. (2008) argue

that the ‘life cycle’ model does not suggest that income changes always affect

indebtedness, but that it is adverse shocks to income, possibly coupled with high debt

commitments, which cause problems.

Jentzsch and Riestra (2006) examine reasons for repayment difficulties in Austria,

Belgium, France and Spain. Unemployment is blamed for repayment difficulties in all

countries, with a fraction as large as 42% in France to a relatively low 21% in Austria

and 19% in Belgium. Divorce is also a widespread factor, mentioned by a fifth of

Austrian, Belgian and French households. In Spain 58% give a reduction in income (due

to unemployment, divorce or other factors) as reason for a default on payment

obligations. Interestingly, the reason most commonly reported in Austria is poor

household management. Similarly, 12% of Spanish households admit bad financial

management to have caused their payment problems. Excessive charges or unexpected

charges (as mentioned by 16% and 3% of Belgian households, respectively) may as well

be indicators of poor financial management, as can be the lack of information reported by

26% of Spanish debtors.

Conclusion/Discussion

Literature finds that different factors influence the risk of over-indebtedness on the

macro and the micro level: Most cited are a rise in unemployment and changes in the

family structure. Even so, there are contradicting opinions on which of the factors has the

highest impact. This may be due to differences in credit markets. Literature shows that

there is a range of factors besides interest rate restrictions which can lead to adverse

effects on default rates/over-indebtedness in case of negative shocks to the economy.

270 See Kempson et al. (2004), p. 32.

271 Compare Dominy and Kempson (2003).

302 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5.8 H6: The average consumer (or even more so: low-risk consumer)

would be granted cheaper credit in the presence of IRR

2.5.8.1 An Overview

A necessary condition for H6 to hold would be a decline in the average interest rate after

the introduction of restrictions. Since in general no data on individual credit cost before

and after the introduction of IRR are available, aggregate data are taken into account.

Drawing on these data and findings from the Stakeholder Questionnaire, we find

inconclusive results with respect to H6.

The cross-country comparison of interest rates is difficult due to differing loan

values and maturities of the loans as well as differing market structures, taxation

and the cost of funding by lenders. A cross-country comparison of the case

study countries does not deliver patterns which can be explained by

interest rate regulation.

Based on the data on interest rates presented above we find some evidence in

favour of the decreased average rates immediately after the introduction of

interest rate restrictions in Poland. The observations from the year 2006/2007 in

the Netherlands suggest that one cannot expect a decline in average interest

rates on instalment loans after lowering interest rate restrictions, as H6 implies.

The German example in the 1980s suggests that interest rate restrictions may

lead to a lower degree of heterogeneity of charged interest rates: this means

that the most expensive loans offered are closer to the average interest rates

than before; this does not, however, necessarily confirm H6.

Stakeholders report their view that interest rate restrictions are likely to make

credit more expensive than without; this is the opposite of H6.

2.5.8.2 Introduction

When banks apply risk-adjusted pricing, there is a broad range of different interest rates

in the market. In theory, interest rate caps may reduce the scope of risk-adjustment,

potentially driving higher-risk groups out of the market (see Chapter 2.5.1). As a

statistical consequence, the average interest rate after the introduction of interest rate

restrictions is lower. This does not necessarily imply that an individual borrower (who

remains in the market) gets better rates as before. In addition, as discussed in Chapter

2.1.3, there are also theoretical arguments that credit should be cheaper for the

custumers remaining in the market.

Unfortunately, data limitations do not permit to distinguish whether a drop in average

rates is due to the statistical effect, or whether an individual borrower ultimately faces

lower credit costs.272 To evaluate the hypothesis (which has the latter effect at its core),

we thus consider the drop of average rates to be necessary (but not sufficient) condition

for the validity of the hypothesis.

2.5.8.3 Cross-country comparison

Interest rates are difficult to compare across countries. Differing loan values and

maturities of the loans as well as differing market structures and taxation of interest

272 This would only be possible based on micro data which records not only the credit positions of households,

but also the prevailing credit conditions before and after the introduction of interest rate restrictions.

iff/ZEW – Final Report on interest rate restrictions in the EU 303

rates between countries make it very difficult if not impossible to draw conclusions with

regard to relative interest rate levels (Guardia, 2002). Casolaro et al. (2006) briefly

describe dynamics of interest rates in Italy. They find a continuous reduction in the level

of both nominal and real interest rates in Italy starting in 1995, which they attribute to

the decrease in the interest realised to achieve Maastricht criteria for admission to the

single currency, worldwide reduction in the interest rates during the 1990s, as well as

financial liberalisation. Casolaro et al. (2006) also find a decline in the spread in the longterm

interest rates vis-à-vis average interest rates on long-term bank funding from 5%

in 1996 to 2% in 2000. The authors state that the decline may be attributed to the

increased completion brought about by the financial liberalisation. Except for Hartmann-

Wendels and Spörk (2008), we are not aware of studies analysing consumer credit

interest rates in the six case study countries. However, Hartmann-Wendels and Spörk

(2008) focus on German consumer credit interest rates only and their discussion is

centered on comparability of the interes rates reported after the introduction of the EMU

guidelines for interest rate statistics in 2003.

Below we illustrate the dynamics of the consumer credit interest rates in the selected six

EU countries. Figure 130 presents real interest rates on consumer credit with initial

fixation of up to 1 year. Due to lack of data for Sweden, Poland and the UK in the early

2000s, the time period for which data for all six countries are available is limited to

August 2005 to January 2010. As we can see from Figure 130, for most of this period,

the highest real interest rates are observed in Poland while the lowest interest rates are

found in Sweden. Interest rates in France, Germany and the Netherlands remained below

those in the UK until the period of the recent financial turmoil. French, German and

Dutch interest rates exceeded those of the UK in 2007, 2008 and 2009, respectively.

Notably, after the start of the recent financial crisis in 2007, the interest rates on

consumer credit declined somewhat in all countries except for Germany.

Figure 130: Real interest rates on consumer credit with initial fixation of up to 1 year in the six EU

countries, 2003-2010, in percent.

-6

-4

-2

0

2

4

6

8

10

12

2003-01 2004-01 2005-01 2006-01 2007-01 2008-01 2009-01 2010-01

France Netherlands Poland UK Sweden Germany

Source: National Central Banks, Eurostat, own calculations. Notes: Interest rates are for new business, except

for Sweden.

304 iff/ZEW – Final Report on interest rate restrictions in the EU

A similar picture is observed in case of interest rates with an initial fixation of more than

1 and less than five years (Figure 131). The highest and lowest interest rates were

observed in Poland and Sweden until 2007 and in the Netherlands since 2007. German

and French interest rates remained well below those of the UK for the whole period.

Figure 131: Real interest rates on consumer credit with an initial rate fixation of more than 1 year

and less then 5 years in the six EU countries, 2003-2010, in percent.

0

2

4

6

8

10

12

14

2003-01 2004-01 2005-01 2006-01 2007-01 2008-01 2009-01 2010-01

Germany France Netherlands Poland Sweden UK

Source: National Central Banks, Eurostat, own calculations. Notes: Interest rates are for new business, except

for Sweden.

Finally, Figure 132 shows interest rates with the initial fixation of more than five years.

As in the case with the two previous interest rates, Poland showed consistently higher

interest rates than other countries. UK interest rates are similar to those of Germany and

Swedish interest rates are somewhat below those of France. The lowest interest rates for

this type of credit are observed in the Netherlands.

iff/ZEW – Final Report on interest rate restrictions in the EU 305

Figure 132: Real interest rates on consumer credit with an initial rate fixation of more than 5 years

in the six EU countries, 2003-2010, in percent.

0

2

4

6

8

10

12

14

16

2003-01 2004-01 2005-01 2006-01 2007-01 2008-01 2009-01 2010-01

Germany France Netherlands Poland Sweden UK

Source: National Central Banks, Eurostat, own calculations.

Notes: Interest rates are for new business, except for Sweden.

Discussion

Based on the facts above, we conclude that the evidence with respect to the

interest rate levels in the countries with and without IRR is mixed. Two issues

should be borne in mind. First, the case of Poland is special since this is the only new

Member State in the group of the considered six EU countries. As a country with a

relatively new and fairly concentrated banking system, where competition may still

remain in a nascent state, it is not surprising that relatively high interest rates are

observed. Second, the recent financial crisis resulted in a decline in interest rates

internationally which means that the observed figures for 2007-2009 reflect not only the

interplay between demand and supply factors, but also broader macroeconomic

situation.273 Finally, as interest rates may be affected by a number of the abovementioned

factors, it is difficult to identify whether and to what extend the observed

changes result from differences in interest rate regulation. Therefore observations

presented here should be treated with caution.

2.5.8.4 Past experience

In this section we provide an illustration of the development of the levels of

interest rates for consumer credit following the introduction of interest rate

restrictions in Poland in 2005 and the change in interest rate regulations in the

Netherlands in 2006. A brief summary of the results is given in Table 62:

Evidence from the past.

273 In addition to real interest rates we also considered spreads between interest rates and rates on

government bonds (as a proxy for bank cost of funding). The results are largely the same.

306 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 62: Evidence from the past – average prices (H6)

Country Year Event Trend in interest rates

one year after change

Summary

Poland 2005 Introductio

n of IRR

Decrease in three out of

four considered rates

Evidence in favour of

H6

Netherlands 2006 Lowering of

the legal

allowed

rate

No decrease in two out

of three rates

Evidence against H6

Furthermore, the introduction of interest rate restrictions in Germany in the years after

1981 allows an assessment of the development of the average interest rate compared to

the 5% most expensive interest rates at each point in time. This investigation reveals

that the heterogeneity of consumer credit with respect to the offered interest rate

declines after the introduction of interest rate restrictions.

Example 1: Poland274

Facts and Figures

As Figure 133 shows, during 2006, the year after the introduction of interest rate

restrictions in Poland, we do not observe an increase in interest rates. In fact, average

interest rates in 2006 were slightly lower than those of 2005, with the exception of the

interest on consumer loans with an initial rate fixation of over three months and up to 1

year. Interest rates on loans with an initial fixation of more than 1 year and less than 5

years decreased by 1.12 percentage points; loans with an initial period of fixation of more

than five years decreased by 0.04 percentage points. APRC showed the most dramatic

decline of 2.51 percentage points. It should be pointed out, however, that this decline in

interest rates observed in 2006 could be a continuation of the downward trend in

consumer credit interest rates that has been observed since at least 2002 (Pruski and

Zochowski, 2006).

274 The background of the event, here for Poland as well as later for the Netherlands and Germany, is described

in Section 2.5.2.5 on page 241ff, where the events are also studied with respect to the influence on the

development of credit volume.

iff/ZEW – Final Report on interest rate restrictions in the EU 307

Figure 133: Selected consumer credit interest rates in Poland, 2005-2010, in percent.

0

5

10

15

20

25

Jan 05

Jan 06

Jan 07

Jan 08

Jan 09

Jan 10

Variable rate and up to 3 months initial rate fixation Over 3 months and up to 1 year initial rate fixation

Over 1 and up to 5 years initial rate fixation Over 5 years initial rate fixation

APRC

before after

Source: National Bank of Poland, www.nbp.pl

Note: interest rates provided are agreed annual rates (AAR), unless indicated otherwise, in real terms. Loans

are denominated in Polish Zloty, new business.

Discussion

Based on the data on interest rates presented above we find some evidence in favour of

the decreased average rates during the immediate aftermath of the introduction of

interest rate restrictions in Poland. However, it should be pointed out that since no

adjustment for the bank cost of capital is done, it is difficult to deduce to what extent this

decline might be driven by the prevailing macroeconomic climate.

Example 2: The Netherlands

Facts and Figures

Figure 134 depicts real interest rates on consumer credit loans with initial period of

fixation of less than 1 year, more than one year and less than five years and more than

five years. As we can see, interest rates have increased slightly between 2006 and 2007.

Interest rates on loans with an initial rate fixation of less than one year increased by 0.9

percentage points, rates on loans with an initial rate fixation of more than five years

increased by 0.30 percentage points. Interest rates on loans with more than one and less

than five years initial fixation declined by 0.6 percentage points during this period.

308 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 134: Selected consumer credit real interest rates in the Netherlands, 2003-2010

0

2

4

6

8

10

12

14

16

2003-01 2004-01 2005-01 2006-01 2007-01 2008-01 2009-01 2010-01

Instalment loans: duration 1 year Instalment loans: duration 1-5 years

Instalment loans: duration >5 years IR Ceiling (after 2006)

before IRR after IRR

Source: http://www.statistics.dnb.nl/index.cgi?lang=uk&todo=Rentes

Note: interest rates provided are agreed annual rates (AAR), in real terms. Loans are in Euro, stocks. The

interest rate ceiling of 15% presented on the graph is the ceiling which came into effect in 2006 (see Chapter

1.2.1.3.8 for details). An earlier, higher ceiling which existed before 2006 is not depicted.

Discussion

These observations suggest that one cannot expect a decline in average interest rates on

instalment loans after lowering interest rate restrictions, as H6 implies. While these

interest rates do not cover the entire consumer credit market (as they apply to instalment

credit), this finding is still important for an important fraction of the consumer credit

market.

Example 3: Germany

Unlike the considered events in the Netherlands and Poland, where interest rate caps

were introduced or tightened in a single-point-in-time legislative decision, the

introduction of interest rate caps in Germany took place as a direct consequence from a

sequence of judical decisions starting in 1981. As this took time to become effective, it is

impossible to identify only one “relevant” year to look at. Instead, we will consider how

the spread between the average and the most expensive credit contracts evolve in the

years after the introduction of interest rate restrictions compared to the situation before.

Facts and figures

The following graphs depict effective annual interest rates on instalment credit (excluding

mortgages, new business) from March 1968 to June 2003.275 The legal cap rate was

calculated as two times the current average interest rate.

275 The data is taken from “Erhebung über Soll- und Habenzinsen” by Deutsche Bundesbank, which was

suspended in 2003. The unweighted average interest rate was calculated after excluding the highest and the

lowest 5th percentile of interest rates. The value of the surveyed contracts changed twice during the

surveyed period: from 2000 to 5000 DM in 1968 and from 5000 to 15000 Euro in 2003.

iff/ZEW – Final Report on interest rate restrictions in the EU 309

Since effective annual interest rates are only available from June 1986, we estimated

them from monthly interest rates where necessary.276 As can be seen from Figure 135,

our estimate of the interest rate ceiling lies slightly below the ceiling provided by

Bundesbank.277 The difference is relatively constant and rather small at 0.3 percent.

Figure 135: Instalment consumer credit interest rates in Germany, average cap and bond yields (in

percent), 1968-2003.

0

8

16

24

32

40

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

00

01

02

03

Average legal cap legal cap - own calculation 4 yrs. bond yield

before after

Source: www.bundesbank.de, own calculations. Interest rates used are effective annual interest rates (new

business).

Note: “Average” stands for unweighted average interest rates calculated after excluding the bottom and the top

5th percentile of interest rates. “Legal cap” stands for interest rate ceiling as provided by the

Bundesbank. “Legal cap – own calculation” stands for the interest rate ceiling based on own

calculations. 4 yrs. bond yield reflect the yield of a government security.

Figure 135 shows substantial fluctuations in average interest rates on instalment

consumer credit over time. Most of these fluctuations originate from the general level of

interest rates (which directly affects refinancing costs of lenders), as captured by the line

“4yrs bond yield”. Note that the spread between the average instalment interest rate and

the bond yield had already widened before the introduction of interest rate restrictions in

1981.

The German data not only includes the average rate of instalment credits, but also the

level of the 5% lowest and highest interest rates, respectively. Accordingly, Figure 136

allows a comparison of the interest rates of the 5% most expensive credits with double

the average interest rate, which has been defined as interest rate restriction by court

decisions after 1981.

276 Until 1986, interest rates were only given as monthly rates on the initially borrowed amount. In the time

period under consideration, the conversion of these monthly rates to effective annualised rates was mostly

carried out using calculation tables. Interest rate conversion tables recommended by the “Bund-Länder-

Ausschuss” were used in calculations. Based on Bundesbank data, we assumed a contract fee of 2.1%. See

also Sievi et al (1980). To compute the effective rates, one requires information about the maturity of the

considered contracts. Unfortunately, the surveyed maturities changed over time, starting from an interval of

12 and 24 months in 1968 and ending with an interval of 36 and 60 months in 2003. Sievi, F., G. H., and C.

R. Sievi, 1980. Effektivzinssätze für Ratenkredite mit monatlichen Raten(Gillardon-Verlag, Bretten). For the

sake of comparability, we assume alternative average maturities of 24 and 48 months, respectively. The

results of both specifications are qualitatively similar.

277 This graph reflects the findings obtained when assuming an average maturity of 48 months. Assuming an

average maturity of 24 months, the approximation is even closer.

310 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 136: Instalment consumer credit interest rates in Germany, in percent, 1968-2003.

0

8

16

24

32

40

6 8

6 9

7 0

7 1

7 2

7 3

7 4

7 5

7 6

7 7

7 8

7 9

8 0

8 1

8 2

8 3

8 4

8 5

8 6

8 7

8 8

8 9

9 0

9 1

9 2

9 3

9 4

9 5

9 6

9 7

9 8

9 9

0 0

0 1

0 2

0 3

Average lower bound upper bound legal cap - own calculation

before after

Source: www.bundesbank.de, own calculations. Interest rates used are effective annual interest rates

(new business).

Note: “Average” stands for unweighted average interest rates calculated after excluding the bottom

and the top 5th percentile of interest rates. “Lower bound” and “Upper bound” stand for the bottom

5th and the top 5th percentile of interest rates, respectively. “Legal cap – own calculation” stands for

the interest rate ceiling based on own calculations.

It is striking that, after 1981, the 5 percent most expensive credits have constantly been

substantially lower than the interest rate ceiling. Before the introduction of the rate

caps, however, the upper bound of interest rates was at some points in time (in

particular in time periods with lower average levels) very close to double the average

rate. This implies that before the introduction of interest rate restrictions, the

heterogeneity among credit interest rates was higher than after. Figure 137 points to

the same direction: as we can see, the “upper bound - average” spread, which is

calculated as a difference between the 5 percent most expensive credits and the average

interest rates, is substantially lower in the post-1981 than in the pre-1981 trend.

Figure 137: Consumer credit interest rates in Germany, spreads and interest rate cap, in percent,

1968-2003.

0

4

8

12

16

20

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

00

01

02

03

upper bound - average average

before after

Source: www.bundesbank.de, own calculations. Interest rates used are effective annual interest rates

(new business).

iff/ZEW – Final Report on interest rate restrictions in the EU 311

Note: “Upper bound – average” stands for a spread calculated as a difference between 95% percentile

of interest rates and the average interest rates. “average” stands for the average interest rate.

Discussion

We are not able to find a clear pattern with respect to H6: we demonstrate that the

heterogeneity of credit contracts (with respect to their interest rate) was reduced after

1981. This could have various alternative reasons:

More homogeneity in credit contracts could imply less risk-adjusted pricing and

less particular (eg. small amount) credit. This would exclude specific consumer

groups and confirms H1.

IRR could bring the level of interests charged by, eg. local monopolists down to a

“more competitive” level. This would be to the benefit of their debtors, which could

make a point in favour of H6.

Providers could have abandoned the interest rate as an instrument to differentiate

between different consumer types, and could have introduced alternative sources

of revenue which are not captured by IRR. These sources could, for example,

include insurances. This strategy would imply that lenders do not necessarily

exclude specific consumer groups, but charge their margin by other means than

interest.

2.5.8.5 Views of Stakeholders

Due to the problems to identify the causal effect of interest rate restrictions on the price

of credit for consumers, we also collect the stakeholders’ views on this issue.

Figure 138 demonstrates this view with respect to credit for average borrowers: A large

majority of stakeholders (95%) does not expect the cost of credit for the average

borrower to decrease, which contradicts hypothesis H6. In contrast, 77% of the

responding provider associations further claim that interest rate restrictions would

definitely lead to an increase in the costs of credit even to the average borrower.

Figure 138: Impact of a floating IRR on the cost of credit for the average borrower

0 2 4 6 8 0 2 4 6 8

decrease

not change

increase

decrease

not change

increase

decrease

not change

increase

no IRR, Provider Associations no IRR, Consumer Organisations no IRR, Other Activity

MS with IRR, Provider Associations MS with IRR, Consumer Organisations MS with IRR, Other Activity

Source: SQ Question: Imagine a country with no IRR at all. If interest rate restrictions were introduced and

effective and set at a level that was double the average of the interest rates offered on the entire

credit market for average consumers cost of credit would…[decrease/not change/increase]?

312 iff/ZEW – Final Report on interest rate restrictions in the EU

In line with these results, the majority of provider associations (77%) expect an

introduction of interest rate restrictions to be largely ineffective in reducing cost of credit,

leading to an increase in the latter instead (see Figure 138). The majority of respondents

further considers all listed alternative regulation measures as ineffective or even having

the opposite effect with regard to a reduction in the costs of credit. Notably, consumer

organisations show inconsistencies in their abovementioned opinion by unanimously

claiming that interest rate restrictions will have more than just a little positive effect in

reducing costs of credit.

Discussion

Stakeholders express their view that H6 does not hold, since, in their opinion, interest

rates will either remain constant or increase.

iff/ZEW – Final Report on interest rate restrictions in the EU 313

2.5.9 H7: IRR lead to increased charges as providers will try to compensate

the reduced interest revenues by increased charges

2.5.9.1 An Overview

H7 is plausible and confirmed by several examples in selected countries. Still, one

should note that there are no comprehensive data available.

There is a variety of additional charges which are generally not affected by

existing interest rate regulations. Furthermore, complementary agreements such

as life insurances until credit maturity can be applied to compensate for

decreasing interest revenues caused by the interest rate regulation.

Interest rate restrictions are expected to increase the price complexity of credit

products.

Experience from the UK, Germany and Poland confirms the expectations that a

ceiling on interest rates or on specific fee components will lead to an introduction

of new charges and fees by credit providers.

Stakeholders who favour the introduction of interest rates restrictions hold the

opinion that charges and fees should be regulated as well.

It is obvious that the degree to which H7 has relevant consequences depends on

the way the interest rate restrictions are forumulated: eg. an interest rate cap

based on APR covers more cost components than an interest rate cap based on an

interest rate (without taking account of fees).

2.5.9.2 Introduction

Existing theoretical literature suggests that, depending on the price elasticity of credit

demand, interest rate regulation would lead to an increase in or introduction of charges,

not captured by the interest rate cap, or to the signing of additional agreements in

connection to the credit. This would imply that, credit contracts are more complicated in

countries with IRR, as these need to incorporate more elements (charges,

complementary agreements) than those captured by the regulation on interest rates. The

increased price complexity would damp down the decrease in credit supply and thus an

expected credit exclusion of low-income borrowers (according to H1) as well as a

decrease in volumes of credit granted (according to H2).

There are different types of charges which are not captured by the interest rate

regulation in most of the Member States:

upfront interest payments,

fees for issuing a card (Białowolski 2009),

current account fees (IFF Research 1998, Białowolski 2009, DG SANCO 2009),

charges attached to specific payment behaviour such as:

o late payment fees on credit cards (OFT 2006, Białowolski 2009),

o unarranged overdraft charges (OFT 2010b),

o low-activity fees (PwC 2008),

314 iff/ZEW – Final Report on interest rate restrictions in the EU

o connected fees for issuing notices for delayed payments or low activity

(Białowolski 2009).

Complementary agreements include

additional insurances such as life insurances,

payment protection insurances (IFF Research 1998),

savings agreements capturing repayment amounts of the credit (IFF Research

1998),

currency swaps (Białowolski 2009),

etc.

It is apparent that circumvention is more difficult when interest rate ceilings are applied to the

APR rather than an interest rate excluding fees and charges. Although charging of non-mandatory

costs (those not included in the APR such as voluntary insurance) can still take place, the problem of

circumvention is of concern primarily to high cost credits like home credit where seperate costs can be

identified and extracted from the interest rate ceiling and justified by associating the additional fee

with an additional service (such as home collection, direct debit).

2.5.9.3 Literature Review

To our knowledge, there is no comprehensive data on the change in pricing behaviour of

suppliers as a response to the introduction of interest rate restrictions. However, there

are some cases which underline the potential importance of the hypothesis.

In Germany, an alternative to upfront interest payment has been documented by IFF

Research (1998): a zero interest credit provided by banks owned by automobile

companies. Thereby, although no explicit upfront payment is collected, consumers

granted a zero interest credit do not benefit from discounts made to cash-paying

customers. This strategy is also a way to circumvent interest rate restrictions, as

the interest payments are not declared as such; rather, they formally are a component of

the price charged on a product.

Since the EU regulation on the APRC has been implemented in Germany also for usury

ceilings after 1998 (Art. 1A CCD), banks have increasingly cross sold insurance products

with hidden bank provisions. The extra price for these insurances lead to credit contracts

with costs beyond the interest rate ceiling. With these practices lately accepted by OLG

Hamburg Dec. 12.5.2010 13 U 21/09, the rate ceiling was de facto lifted and

gradually circumvented in practice. This evidence also speaks in favour of H7. In this

context, interest rate ceilings may further set an incentive for a constant refinancing of

existing credit contracts, as they do not reimburse the hidden provision in the cancelled

insurance contracts in early repayment cases.

In the UK, unarranged overdraft charges on personal bank accounts account for onethird

of the banks’ revenues (OFT 2010b). Furthermore, in a different statement, the OFT

addresses the issue of fairness of default charges on credit cards (OFT 2006) in a survey

of major credit card issuers. Thereby, the level of default charges is considered unfair if it

exceeds a reasonable estimate of the marginal administrative costs in case of default.

OFT (2006) conclude that the level of credit card default charges in the UK was unfairly

high. Subsequently to the report, some credit card issuers have decided to reduce the

credit card default charges to the suggested threshold of GBP12 (see PwC 2008). As

argued by PwC (2008), credit card issuers have attempted to replace the income thereof

by introducing fees in other areas, for example on dormant or low-activity accounts. The

report advocates that direct price regulation applied on specific charges will have a

iff/ZEW – Final Report on interest rate restrictions in the EU 315

“waterbed effect” (PwC 2008) on other charges. From this, we learn the following: there

is a variety of charges even in the UK, a country without IRR. This variety of

charges offers sufficient room for an efficient adaptation of the pricing structure to

any kind of political pressure or regulation, including interest rate restrictions. This

makes it likely that providers react with the introduction or strengthening of unaffected

income sources when interest rate restrictions are introduced, as suggested by H7.

Since the introduction of the ceiling in Poland, the products for home lending have had to

be reworked in a way that a separate optional convenience fee was introduced (Chapter

2.4.2.4.2, p.213). The evidence is, again, in line with H7. It should be noted that in

Poland administrative fees and charges are additionally capped at a level of 5%. The

convenience fee, however, is not captured by the definition of administrative fees and

charges.

DG SANCO (2009) focuses on the simplicity and transparency in the charges on bank

accounts and conducts a cross-country comparison of the EU Member States.

Unfortunately, it does not consider credit contracts explicitly. For the bank accounts, its

evidence shows that the level of transparency and simplicity of charges278 on personal

bank accounts do not depend on the interest rate regulation of the country. For

example, France and Poland are distinguished by both above average intransparency

and above average complexity of the charges on personal bank accounts, whereas in

Germany and the Netherlands the opposite is true. All these are countries with IRR.

The Netherlands is even reported to be the country with the highest level of both

transparency and simplicity of charges. The report suggests that there are factors other

than interest rate restrictions influencing the structure and assessment of bank account

charges such as the level of the bank account charges themselves. It is an open

question, however, whether these findings can also be transferred to credit markets. We

thus recommend to evaluate the transparency and complexity of consumer credit

contracts in an international comparison.

2.5.9.4 Views of Stakeholders

Among the stakeholders participating in the survey who claim that interest rates should

be capped, 73% hold the opinion that charges and fees should be capped alongside

interest rates (Figure 139). It should be noted that 91% of the stakeholders who are in

favour of interest rate restrictions represent consumer organisations and “other”

activities such as banking authorities, financial regulators and government officials. The

few responses from provider associations in favour of an interest rate cap unanimously

argue against an additional cap on charges and fees. However, the tight connection

between a desire to regulate interest rates on the one hand and fees and charges on the

other highlights the importance of a potential restructuring of pricing schemes by

providers in case of an introduction of interest rate restrictions. This finding is also in line

with H7.

278 Transparency is measured by the need for an in-depth search in order to fully identify the total price of

credit paid by the customer. Simplicity captures the number of components of a bank account tariff.

316 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 139: Acceptance of regulation of fees and charges

0% 20% 40% 60% 80% 100%

Other Activity

Consumer

Organisations

All Stakeholders

Yes

No

Don't know

Source: SQ Question: Level of fees and charges should be regulated. [Yes/No/Don’t know]. Note: A separate

category showing provider associations responses was not meaningful seeing as too few responses to

this particular question were received from them.

Also, the majority of respondents to the PQ agree that interest rate restrictions would

lead to an increase in the level of administrative and other fees, as implied by H7.

Exceptions are commercial banks from Germany and the Netherlands who argue that the

level of fees would not change as a result of the regulation.

Home credit providers participating in the survey further note that, as there are no

administrative fees charged in home lending, an interest rate ceiling would far more lead

to a maturity extension of the products.279 In a personal interview a major home credit

provider from Poland further claims that the regulation would increase the intransparency

of the products. In connection with the evidenced negative correlation between

transparency and the level of charges on bank accounts (see DG SANCO 2009), this

expectation would be in line with H7.

279 In case the restriction (regardless of whether absolute or relative) is set based on APR.

iff/ZEW – Final Report on interest rate restrictions in the EU 317

2.5.10 H8: IRR represent barriers to consumer credit market integration

2.5.10.1 An Overview

Given the multitude of factors determining cross-border market entry or international

diversification of credit companies, the evidence on H8 is inconclusive.

It is not the interest rate restrictions as such, but the variety of different

interest rate regulations which may be a barrier to consumer credit

market integration in Europe. However, since there are multiple differences in

credit regulation, it is likely that barriers will remain despite a potential

harmonisation of interest rate regulations.

The Stakeholder questionnaire shows that provider associations support H8.

Consumer associations are more critical of this point. From provider associations,

however, there are also voices deemphasising the role of interest rate restrictions

for cross-border activity, saying that the variety of interest rate restrictions is

deeply rooted in the cultural background and the national attitude towards credit.

2.5.10.2 Literature

Several studies have argued that the variety of regulatory measures across the EU has

lead to lower levels of credit market integration. In the context of mortgage suppliers,

CEC (2007) points at the legal “costs of adapting the products and producing different

materials in accordance with different national frameworks” as a limitation to enter

markets in different Member States. CEC (2007) acknowledges that may of these legal

institutions are created as devices of consumer protection and calls for “balancing the

benefits of product diversity with the need to protect consumers”. In this context of

consumer credit, one may also make this argument with respect to interest rate

restrictions, for which we have extensively demonstrated the heterogeneity in the first

part of this report. In the same vein, CEC (2002) sees “a lack of adequate harmonisation

as regards national legislation” as one of the causes for the “sluggish development of the

European cross-border credit market“. In line with this, DG SANCO (2008) sees the “Lack

of harmonisation of relevant MS legislation or absence of EU legislation“ to be a „very

significant barrier“ for all retail financial services. In line with this view, DG Internal

Market and Services (2006) report that financial institutions overwhelmingly express that

usury rules have a negative impact on integration. In that report, which summarizes the

contributions provided to the Commission in the context of the consultation on its Green

Paper on Mortgage Credit (2005), it is also apparent that other stakeholders, such as

consumers or Member State authorities, provided an unclear pattern of responses. This

report concludes that the main message of the stakeholders was that there is “no need to

introduce any usury rules at the EU level”.

On the other hand, one has to acknowledge that interest rate restrictions are only a small

part of the overall regulation which is meant by the remarks above. For this reason, it is

likely that barriers to integration will still persist even if interest rate restrictions are

harmonised. In this vein, EUROFINAS (2010)280 therefore deemphasise the importance of

interest rate restrictions by pointing out that they “do not believe that diversity of policy

of interest rate restrictions has a decisive impact on lenders’ cross-border trade

strategy”. They also acknowledge that the divergence of interest rate restrictions across

the EU Member States “can largely be explained by distinct national cultural

preferences”, which “[…] correspond to different cultural approaches towards access, use

of credit products and regulation of lending activities”.

280 Note that EUROFINAS is the association of specialised consumer credit providers in Europe and is thus to be

seen as a particular stakeholder, which makes these insights above inherently relevant from a practitioners’

perspective.

318 iff/ZEW – Final Report on interest rate restrictions in the EU

2.5.10.3 Views of Stakeholders

Facts and Figures

As demonstrated in Figure 140, stakeholders’ responses substantially differ among

Member States with IRR and Member States without IRR. As can be seen from Figure

140 stakeholders from countries with IRR expect these to be stronger barriers (55%) to

market integration than stakeholders from countries without IRR.

Figure 140: IRR as a barrier to the cross-border provision of consumer credit

0% 20% 40% 60% 80% 100%

MS without IRR

MS with IRR

not a barrier/insignificant

small barrier

significant/very significant

SQ Question: In your view, how important are interest rate restrictions (incl. restrictions on fees) as barriers to

the cross-border provision of consumer credit in the EU? [not a barrier/insignificant/small

barrier/ significant/ very significant]

Interestingly, as can be observed from Figure 141, provider associations assign a greater

importance to interest rate restrictions in the discussed context than consumer

organisations and “other” stakeholders such as financial regulators and banking

authorities.281 In the case of the latter, a majority of 72% even assert that IRR are at

most only a small barrier to the cross-border provision of credit. The majority of them

further indicates that consumer preferences for local creditors as well as foreign

providers’ lack of information about customers and language barriers are significant or

very significant barriers to the cross-border provision of credit.

Figure 141: IRR as a barrier to the cross-border provision of consumer credit (divergence of

opinion among stakeholder categories)

0% 20% 40% 60% 80% 100%

Other

Consumer

Organisations

Provider

Associations

not a barrier/insignificant

small barrier

significant/very significant

Discussion

The Stakeholders (in particular provider associations) support H8.

281 In a written feedback individual providers from UK and Poland even indicate considerations a cross-border

market entry in the past. They indicate that existing interest rate restrictions in the foreign country were

among the main reasons to refrain.

iff/ZEW – Final Report on interest rate restrictions in the EU 319

2.5.11 H9: IRR lead to lower levels of competition in the consumer credit

industry.

2.5.11.1 An Overview

Comparing markets with and without interest rate restrictions with respect to

competition, stakeholders’ concerns that tough regulation might lead to less supply and

therefore lower competition cannot be confirmed. H9 is thus unlikely to hold:

A recent study has reported concerns that specific segments of the high-cost

credit market do not function competitively due to limitations of the demand side.

As this has been found in the UK, a country without interest rate restrictions, this

finding deemphasizes the role of interest rate restrictions in the context of

competition: the absence of interest rate restrictions does not appear to be

sufficient to come to competitive prices.

There are also concerns that tougher regulation (including interest rate

restrictions) could lead to an additional reduction of the supply side, making the

problem even worse.

While we are not able to quantify the degree of competiveness of specific market

segments due to data limitations, we discuss the competiton in the banking sector

in various countries. We argue that from this sector, players could easily enter

any part of the consumer credit market if excessive returns were to be earned

there.

There is no evidence that different approaches to regulate or not regulate

interest rates in credit markets affect the degree of competitiveness of the

banking segment, as H9 suggests.

2.5.11.2 Introduction

OFT (2010) expresses concerns that the market of high cost credit in the UK does not

function in a competitive manner due to limitations of the demand side in shopping for

best prices. As a consequence, this finding could imply that consumers pay excessively

high prices when taking out high cost credit. This argumentation refers to the low

elasticity of demand, which allows suppliers to engage in non-competitive pricing. This

holds despite the fact that there are no interest rate restrictions in the UK. One can

therefore conclude that, in specific markets, the absence of interest rate restrictions is

not sufficient to come to competitive prices. This is no direct evidence against the

hypothesis, but deemphasizes the role of interest rate restrictions in the context of

competitive pricing.

Another argumentation stresses that the presence of interest rate restrictions even

worsens this problem: interest rate restrictions could lower the profitability of providing

services in specific segments of the consumer credit market, which may drive

competitors out of the market. As a result, the remaining suppliers would face too little

competition to keep prices at competitive levels. This reasoning refers to the

concentration of the credit supply side, which is a traditional way of thinking about

competition in markets.

320 iff/ZEW – Final Report on interest rate restrictions in the EU

H9 postulates that competition is lower in countries with IRR. In the following

paragraphs, we therefore discuss various measures of market competitiveness in the

banking sector.282

2.5.11.3 Cross-country comparison

Facts and Figures

A traditional measure of competition is the Herfindahl-Index. It is a weighted

concentration measure which over-weights large players.283 Figure 142 depicts the ECB

(2004) estimates of this measure for different Member States. According to the

Herfindahl-Index, Germany is ranked the country with the highest level of competition in

the banking sector, as it exhibits low market concentration. At the other end of the

ranking is Belgium, with the highest Herfindahl-Index and thus the lowest competition.

Interestingly, countries with IRR are very dissimilar in terms of their market

concentration: Germany and the Netherlands, countries with IRR, are found at different

ends of the ranking in Figure 142. France (another country with IRR) and Sweden (a

country without) are relatively close in their Herfindahl-Index.

Figure 142: Competition measures by the Herfindahl-Index

0

500

1000

1500

2000

2500

Germany

Italy

Luxembourg

U.K.

Spain

EU 15

Austria

Ireland

France

Sweden

Portugal

Denmark

Greece

Netherlands

Belgium

Source: ECB (2004), p.36.

However, concentration measures are problematic to determine competition in the

banking sector when there is limited connection between regional markets within a

country.284 This is the case for, eg. Germany, where public banks and cooperative banks

are restricted to specific geographical units. As a consequence, different measures are

282 The banking sector may be too broad to capture the competition in a specific segment of the credit market.

However, the banking sector can be assumed to include sufficient agents who could easily enter a specific

market if excessive profits wetre to be earned in it. The exact determination of competition in specific

segments requires econometric analyses and extensive data input, which is beyond the scope of this study.

283 See, eg. Sachverständigenrat (2008).

284 See Sachverständigenrat (2008).

iff/ZEW – Final Report on interest rate restrictions in the EU 321

applied to determine the mark-up banks can charge over their marginal costs. This idea

is reflected by the Lerner-Index, which depends on the price elasticity of demand.285

The numbers presented in Figure 143 capture the banking business in a broader sense,

since they include interest and non-interest income. A low Learner-Index reflects a high

level of competition. According to this measure, the UK is ranked second concerning

competition and Sweden is ranked fourth across the European countries considered.

While these two countries have no interest rate restrictions in place, one has to note that

also Germany and France are also above the EU 15 average. The Netherlands, in

contrast, exhibits a relatively high level of market power of its banks.

Another measure, the H-Statistic is frequently being used as a measure of market power

in the banking industry.286 It reflects the relationship between input prices and revenues

prevailing in a market: if changes of costs transmit efficiently into prices, the market is

considered to be competitive.287 The H-Statistic is unity under perfect competition,

between zero and one for monopolistic competition, and smaller than zero under

monopoly. Figure 144 illustrates that the H-Statistic provides a different picture about the

competitiveness in the banking sector in different countries. In this ranking, the

Netherlands (country with IRR) is found to be a rather competitive country, whereas

Sweden (a country without interest rate restriction) is found to have the least

competition across countries. However, the UK (without IRR) is ranked above average,

while Germany and France (with IRR) are ranked below average.

Figure 143: Competition measured by the Learner Index

0

5

10

15

20

25

Luxembourg

U.K.

Belgium

Sweden

France

Germany

Ireland

Italy

Portugal

EU 15

Austria

Greece

Netherlands

Spain

Denmark

Source: Carbó et al. (2009).

285 If the price elasticity of demand is infinitely large, the Lerner Index converges to zero and implies perfect

competition (Sachverständigenrat, 2008).

286 See Sachvertändigenrat 2008.

287 For a more detailed discussion, see Carbó et al. (2009).

322 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 144: Competition measured by the H-Statistic

0

0.2

0.4

0.6

0.8

1

1.2

Luxembourg

Portugal

Netherlands

Ireland

Austria

U.K.

EU 15

Italy

France

Spain

Germany

Denmark

Belgium

Greece

Sweden

Source:Carbó et al. (2009).

Discussion

The paragraphs above demonstrate that different competition measures lead to different

conclusions about the relative competitiveness in different markets. The UK is

consistently found to have a relatively competitive banking industry. Other countries

show more diverse patterns across different measures.288 Looking at these other

countries, there is no evidence that different approaches to regulate or not to

regulate interest rates in credit markets affect the degree of competitiveness of

the banking segment, as H9 suggests. Rather, it appears that other factors (banking

regulation, the role of state-owned banks, historical developments) are more important

in determining the competitiveness of those markets.

288 Carbó et al. (2009) also emphasise this point.

iff/ZEW – Final Report on interest rate restrictions in the EU 323

2.5.12 H10: IRR lead to a convergence of all consumer credit interest rates at

the level of the interest rate cap

2.5.12.1 An Overview

Considering findings from the literature and the analysis of the development of interest

rates in the Netherlands and Poland where IRR were introduced, does not allow drawing

clear-cut conclusions, since there are opposing findings. The results on H10 are therefore

inconclusive”.

Theory argues that certain types of interest rate caps may be used by competitors

to collude on prices.

Other studies have documented that H10 holds for some market segments in

France.

It does not appear that the introduction or lowering of an interest rate cap in

Poland or the Netherlands, respectively, has lead to an increased clustering at the

level of interest rate cap. This evidence speaks against H10.

We stress the point (already mentioned on p.152) that the possibility of a

convergence to the level of the interest rate caps is only given for specific forms

of interest rate caps, such as exogenously determined rates. Thus, the problem

implied in H10 could be avoided by the right design of interest rate caps.

2.5.12.2 Literature

IFG/IGAS (2009) analyses the distribution of interest rates of different credit types in the

French consumer credit market. With respect to small revolving credits and overdraft

loans (<EUR 1524), they demonstrate a clustering of interest rates slightly below the

legal ceiling. With respect to revolving credit and overdrafts of higher amounts

(>EUR 1524), they document a clustering of interest rates at this level for specialpurpose

banks, but not for general banks. These findings suggest that the consequences

described in H10 may, in fact, become reality.

2.5.12.3 Past experience

Figure 145 shows the development of consumer credit interest rates in Poland in relation

to the interest rate ceiling during the period from 2005 to 2010. As can be seen from the

graph, we do not observe a convergence of interest rates to the ceiling until 2009, as

H10 implies. The convergence observed in 2009 is more likely to be due to an abrupt

reduction of the ceiling rather than to an increase in interest rates.

324 iff/ZEW – Final Report on interest rate restrictions in the EU

Figure 145: Consumer credit interest rates and ceiling in Poland, 2005-2010

0

5

10

15

20

25

30

35

31.03.2005

31.05.2005

31.07.2005

30.09.2005

30.11.2005

31.01.2006

31.03.2006

31.05.2006

31.07.2006

30.09.2006

30.11.2006

31.01.2007

31.03.2007

31.05.2007

31.07.2007

30.09.2007

30.11.2007

31.01.2008

31.03.2008

31.05.2008

31.07.2008

30.09.2008

30.11.2008

31.01.2009

31.03.2009

31.05.2009

31.07.2009

30.09.2009

30.11.2009

IR Ceiling

Consumer loans: variable rate and up to 3 months initial rate fixation

Consumer loans: over 3 months and up to 1 year initial rate fixation

Consumer loans: over 1 and up to 5 years initial rate fixation

Consumer loans: over 5 years initial rate fixation

Source: National Bank of Poland, www.nbp.pl

Note: interest rates provided are agreed annual rates (AAR). Loans are in Polish Zloty, new business. Interest

rate ceiling is calculated as four times the Lombard rate. Due to gaps in the Lombard rate

data, all rates are presented for those dates only for which Lombard rate is available.

Similar findings can also be made in the Netherlands: it can also be seen from Figure

134 on page 308 that the interest rate ceiling of 15% applied to the interest on consumer

credit in the Netherlands after 2006 is well above the shown interest rates, none of which

exceeds 9% at any time after 2006. This contradicts the hypothesis that interest rate

restrictions are used as a reference point for implicit collusion, which would imply that

interest rates converge to the rate cap (H10).

Discussion

While findings from France suggest that H10 is likely to hold in some market segments,

it does not appear that the introduction or lowering of an interest rate cap in Poland or

the Netherlands, respectively, has lead to an increased clustering at the level of interest

rate cap. There is thus mixed evidence.

In this context, it has to be noted that clustering at the level of the interest rate ceiling is

only given for specific forms of interest rate caps, such as exogenously

determined rates. For interest rates caps tied at an average of the targeted interest

rates (eg. twice the typical consumer loan interest rate), the cap would keep increasing

when all actors are rising their rates towards this level. This is unlikely in a market

structure with several players, as they are threatened to be underbid by competitors

easily.

iff/ZEW – Final Report on interest rate restrictions in the EU 325

2.5.13 Summary H1-H10

2.5.13.1 Introduction

For most of the above-mentioned hypotheses, theoretical reasoning comes to rather

clear-cut conclusions. In contrast, collecting evidence which can be observed in the

reality of European markets (as done in the paragraphs above) reveals that, for many

hypotheses, the impact of interest rate restrictions may be less obvious than postulated

by theory. The above subchapters thus carefully analyse to what extent theoretical

considerations can be confirmed or rejected. In fact, there are three out of the twelve

hypotheses and sub-hypotheses which can be broadly confirmed to be plausible by our

investigations. Three different hypotheses are found to be relatively unlikely. In six more

cases, conflicting evidence or the lack of data leads to inconclusive results.

2.5.13.2 Summary of the hypotheses

The first hypothesis (H1: IRR reduce credit access, in particular for low-income

borrowers) is generally found to be plausible: High-risk borrowers requesting smallamount

credit can only be served when a certain threshold interest rate is exceeded.

Hence, they may not be served credit in the presence of interest rate restrictions.

However, one needs to keep in mind that, due to the relatively high levels of interest rate

caps in most European countries, the scope of the interest rate restrictions is not

expected to be equivalent to the ones documented in the US a few decades ago. Rather,

it is likely that access to mainstream credit (including overdrafts and revolving credit)

remains rather unaffected by IRR, while there may still be missing credit options to lowincome

borrowers which are served in the high-cost credit segment in some countries.

Note that the desirability of such credit access to this specific population group is subject

to political controversy.

In the light of our analysis, a second hypothesis (H2: IRR lead to a decline in the

volumes of consumer credit granted) appears unlikely to hold in an economically

significant way. This may be explained by the fact that the relevant market segments of

high-cost credit (which are most affected by interest rate restrictions) only constitute a

relatively small fraction of the entire consumer credit market in which they exist. A lack

of these market segments is unlikely to lead to an economically significant drop in the

volume of credit markets, as the latter is affected by a multitude of other factors, as well.

This finding implies that economic activity is unlikely to be significantly supported by the

presence of high-cost credit (ie. the absence of interest rate restrictions). In contrast,

hypothesis H2a (Without IRR, more product types exist in the market) appears plausible:

Countries without interest rate restrictions tend to have a higher prevalence of personal

loans/auto loans than countries with interest rate restrictions. It is also likely that

interest rate restrictions prevent the market entry of one or more forms of high-cost

credit. Note in this context, however, that a relationship between the importance of

certain credit types and interest rate restrictions is unlikely to be purely mechanistic.

Rather, lower levels of consumer credit and the tougher regulation of credit (including

interest rate restrictions) may both be consequences of a third factor- the country’s

attitude towards credit.

With respect to another hypothesis (H3: IRR lead to credit from non-bank sources, such

as paying bills late), our results remain inconclusive. Some argue that the existence of

high-cost credit helps households to avoid obtaining credit from (potentially expensive)

non-bank sources, such as utility providers. However, it does not appear that there are

systematic differences in lending from non-bank sources between countries with and

without interest rate restrictions. A related claim in the context of interest rate

restrictions is addressed in H4 (IRR lead to a substantial illegal market in lending).There

326 iff/ZEW – Final Report on interest rate restrictions in the EU

is no convincing comprehensive data to evaluate this hypothesis. Due to this lack of data,

we evaluate the evidence on H4 to be inconclusive.

The related hypotheses H5 and H5a address the effect of interest rate restrictions on the

level of over-indebtedness and on its growth rate after an economic shock,

respectively.289 We conclude that a direct influence of interest rate restrictions on the

level of over-indebtedness, as H5 suggests, is unlikely. With respect to H5a, we obtain

inconclusive results: the most current figures do not reveal a particularly pronounced

increase of over-indebtedness in the aftermath of the financial market crisis. However,

there are no official data covering the years 2009 and 2010, and the views collected from

our stakeholders indicate that there might be a link as postulated by H5a in those years.

We therefore recommend re-evaluating this issue in the next year when the relevant

data are available.

With respect to the hypothesis H6 (The average consumer - or even more so: low-risk

consumer - would be granted cheaper credit in the presence of IRR), we find inconclusive

results. Due to the lack of micro data on individual credit cost before and after the

introduction of IRR , we focus on average interest rates. Drawing on these data and

findings from the Stakeholder Questionnaire, there is no unambiguous evidence that

average rates are declining, as H6 postulates. In contrast, it appears plausible that H7

holds (IRR lead to increased charges as providers will try to compensate the reduced

interest revenues by increased charges). There are examples from different countries

illustrating that providers react to regulation by charging fees for which the regulation

does not apply.

The evidence on H8 (IRR represent barriers to consumer credit market integration) is

inconclusive for two reasons. Firstly, it appears that not the mere presence of an interest

rate restriction, but rather the variety of interest rate restriction is potentially an

impediment to market entry. Secondly, the institutional setting is also diverse across

countries along several other dimensions, such that it is hard to assess how important

interest rate restrictions are in this environment. Hypothesis H9 (IRR lead to lower levels

of competition in the consumer credit industry) is found to be unlikely to hold, as the

level of competition differs (according to several criteria) across countries regardless of

the presence of interest rate restrictions. It has also been documented in other studies

that there is a low level of competition in high cost credit markets in the UK, a country

without interest rate restrictions in these segments. This also deemphasises the role of

interest rate restrictions in the context of competition.

The last hypotheses (H10: IRR lead to a convergence of all consumer credit interest

rates at the level of the interest rate cap) implies that providers use an exogenously

given interest rate cap to coordinate their (non-competitive) price setting at a rate just

below the cap. Other studies have found some evidence on this issue. We demonstrate

that the results on this hypothesis are inconclusive: whether or not the phenomenon

captured in H10 occurs crucially depends on the market structure and the exact way in

which interest rate restrictions are implemented.

2.5.13.3 Concluding remarks

In summary, we find that there are less clear-cut implications of economic significance of

interest rate restrictions than it is sometimes argued. However, it is apparent that

interest rate restrictions do shape the supply side of the consumer credit market in three

respects: Firstly, it is likely that the existence of interest rate restrictions excludes some

289 (H5: The lack of IRR leads to a higher level of over-indebtedness; H5a: The lack of IRR has particularly

adverse effects on default rates/ over-indebtedness in the presence of negative shocks (eg. recessions) to

the economy).

iff/ZEW – Final Report on interest rate restrictions in the EU 327

customer groups from credit access (which might or might not be an explicit objective of

the introduction of interest rate restrictions). Secondly, there might be a reduced set of

credit types, as some credit types with very high interest rates are not offered in the

presence of interest rate restrictions. Thirdly, one needs to keep in mind that suppliers

can (to some extent) structure their credit product in a way so that the existing interest

rate regulation does not apply.

We also conclude that many observations on credit markets are not only driven by

regulatory conditions (affecting the supply side), but also by the actual behaviour of the

demand side. This aspect is important as typical strategies of quantification of the effects

of interest rate restrictions on the basis of cross-country comparisons rely on the bold

assumption that demand patterns are identical across countries. In contrast, as

particularly explained in our discussion of H1 and H2a, it is more realistic to assume that

there are clear patterns in the attitude towards credit across countries which may explain

both the preferences for strict interest rate regulations prevailing in some countries and,

eg. reduced incidence of credit of their consumers.

328 iff/ZEW – Final Report on interest rate restrictions in the EU


3 What can the European Commission learn from this study?

1. IRR in general do not affect ordinary car loans, mortgage loans, home

improvement loans, ordinary overdraft for short-term liquidity because

a. Their interest rates are by definition far removed from the ceiling between

33% (France), 50% (Italy), 100% (Germany), 300% (Slovenia) because

ceilings are defined by average interest rates for these products.

b. Borrowers can exercise a certain amount of choice (sufficient competition).

2. IRR do shape the supply side of the credit market in the sense that they affect

high-cost credit.

a. Credit types operating with high annualised interest rates are not able to

operate in an environment of interest rate caps.

b. The lack of high cost credit reduces credit access to those parts of the

population which are considered to be high-risk and which demand small amount

credit. This reduction of these subgroups’ credit access may or

may not be an objective of a government’s policy.

c. Credit providers may tend to offset these effects by designing credit

contracts or additional service contracts charging fees which are not

captured by the interest rate regulation.

3. The mere existence of certain credit types or differential volumes of credit in

different countries is not a sufficient indication about the restrictiveness of credit

(including interest rate) regulation. Rather, these also reflect different attitudes

towards credit.

a. A country in which the population is more sceptical with respect to

consumer credit is likely to have lower total volumes of credit simply

because of low demand.

b. A country with these properties is more likely to opt for stricter legislative

measures to reduce the level of credit.

4. However, in these market segments we are increasingly confronted with “invisible

high interest credit” for consumers, the effective cost of which are much higher

than disclosed in the APRC, which is also the basis for comparison with the Usury

Ceiling.

a. Credit and payment protection insurance contain high kick-back provisions

and take unusual forms, such as long-term prepayment of premiums to

facilitate their financing, relative short-term of risk coverage, unfavourable

early repayment conditions where much of the premium is withheld.

b. Endowment credit, where additional borrowing is necessary in order to

fund an investment or savings.

c. Credit card fees on, for example, cash withdrawal which adds invisible interest.

5. If the internal market favours increased competition and choice, the following

differentiation could take place:

iff/ZEW – Final Report on interest rate restrictions in the EU 329

a. Ordinary investment credit is not affected by IRR – there is no need for the

harmonisation of such rules to further transborder commerce.

b. Circumventions with the effect of high-price credit in this segment occur

irrespective of the existence of IRR – improved EU regulation on price

disclosure including all price elements which burden the borrower would

help to clarify the impacts.

c. Since it appears that short-term credit is often extended under conditions

of little or no competition, removal of IRR would not be conducive to

consumer benefit and confidence.

6. If regulation is deemed necessary it should be modern and adequate for market

conditions.

a. Instead of criminal law with moral and subjective rules, private law with

objective market-related ceilings specific to certain products would be

more self-executing.

b. Rules should carefully observe the impact on the distribution of certain

regulated products. Differentiation by marketing practice, product, life-time

and amount is more promising than unified approaches.

c. Rules should be immune to circumvention. IRR require an objective

consumer view of the amount payable, irrespective of the purposes to

which these payments are attributed by the supplier side.

d. Sanctions should also be clear and easy to understand and develop enough

threat to provide the underlying rules with sufficient deterrent effect.

7. IRR need a common ground which develops the traditional concept of usury and

good morals into a market-driven concept. It could be derived from the new

market-related concepts of relative rate ceilings, which replace the subjective and

moral principle of exploitation by a more objective concept of market dysfunction,

possibly developed from cartel law. The EU could play a special role in providing a

framework for such a concept, leaving its operational form for the Member States,

to determine according to their different traditions and stage of market

development as indicated in the opinion of the Attorney General in ECJ Case C-

484/08.

a. Art. 101 (1) (a) of the Lisbon Treaty creates the idea of unfair prices

because of a lack of competition through the creation of monopolies and

cartels. State agencies intervene with “as-if-prices”, representing prices

from markets which are still functioning, in a manner which is comparable

to interest rate ceilings derived from average market rates.

b. The concept of unfairness should cover all forms of distorted competition

with regard to prices, and include both opaque pricing and extortionate

pricing in consumer credit.

c. Such propositions could be made within the actual process of proposing a

Common Frame of Reference for EU consumer contract law.

330 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex I: Bibliography

Agarwal S., Skiba P. M., Tobacman, J. (2009). “Payday Loans and Credit Cards: New

Liquidity and Credit Scoring Puzzles?” American Economic Review: Papers & Proceedings

99(2), pp. 412–417.

Aidit bin Hazi Ghazali (1994). “Consumer Credit from the Islamic Viewpoint”, Journal of

Consumer Policy 443 et seq.

Anderloni, L., Vandone D. (2008). “Households Over-Indebtedness in the Economic

Literature”, Departmental Working Papers No. 2008-46, Department of Economics

University of Milan Italy.

Asteres (2007). “Credit a la consommation et lois sur I 'usure - faut-il modifier les taux

plafonds?”, etude realisee pour LaserCofinoga.

Athling (2008). “Pour un developpement responsable du credit renouvelable en France”,

Rapport realise pour le Comite consultatif du secteur financier, available at

http://www.responsible-credit.net/media.php?id=3390.

Attanasio, O.P., Goldberg P.K., Kyriazidou E. (2008). “Credit Constraints In The Market

For Consumer Durables: Evidence from Micro Data on Car Loans”, International Economic

Review 49(2), pp. 401-436.

Ausubel L.M. (1991). “The Failure of Competition in the Credit Card Market” The

American Economic Review 81, pp. 50-81.

Babeau, A. (2006). „”La demande des menages en mattere de credit a la consommation

et les ajustements necessaires po ur y repondre, rapport realise par le EIPE pour le

cornite consultatif du secteur bancaire“.

Barberis, N., Thaler, R. (2003). “A Survey of Behavioral Finance”, in: Constantinides,

G.M., Harris, M., Stulz, R. (Ed.), Handbook of the Economics of Finance, Chap. 18, pp.

1054-1123.

Barr, M. S. (2004) “Banking the Poor.” Yale Journal on Regulation 21(1), pp. 121-237.

Barr, M. S., Mullainathan, S., Shafir, E. (2005). “Credit Where It Counts.” New York

University Law Review 80(2): pp 513–652.

Barr, M. S.; Mullainathan, S.; Shafir, E. (2008). “Behavioural Informed Financial Services

Information”, New America Foundation.

Benmelech, E., Moskowitz, T.J. (2010). “The Political Economy of Financial Regulation:

Evidence from U.S. Usury Laws in the 19th Century” Journal of Finance, forthcoming.

Berlin, Mitchell, Mester, Loretta J. (2004). “Credit Card rates and consumer search”.

BERR (2007). “Tackling Over-Indebtedness – Annual Report 2007”, URN 07/445.

Berthoud, R., Kempson, E.: (1992). “Credit and Debt: The PSI Report”, Policy Studies

Institute, London.

Bester H. (1985). “Screening versus Rationing in Credit Markets with Imperfect

Information” American Economic Review 75, pp. 850-855.

Betti, G., Dourmashkin, N., Rossi, M., Ying, Y.P. (2007). “Consumer over-indebtedness in

the EU: measurement and characteristics” Journal of Economic Studies 34(2), pp. 136-

156.

Białowolski, P. (2009). “Raport o skutkach ekonomicznych ustawy o maksymalnym

oprocentowaniu, Konferencja Przedsiębiorstw Finansowych w Polsce”, Warszawa.

iff/ZEW – Final Report on interest rate restrictions in the EU 331

BIPE (2006). “La Demande des Ménages en Mariére de Crédit á la Consommation et les

Ajustements Nécessaires pour y Répondre”, Rapport réalisé par le BIPE et présenté par

André Babeau pour le Comité consultatif du secteur financier.

BIS (2007). “Principles for Remittance”, available at http://www.responsiblecredit.

net/index.php?id=1980&viewid=43128.

Blitz, R.C., Long, M.F. (1965). “The Economics of Usury Regulation” Journal of Political

Economy 73(6), pp. 608-619.

BMAS (2008). “Life Situations in Germany: The German Federal Government's 3rd

Report on Poverty and Wealth”, July 2008, Art.-Nr.: A332.

Bodenhorn, H. (2007). “Usury Ceilings and Bank Lending Behavior: Evidence from

Nineteenth-Century New York” Explorations in Economic History 44:2, 179-202.

Bond, P., Musto, D.K., Yilmaz, B. (2009). “Predatory Mortgage Lending”, Journal of

Financial Economics 94(3), pp. 345-508.

Brennecke, H., Otépková, E. (2009). “Insolvenz und Restschuldbefreiung in Europa - ein

Vergleich der Insolvenzordnungen der Länder der EU“, Karlsruhe.

Bridges, S., Disney, R. (2006). “Debt and depression” Centre for Finance and Credit

Markets Working Paper 06/02, University of Nottingham.

Brouwer, H.J. (2008). “The Dutch housing market: stable or stagnant?” Speech at the

DBV symposium, Zeist, 19 May 2008.

Brown, J.D., Taylor, S.E. (1988). “Illusion and Well-Being: A Social Psychological

Perspective on Mental Health.” Psychological Bulletin 103(2), pp 193-210.

Brown, M., Peter, M., Wehrmüller, S. (2009). Swiss Franc Lending in Europe, Swiss

National Bank.

Calem, P.S., Mester, L.J. (1995). “Consumer Behavior and the Stickiness of Credit-Card

Interest Rates”, American Economic Review 85(5), pp. 1327-1336.

Caplovitz, D. (1963). “The Poor Pay More - Consumer Practices of Low Income Families -

The Cost of Paying Later, The Instalment Plan”, Free Press of Glencoe: New York.

Caplovitz, D.: (1974). “Consumers in Trouble - A Study of Debtors in Default”, The Free

Press: New York 1974.

Carbó, S., Humphrey, D., Maudos, J., Molyneux, P. (2009). “Cross-county comparisons of

competition and pricing power in European banking” Journal of International Money and

Finance 28, pp. 115-134.

Casolaro, L., Gambacorta, L., Guiso, L. (2006). “Regulation, Formal and Informal

Inforcement, and the Development of the Household Loan market: Lessons from Italy”

The Economics of Consumer Credit by Bertola, G., R. Disney and C. Grant (Eds.), The

MIT Press, Cambridge, Massachussets.

Cayne, D.J.; Trebilcock, M. (1973). “Market Considerations in the Formulation of

Consumer Protection Policy” 23 University of Toronto Law Journal 396 et seq.

CEC (2002). “Proposal for a Directive of the European Parliament and of the Council on

the harmonisation of the laws, regulations and administrative provisions of the Member

States concerning credit for consumers”, Commission of the European Communities,

Brussels.

332 iff/ZEW – Final Report on interest rate restrictions in the EU

CEC (2007). “Impact Assessment, Commission Staff Working Document, Accompanying

the White Paper on the Integration of EU Mortgage Credit Markets”, Commission of the

European Communities, Brussels.

Cetelem (2006). “Responsible lending Principles”, available at http://www.responsiblecredit.

net/index.php?id=1980&viewid=43128.

Cherbonnier, F. (2008). “Faut-il faire evoluer la legislation de l'usure en France?” Rapport

moral sur l'argent dans le monde.

Coco, G., de Meza, D. (2009). “In Defense of Usury Laws” Journal of Money, Credit and

Banking 41(8), pp. 1691-1703.

Collard, S., Kempson, E. (2003), "Pawnbrokers and their customers (report)", The

Personal Finance Research Center, University of Bristol, Bristol.

Collard, S., Kempson, E. (2005). “Affordable Credit. The Way Forward”, The Policy Press,

Bristol.

Competition Commission (2006). Home Credit Market Investigation.

Conseil de la Consommation, Belgique (2006). “Avis concernant le mode de fixation des

taux annuels effectifs globaux en matière de crédit à la consummation” available at

http://www.responsible-credit.net/media.php?t=media&f=file&id=3861.

Conseil de la Consommation, Belgique (2006). “Avis concernant le mode de fixation des

taux annuels effectifs globaux en matière de crédit à la consommation“, available at:

http://statbel.fgov.be/fr/binaries/354_tcm326-41864.pdf.

Corr C. (2007). “Alternative Credit Providers in Europe, December 2007 edition”

(European Financial Inclusion Network, available at:

http://www.fininc.eu/gallery/documents/workingpaper-alternativeproviders-final.pdf).

Creditreform (2009). “SchuldnerAtlas Deutschland, Jahr 2009”, Neuss.

Creditreform (2010). “Insolvencies in Europe”, Neuss.

Deaton A. (1992). "Understanding Consumption" Oxford: Clarendon Press.

Department for Business Innovation and Skills (2009). “A Better Deal for Consumers.

Review of the Regulation of Credit and Store Cards: A Consultation”.

Der Spiegel (2007). “Geschäft mit der Armut”, 21/2007, available at:

http://www.spiegel.de/spiegel/print/d-51644679.html.

Deutsche Bundesbank (2004). “Die neue EWU-Zinsstatistik – Methodik zur Erhebung des

deutschen Beitrags“, Deutsche Bundesbank Monatsbericht, Januar 2004, available at:

http://www.bundesbank.de/download/volkswirtschaft/mba/2004/200401mba_ewuzinsst

atistik.pdf.

DG MARKT (2006). “Feedback on the consultation on the Green Paper on Mortgage

Credit”, available at: http://ec.europa.eu/internal_market/finservices-retail/docs/homeloans/

feedback_gp-en.pdf.

DG SANCO (2008). “Final Report to DG SANCO – Analysis of the Economic Impact of

Directive 2002/65/EC concerning the distance marketing of consumer financial services

on the conclusion of cross-border contracts for financial services between suppliers and

consumers within the Internal Market”, DG SANCO Framework contract Lot 2 – Economic

part, by Alleweldt, F., Milne, A., Schubert, K., Gonzalo, S., Kara, S. and Buccino,S.

iff/ZEW – Final Report on interest rate restrictions in the EU 333

DG SANCO (2009). “Data collection for prices of current accounts provided to consumers,

prepared by Van Dijk Management Consultants in partnership with Centre for European

Policy Studies”.

Dilcher, J. (2002). “Die Zins-Wucher-Gesetzgebung in Deutschland im 19. Jahrhundert :

eine entwicklungsgeschichtliche Betrachtung unter besonderer Berücksichtigung

wirtschaftlicher und wirtschaftspolitischer Faktoren”, Frankfurt am Main: Lang.

Dischinger, N., Mögel, N.A. (2004). “Wer Besucht ein Pfandhaus? Nutzer gewerbliche

Mikrokredite in Deutschland“ Arbeitsbericht Nr. 29, Otto-von-Guericke Universität

Magdeburg.

Disney, R., Bridges, S., Gathergood, J. (2008). “Drivers of Over-indebtedness” Report to

the Department for Business, Enterprise and Regulatory Reform, October 2008.

Dominy, N., Kempson, E. (2003). “Pay Day Advances. The Companies and Their

Customers”, Personal Finance Research Centre, British Cheque Cashers Association,

Liphook.

Domont-Naert, F. (1993). “Le Surendettement des Consommateurs en Belgique -

Rapport Final”, Centre de Droit de la Consommation: Lou-vaine-La-Neuve.

DTI (2003). “Fair, Clear and Competitive. The Consumer Credit Market in the 21st

Century” presented to Parliament by the Secretary of State for Trade and Industry by

Command of Her Majesty December 2003.

DTI (2006). “Illegal lending in the UK. Research report.” URN 06/1883.

Dubois, H., Anderson, R. (2010, forthcoming) “Managing household debts. Social service

provision in the EU.”

Duhaime, G. (2003). “La vie à credit”, Montreal.

ECB (2001). “Regulation (EC) No. 63/2002 of the European Central Bank of 20 December

2001 concerning statistics on interest rates applied by monetary financial institutions to

deposits and loans vis-à-vis households and non-financial corporations”, ECB/2001/18,

OJ L 10, 12.1.2002.

ECB (2003a). “Manual on MFI Interest Rate Statistics”, Regulation ECB/2001/18.

ECB (2003b). “MFI Interest Rate Statistics. General Description.”

ECB (2009). “Survey Data on Household Finance and Consumption. Research Summary

and Policy Use”, Occastional Paper Series No. 100/January 2009.

ECRI (European Credit Research Institute) (2009). “Statistical Package 2009. Lending to

Households in Europe”, ECRI, Brussels, Belgium.

ECRI and PfRC (2008). “Consumer Financial Vulnerability: Technical Report”, available

at: http://www.genworth.eu/.

Eisenstein E.M., Hoch S.J. (2005). “Intuitive Compounding: Framing, Temporal

Perspective, and Expertise“, Working Paper.

Ekici, Tufan; Dunn, Lucia (2006) “Credit Card Debt and Consumption: Evidence from

Household-Level Data”.

Elliehausen, G., Lawrence, E.C. (2001). “Payday advance credit in America: An analysis

of consumer demand” Monograph 35, Georgetown University, Washington, D.C.

334 iff/ZEW – Final Report on interest rate restrictions in the EU

Elliehausen, G., Staten, M. (2004). ”Regulation of Subprime Mortgage Products: An

Analysis of North Carolina’s Predatory Lending Law” Journal of Real Estate Finance and

Economics 29(4), pp. 411-433.

Engel, K.; McCoy, P. (2007) “Turning a Blind Eye: Wall Street Finance of Predatory

Lending” 75 Fordham Law Review p. 2039.

Eurofinas (2010). “Preliminary Observations on Interest Rate Restrictions”, Brussels.

European Mortgage Federation (EMF) (2009). “Hypostat 2008 – A review of europe’s

mortgage and housing markets”.

Farinha, L. (2003). “The effect of demographic and socio-economic factors on

household’s indebtedness”, Banco de Portugal, Economic bulletin.

Farinha, L. (2004). “Households’ Debt Burden: An Analysis Based on Microeconomic

Data”, Banco de Portugal, Economic bulletin.

Fehr, E. (2002). “The Economics of Impatience” Nature 415(6869), pp. 269–272.

Fernando, N. (2006). “Understanding and dealing with high interest rates on microcredit”

Asian Development Bank.

Financial Inclusion Taskforce (2010). “Mainstreaming Financial Inclusion – Planning for

the future and coping with financial pressure: access to affordable credit”, available at:

http://webarchive.nationalarchives.gov.uk/+/http://www.hmtreasury.

gov.uk/d/fit_affordable_credit.pdf.

Financial Services Authority (2009). “Mortgage Market Review DP09/3”, available at:

http://www.fsa.gov.uk/pubs/discussion/dp09_03.pdf.

Finlay, S., (2009). “Consumer Credit Fundamentals”, 2nd edition, Palgrave Macmillan,

London.

FIN-USE (2010). “Opinion on national experiences on unfair commercial practices in the

EU”, available at: http://ec.europa.eu/internal_market/finuse_

forum/docs/unfair_commercial_practices_en.pdf.

Ford, J., Kempson, E., Wilson, M. (1995). Mortgage arrears and possessions:perspectives

from borrowers, lenders and the courts. HMSO (Department of the Environment).

Ford, J. (1988). “The Indebted Society: Credit and Default in the 1980s”, Routledge,

London.

Friedman, M. (1957). ”The Theory of the Consumption Function” Princeton: Princeton

University Press.

Galor, O., Zeira, J. (1993). “Income Distribution and Macroeconomics,” Review of

Economic Studies 60(1), 35-52.

Geiger, T. (1964). „Vorstudien zu einer Soziologie des Rechts“.

Ghazali, A.(1994). “Consumer Credit from the Islamic Viewpoint”, Journal of Consumer

Policy 443 et seq.

GHK (2009). “Establishment of a Benchmark on the Economic Impact of the Consumer

Credit Directive on the Functioning of the Internal Market in This Sector and on the Level

of Economic Protection” European Commission, DG Health and Consumer Protection,

Brussels.

Gilmore G. (2008). “Rise in payday loans show credit problems still to come” The Sunday

Times, 26.06.2008.

iff/ZEW – Final Report on interest rate restrictions in the EU 335

Glaeser, E., Sheinkman, J. (1998). “Neither a Borrower nor a Lender be: An Economic

Analysis of Interest Restrictions and Usury Laws”, Journal of Law & Economics l et seq.

Goudzwaard, M.B. (1968). “Price Ceilings and Credit Rationing” Journal of Finance 23,

pp. 177-85.

Grote, H. (2007). “Kredite Ohne „SCHUFA-Auskunft“. Rechtiliche Bewertung und

Handlungsmglichkeiten“, study on behalf of SCHUFA Holding AG.

Guardia, N.D. (2002). “Consumer Credit in the European Union”, ECRI Research Report

No. 1.

Hartmann-Wendels, T., Spörk, W. (2006) “Die Eignung der EWU-Zinsstatistik zur

Bewertung der Marktüblichkeit von Konsumentenkreditkonditionen“, Universität zu Köln,

available at: http://www.responsible-credit.net/index.php?id=1980&viewid=45790.

Hartmann-Wendels, T., Spörk W. (2008). “Die Bedeutung der Marktüblichkeit von

Konsumentenkreditkonditionen anhand der EWU-Zinsstatistik”, Zeitschrift für Bankrecht

und Bankwirtschaft (ZBB), Anlage 1 zum Rundschreiben des Bankenfachverbandes vom

14. Mai 2008.

Heidhues, P., Kőszegi, B. (2010 forthcoming). “Exploiting Naiveté about Self-Control in

the Credit Market”, American Economic Review.

Helfgott, J.B. (2008). “Criminal behaviour: theories, typologies, and criminal justice” pp

13ff.

Helms, B., Reille, X. (2004). “Interest Rate Ceilings And Microfinance: The Story So Far”,

CGAP Occasional Paper.

Herbert, A., Kempson, E. (1995). “Water debt and disconnection”, Policy Studies

Institute.

Hoffmann, D. (1997) „´Gerechtigkeitsprinzipien´ im Europäischen

Verbraucherprivatrecht – Fortschreitende Privatrechtskodifikation als Teil des sozialen

Europas“ in Krämer, l., Micklitz, H., Tonner, K., “Law and Diffuse Interest in the European

Legal Order”, p. 291 ff.

Holzscheck, K., Hörmann, G., Daviter, J. (1982). “Die Praxis des Konsumentenkredits in

der Bundesrepublik Deutschland - Eine empirische Untersuchung zur Rechtssoziologie

und Ökonomie des Konsumentenkredits”, Bundesanzeiger: Cologne.

Hörmann, G. (ed): (1993). “Consumer Debt in Europe - The Birmingham Declaration:

Proceedings of the Third European Conference on Overindebtedness”, Birmingham

Settlement.

Howell, N. (2005). “High Cost Loans: A Case for Setting Maximum Rates”, Centre for

Credit and Consumer Law, Griffith University.

Howells, G., Ramsay, I. , Wilhelmsson Th. (2010) “Handbook of research on international

consumer law” pp 397 ff.

Hynes, R.; Posner, E. (2002). “The Law and Economics of Consumer Finance” American

Law and Economics Review pp.168-207.

Hypostat (2008). “A Review of Europe’s Mortgage and Housing Markets”, European

Mortgage Federation, Brussels.

IFC (2008). "Banking On Sustainability", available at http://www.responsiblecredit.

net/index.php?id=1980&viewid=43128.

336 iff/ZEW – Final Report on interest rate restrictions in the EU

IGF/IGAS (2009). “Rapport sur les modalités de fixation du taux de l’usure”, Inspection

Générale des Finances/Inspection Générale des Affaires Sociales, Paris, available at:

https://www.igf.bercy.gouv.fr/sections/les_rapports_par_typ/les_enquetes_portant/politi

que_economiqu/les_modalites_de_fix/downloadFile/attachedFile/Rapport_sur_le_taux_d

_usure.pdf.

Inderst, R. (2009). „’Irresponsible Lending’ with a Better Informed Lender“, Economic

Journal 118, pp. 1499-1519.

INGO (2005). “Principles” (Accountability Charter, Dec 2005), available at

http://www.responsible-credit.net/index.php?id=1980&viewid=43128.

IOO (2009). Onderbouwing van een maximale vergoeding voor flitskrediet, Eindrapport,

Een onderzoek in opdracht van ministerie van Financiën, Zoetermeer, available at

http://www.responsible-credit.net/index.php?id=1980&viewid=45753.

Jackson, H. E.; Burlingame, L. (2007). “Kickbacks or Compensation: The Case of Yield

Spread Premiums.” Stanford Journal of Law, Business and Finance 12 pp. 289–361.

Jentzsch, N., Riestra, A.S.J. (2006). „Consumer Credit Markets in the United States and

Europe“ in: Bertola, G., Disney, R, Grant, C. (Eds.) The Economics of Consumer Credit,

Cambridge, London.

Johansson, M., Persson, M. (2006). “Swedish households’ indebtedness and ability to pay

– a household level study” Penning – Och Valutapolotik 3/2006, Sveriges Riksbank.

Jolls, C., Sunstein, C. R. (2005). “Debiasing Through Law.” Olin Working Paper 225,

University of Chicago Law and Economics, Chicago. Harvard Law and Economics

Discussion Paper, Cambridge, MA. http://ssrn.com/abstract=590929.

Kahnemann, D., Tversky, A. (1974). “Judgement under Uncertainty: Heuristics and

Biases” Science, Vol 185, pp. 1124-1131.

Karlan, D., Zinman, J. (2009). "Expanding Credit Access: Using Randomized Supply

Decisions to Estimate the Impacts", Working Papers 956, Economic Growth Center, Yale

University.

Kempson, E. (2002). “Over-indebtedness in Britain”, London: Department of Trade and

Industry.

Kempson E., Atkinson A. (2006). “Overstretched: People at Risk of Financial Difficulties”

PfRC, Bristol.

Kennedy, D. (2005). “Cost-Benefit Analysis of Debtor Protection Rules in Subprime

Market Default Situations.”, Building Assets, Building Credit, ed. Nicolas Retsinas and Eric

Belsky, pp. 266–82.

Kierzenkowski, R. (2008). “Bridging the Housing Gap in Poland” OECD Economics

Department Working Papers, No. 639, OECD Publishing. doi: 10.1787/236314440147.

Kilborn, J.J. (2005). “Behavioral Economics, Over-indebtedness & Comparative Consumer

Bankruptcy: Searching for Causes and Evaluating Solutions” Emory Bankruptcy

Developments Journal 22(1), p. 13-45.

Knobloch, M., Reifner, U., Laatz, W. (2008). “iff-Überschuldungsreport 2008”, available

at: http://news.iff-hh.de/media.php?t=media&f=file&id=3056.

Knobloch, M., Reifner, U., Laatz, W. (2009). “iff-Überschuldungsreport 2009”, available

at: http://news.iff-hh.de/media.php?t=media&f=file&id=3723.

Kozuka, S.; Nottage L. (forthcoming 2009). “The Myth of the Cautious Consumer: Law,

Culture, Economics and Politics in the Rise and Partial Fall of Unsecured Lending in

iff/ZEW – Final Report on interest rate restrictions in the EU 337

Japan”, in: Whitford, B., Ramsay, I., Niemi-Kiesiläinen, J. (eds). Consumer Credit, Debt

and Bankruptcy: Comparative and International Perspectives Hart Pub. UK.

Kraft, D (2008). “Private Law Control of Consumer Credit in the UK”, in: Twigg-Flesner,

C., Parry, D, Howells, G., Nordhausen, A (2008). “Yearbook of Consumer Law 2008”,

Aldershot: Ashgate.

Laibson, D. (1997). “Golden Eggs and Hyperbolic Discounting” Quarterly Journal of

Economics 112(2): pp. 443–477.

Landier, A., Thesmar, D. (2009). “Financial Contracting with Optimistic Entrepreneurs:

Theory and Evidence” Review of Financial Studies 22(1), pp. 117-150.

Lefebvre, M. F. (2008), Rapport d’Information sur les emprunts immobiliers à taux

variable, Paris: Assemblé Nationale Treizième Légistlature N° 744 (26 mars 2008)

Liebner, K. (2010). “Wucher und Staat: die Theorie des Zinswuchers im Deutschland des

18. und 19. Jahrhunderts”, Berlin : Duncker & Humblot.

Lord Crowther (1971), “Consumer Credit: Report of the Committee”, HMSO: London.

Lusardi, A. (1998). “On the Importance of the Precautionary Saving Motive” American

Economic Review 88(2), pp. 449-453.

Lusardi A., Mitchell O.S. (2007). “Baby Boomer retirement security: The Roles of

planning, financial literacy, and housing wealth”, Journal of Monetary Economics 54, pp.

205–224, available at:

http://www.pensionresearchcouncil.org/publications/pdf/boomer.pdf.

Lusardi, A., Tuffano, P. (2009). “Debt Literacy, Financial Experiences, and Overindebtedness”,

Working Paper.

Markesinis, B. et al. (2006). “The German Law of Contract”, Oxford: Hart Publishing.

Meier, S., Sprenger, C. (2007). “Impatience and Credit Behavior: Evidence from a Field

Experiment” Federal Reserve Bank of Boston Working Paper, No. 07-3, Boston.

Meier S., Sprenger, C. (2010). “Present-Biased Preferences and Credit Card Borrowing”,

American Economic Journal: Applied Economics 2(1), pp. 193–210.

Miles, D. (2004). “The UK Mortgage Market: Taking a Longer-Term View”, Final Report

and Recommendations, UK Treasury, London.

Mindy-Chen-Wishart, (2008) Contract Law, 2nd ed, Oxford University Press 2008.

Modigliani, F., Ando, A. (1963). “The Life-Cycle Hypothesis of Saving: Aggregate

Implications and Tests” In The Collected Papers of Franco Modigliani, Vol. 6, (2005),

pp.47-78. Cambridge: The MIT Press.

Modigliani, F., Brumberg, R. (1954). “Utility Analysis and the Consumption Function: An

Interpretation of Cross-Section Data.” In The Collected Papers of Franco Modigliani, Vol.

6, (2005), pp.3-45. Cambridge: The MIT Press.

MORI (2005). “Over-indebtedness in Britain: A DTI report on the MORI Financial Services

survey 2004”, London: Department for Trade and Industry.

Musto, D. K. (2007). “Victimizing the Borrowers: Predatory Lending’s Role in the

Subprime Mortgage Crisis.” Wharton working paper, Philadelphia, PA. Published February

20, 2008 http://knowledge.wharton.upenn.edu/article.cfm?articleid=1901.

Netherlands Bureau for Economic Policy Analysis, (2009). “The Credit Crisis and the

Dutch Economy 2009-2009”, available at:

http://www.cpb.nl/eng/pub/cepmev/cep/2009/pdf/summary_uk.pdf.

338 iff/ZEW – Final Report on interest rate restrictions in the EU

New Economics Foundation (2009). “Doorstep Robbery - Why the UK needs a fair lending

law”, available at

http://www.neweconomics.org/sites/neweconomics.org/files/Doorstep_Robbery.pdf.

Niemi-Kiesiläinen, J., Tala, J., Wilhelmsson, T. (1991). “Kuluttajien Velkasaneeraus –

Tarve ja mahdollisuudet (Consumer Debt Advising, Restructuring of Debts and Discharge

in Finland – an Analysis of Consumer Debt Situation and a Proposal for a Model of

Legislation)”, Oikeuspoliittinen Tutkimuslaitos No 106, Helsinki.

Office of Fair Trading (OFT) (2006). “Calculating fair default charges in credit card

contracts”.

Office of Fair Trading (2007). “Payment Protection Insurance”; OFT 899.

Office of Fair Trading (OFT) (2009). “Review of high-cost credit”, Interim Report, Crown

copyright, London (including Annexes).

Office of Fair Trading (OFT) (2009b). “Personal Current Accounts - Unarranged Overdraft

Charges - Decision on an investigation under the UTCCRs and next steps”.

Office of Fair Trading (OFT) (2010). “Review of high-cost credit” (OFT 1232), Final

Report, Crown copyright, London (including Annexes).

Oliver Wyman (2008). “Consumer Finance in Europe: Back to Reality”.

Olivi (1990). “Usure, compere e vendite: la scienza economica del XIII secolo”, Milano:

Europía.

Ott, K.-O. (1998) “Geld- und Geldwerttheorien im Privatrecht der Industrialisierung

(1815 - 1914) : ökonomische Wechsellagen in der sogenannten Begriffsjurisprudenz”,

Duncker und Humblot.

Oxera (2004). “Are UK households over-indebted?” Commissioned by the Association for

Payment Clearing Services, British Bankers Association, Consumer Credit Association and

the Finance and Leasing Association, available at: http://www.oxera.com/.

Persky, J. (2007). “From Usury to Interest”, Journal of Economic Perspectives 21(1), pp.

227-236.

Petersen, M., Rajan, R. (1994). “The Benefits of Lending Relationships: Evidence from

Small Business Data”, Journal of Finance 49(1), pp. 3-37.

Pikulska-Robaszkiewicz, A. (2000). “L'usure dans la législation des empereurs chrétiens”,

in: Au-delà des frontières, pp715-734.

Plender, R. (1991). “The European Contracts Convention, The Rome Convention on the

Choice of Law for Contracts” p. 127.

Policis (2004a). “The Impact of interest rate controls in other countries. Consumer

research to support DTI.”

Policis (2004b). “The Effect of Interest Rate Restrictions on Other Countries”, DTI,

London.

Policis (2006a). “Economic and Social Risks of Consumer Credit Market Regulation”,

London.

Policis (2006b). “Illegal lending in the UK. Research report.” URN 06/1883 (with Personal

Finance Research Center, published by Department of Trade and Industry, DTI), London,

available at: http://www.berr.gov.uk/files/file35171.pdf.

iff/ZEW – Final Report on interest rate restrictions in the EU 339

Posner, E. (1995). “Contract Law in the Welfare State: A Defence of the

Unconscionability Doctrine, Usury Laws, and Related Limitations on Freedom of

Contract”, Journal of Legal Studies 283 et seq.

PriceWaterhouseCoopers (PwC) (2008). “Precious Plastic 2008 Consumer Credit in the

UK”.

Project Associates (2009). “Briefing note on interest rate ceilings”.

Pruski, J., Żochowski, D. (2006). “Changes in the financing structure of the real economy

in Poland - challenges for the banking sector”, BIS Papers chapters, in: Bank for

International Settlements (ed.), The banking system in emerging economies: how much

progress has been made?, volume 28, pages 313-25 Bank for International Settlements.

Rabin, M. (2002). “A Perspective on Psychology and Economics”, European Economic

Review 46, pp. 657–685.

Ramsay, I. (1995) “Consumer Credit Law, Distributive Justice and the Welfare State” 15

OJLS 177 at 192.

Ramsay, I. (2010) “'To heap distress upon distress?: Comparative reflections on interestrate

ceilings”, University of Toronto Law Journal, Vol 60, No.2, pp. 707-730.

Reich, N. (1993). “Europäisches Verbraucherschutzrecht”, Baden-Baden.

Reifner, U (1992). “Das Zinseszinsverbot im Verbraucherkredit”, Neue Juristische

Wochenschrift pp 227-343.

Reifner, U., Ford, J. (eds.) (1992) “Banking for People”, Vol. 2: Overindebtedness,

Unemployment and Policy Responses - National Reports, de Gruyter: Berlin, New York

1992.

Reifner, U., Wüst, M., Haidar, L., Bonhomme, C. (1998). “Harmonisation of Cost

Elements of the Annual Percentage Rate of Charge, APR”, DG SANCO contract No. AO-

2600/97/000169, Hamburg.

Reifner, U. (1999). “Good Faith: Interpretation or Limitation of Contracts? The Power of

German Judges in Financial Services Law”, in: Brownsworth, R, Hird, N.J., Howells, G.

“Good Faith in Contract”, pp 269 ff.

Reifner, U., Niemi-Kiesiläinen, J., Huls, N., Springeneer, H. (2003). “Law and

Overindebtedness in the European Union - Study on consumer and personal bankruptcy

law in 15 European States”, study for DG Health and Consumer Protection Directorate-

General (Contract Reference No. B5-1000/02/000353) = “Overindebtedness in European

Consumer Law – Principles from 15 European States”, BoD Norderstedt 2010.

Reifner, U., Knobloch, M., Laatz, W. (2007). “iff report on over-indebtedness 2007”,

available at: http://news.iff-hh.de/media.php?t=media&f=file&id=2385.

Reifner, U., Knobloch, M. (2009). “Access to Credit for Poor People in Germany - The

impact of usury ceilings as well the bank monopoly in consumer credit extension on

consumers with moderate means”, available at http://www.responsiblecredit.

net/media.php?t=media&f=file&id=3837.

Reifner, U., Nogler, L. (2009). “Lifetime Contracts – Rediscovering the Social Dimension

of the Sales Contract Model”, in: Jubilee Thomas Wilhelmsson, Juridiska Föreningen i

Finland (JFT) Vol. 3/4 2009 pp 437–455.

Reifner, U., Nogler, L. (2010). “Social contracts in the light of the Draft Common Frame

of Reference for a future EU Contract Law”, in: Antoniolli, L., Fiorentini, F. (eds) Common

Core Evaluating Group, A Factual Assessment of the Draft Common Frame of Reference,

pp 365 - 407 = C.S.D.L.E. "Massimo D'Antona (2010) "Working Papers.INT - 80/2010,

340 iff/ZEW – Final Report on interest rate restrictions in the EU

available at http://www.lex.unict.it/eurolabor/ricerca/wp/int/nogler-reifner_n80-

2010int.pdf.

Reifner, U., Knobloch, M., Knops, K, (2010). „Restschuldversicherung und

Liquiditätssicherung“, BoD Norderstedt.

Saare, K., Sein, K., Simovart, M.A. (2010). “Protection of Consumer Rights in SMS Loan

Agreements" (in English), European Review of Private Law 2010 (18).

Schufa (2010). “SCHUFA Kredit-Kompass 2010: Empirische Indikatoren der privaten

Kreditaufnahme in Deutschland – Auswirkungen der Wirtschaftskrise auf den

Konsumentenkredit”.

Seckelmann, R. (1989). “Zinsen in Wirtschaft und Recht : Einführung in die

Finanzmathematik als logische Grundlage des Finanzrechts, in Anwendungen in der

Wirtschaft und Besonderheiten im Geldgewerbe ; Analyse des gegenwärtigen

Finanzrechts”, Frankfurt am Main, Knapp.

Sievi, F., Sievi, C.R. (1980). “Effektivzinssätze für Ratenkredite mit monatlichen Raten”

,Gillardon-Verlag, Bretten.

Skiba, P., Tobacman, J. (2007). “The Profitability of Payday Loans” Working Paper,

Vanderbilt University.

Smith, P.F. (1970). “Pricing Policies on Consumer Loans at Commercial Banks“, Journal

of Finance 25, pp. 517-525.

Spina, R. (2008). “L' usura Padova”, CEDAM.

Standard & Poors (2005). “Dutch RMBS Market Overview And Criteria”.

Stango, V., Zinman, J. (2010a). “Exponential Growth Bias and Household Finance”

Journal of Finance 64 (6), pp. 2807-2849.

Stango, V., Zinman, J. (2010b). “Fuzzy Math, Disclosure Regulation, and Market

Outcomes: Evidence from Truth-in-Lending Reform” Review of Financial Studies,

forthcoming.

Statistisches Bundesamt (2009). “Statistik zur Überschuldung privater Personen“,

Wiesbaden.

Stegman, M.A., Faris R. (2003). “Payday Lending: A Business Model that Encourages

Chronic Borrowing” Economic Development Quarterly, 17(1): 8–32.

Stegman, M.A. (2007). “Payday Lending”, Journal of Economic Perspectives 21(1),pp.

169–190.

Stiglitz, J.E., Weiss A. “Credit Rationing in Markets with Imperfect Information”,

American Economic Review 71(3), pp. 393-410.

Taffin, C, Vorms, B. (2007). “Elargir L'acces au credit au logement des emprunteurs

atypiques“.

Temin P., Voth, J. (2008). “Interest Rate Restrictions in a Natural Experiment: Loan

Allocation and the Change in the Usury Laws in 1714”,The economic journal, 118, pp

743-758.

UK British Bankers’ Association (2005). “BBA Statement of Principles”, available at

http://www.responsible-credit.net/index.php?id=1980&viewid=43128.

Vandone, D. (2009). “Consumer Credit in Europe: Risks and Opportunities of a Dynamic

Industry” Phyisica Verlag, Berlin/Heidelberg.

iff/ZEW – Final Report on interest rate restrictions in the EU 341

Varian, H. (2003). "Intermediate Microeconomics: a modern approach." 6th ed. New

York: W.W. Norton & Company.

Veritec Solutions (2005a). “Florida Trends in Deferred Presentment, State of Florida

Deferred Presentment Program”.

Veritec Solutions (2009). “Oklahoma Trends in Deferred Deposit Lending”, Oklahoma

Program Status Report.

Villegas, D.J. (1982). “An Analysis of the Impact of Interest Rate Ceilings” The Journal of

Finance 37(4), pp. 941-954.

Villegas, D.J. (1987). “The Impact Of Usury Ceilings On Revolving Credit” Economics

Letters 23 ,pp. 285-288.

Villegas, D. J. (1989). “The Impact of Usury Ceilings on Consumer Credit”, Southern

Economic Journal 56(1), pp. 126-141.

Vitoria, F. (2006). “De Contratos y usura de Pamplona”, Ediciones Universidad de

Navarra.

Wagenaar W.A., Sagaria S. (1975). “Misperception of Exponential Growth” Perception

and Psychophysics 18, pp. 412-422.

Warren, E., Tyagi, A. (2003). “The Two Income Trap” New York: Basic Books, pp. 144-

52.

Warren, E. (2007). “Unsafe at any Rate.” DEMOCRACY: A Journal of Ideas 8.

Weinstein, N.D. (1980). “Unrealistic Optimism About Future Life Events.” Journal of

Personality and Social Psychology 39 (5), pp. 806-820.

Whitford, B., Ramsay, I., Niemi-Kiesiläinen, J. (eds). (2009). “Consumer Credit, Debt and

Bankruptcy: Comparative and International Perspectives” Hart Pub. UK.

Wilhelmsson, T. (1994). “Social Contract Law and European Integration” p 163.

WOCCU (2008) “International Credit Union Consumer Protection Principles”, available at

http://www.responsible-credit.net/index.php?id=1980&viewid=43128.

Wolken, J. D., Navratil, F.J. (1981). “The Economic Impact of the Federal Credit Union

Usury Ceiling” Journal of Finance 36(5), pp.1157-1168.

Yang, S., Markoczy, L., Min, Q. (2007). “Unrealistic optimism in consumer credit card

adoption” Journal of Economic Psychology 28, pp. 170–185.

Zajączkowski, S., Żochowski, D. (2006). “The distribution and dispersion of debt burden

ratios among households in Poland and its implications for financial stability” IFC Bulletin

no. 28.

ZDP (2009), “Pressemappe des Zentralverbands des Deutschen Pfandkreditgewerbes

2009”, available at:

http://www.pfandkredit.org/dl.php?download=Pfandkredit_Pressemappe_2009_aktuell.p

df.

Zeldes, S.P. (1989). “Consumption and Liquidity Constraints: An Empirical Investigation"

Journal of Political Economy, Vol. 97 (2), pp. 305-346.

Zinman, J. (2008). “Restricting Consumer Credit Access: Household Survey Evidence on

Effects Around the Oregon Rate Cap” Journal of Banking and Finance 34(3), pp. 546-556.

342 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex II: Mandate of the European Commission and objectives

This study is intended to broaden the European Commission’s knowledge of interest rate

restrictions in place across Member States, and help it to form an assessment of the

economic, financial, and social impacts of these.

According to the terms of reference of the study, the objectives are to:

Identify the different forms that interest rate restrictions may take,

Provide an inventory of the Member States that apply them,

Assess the economic, financial, and social impacts on identified stakeholder

groups and on the functioning of the single market.

There are a wide range of possible forms of interest rate restrictions that fall within the

scope of the study, including:

Rate ceilings (caps),

Limits on interest rate variability, and

Restrictions on the use of compound interest rates.

The stated objective of the application of interest rate restrictions by a number of

Member States is to protect borrowers against exorbitant charging by lenders that could

make repayment of loans difficult or impossible for some of them. Rate ceilings are

perceived to prevent the charging of unreasonable or excessive interest rates, while the

restrictions on interest variability aim to shield borrowers from large shifts in interest

rates. The rules on compound interest rates are often designed to impose restrictions on

the application of compound interest290. In this sense, rate restrictions are designed to

protect consumers, either in general or specific sets of consumers who have a weak

bargaining position.

However, the existence of interest rate restrictions may constitute an impediment to

product innovation, product diversity, and cross border activity of lenders. Certain

products, such as equity release products, may become more difficult to market given

that they are based on (regular) interest compounding. In addition, certain borrowers

with high risk profiles, including many self-employed and SMEs, may theoretically be

prevented from having access to credit as lenders are unable to charge them risk-based

rates, since doing so would exceed the rate ceiling.

Furthermore, the existence of interest rate restrictions in some Member States could also

theoretically dissuade lenders from other Member States from offering their services in

those countries. It follows that interest rate restrictions may therefore limit, at least in

theory, the range of customers served, consumer choice, and cross-border activity.

Interest rate restrictions may therefore have negative impacts on the businesses, credit

markets themselves, and wider economic interests.

In the White Paper on the Integration of EU Mortgage Credit Markets and its

accompanying Impact Assessment, the Commission identified interest rate restrictions as

an issue that warrants further examination, and for an assessment to be undertaken

which includes the impacts on both consumer and provider stakeholders including the

wider social and economic environment.

290 Study on Equity Release Schemes, p. 111, at http://ec.europa.eu/internal_market/finservicesretail/

credit/mortgage_en.htm#studies.

iff/ZEW – Final Report on interest rate restrictions in the EU 343

On the basis of the findings of the study, the Commission has stated that it will examine

whether it is necessary to take any action on the issue at the EU level.291

The conduct of the study is divided into two main parts, a legal part, and an economic

part, the objectives for which have been set out by the Commission as below.

Objectives of Task 1: Legal part

Task n°1: Identify the different types of interest rate restrictions, the Member States that

apply them, and their reasons for doing so.

1) Identify all the different forms that interest rate restrictions may take and the levels at

which they are set and how they function, including interest rate ceilings (caps),

interest rate variations, and compound interest;

2) Identify those Member States that apply statutory or administrative provisions, case

law, or have self regulation or other instruments that constitute interest rate

restrictions or have an effect equivalent to interest rate restrictions;

3) Inquire for the reasons and concerns that led the competent authorities to adopt such

provisions, particularly any data relating to abusive interest rate charging in the past.

Objectives of Task 2: Economic part

Task n°2: Analyse the economic, financial, and social impacts of interest rate restrictions

on various stakeholders.

1) The study will analyse all the existing and potential economic, financial, and social

impacts (in quantified terms) of provisions on interest rate restrictions existing in the

Member States on: consumers in general, and specific classes of consumers such as

the self-employed, microenterprises, and SMEs; credit providers and, where relevant,

credit intermediaries.

2) The study must further analyse the economic, financial, and social impact that

provisions on interest rate restrictions have on: the pursuit of cross-border business

by credit providers; access to credit products and product diversity; the price paid by

consumers; the functioning of the Single Market.

3) Finally, the study shall provide an analysis of how the different types of interest rate

restrictions identified in the 27 Member States compare to each other in terms of

efficiency and effectiveness.

291 As stated in the European Commission’s terms of reference. To see a summary of the Commission's Call for

Tender (MARKT/2009/08/H) see Contract notice. (In order to facilitate the finding of material underlying this

research iff has developed a blog where relevant materials can be found under http://irr-blog.responsiblecredit.

net/).

344 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex III: Methodology of the research

Tasks, existing literature and stakeholders

Approach to the literature, data review and stakeholder meetings

Direct contact with stakeholders was an integral part of the research and covers a broad

range of stakeholders as identified in the next Annexes of this report.

The research has taken into account existing studies completed at both international and

EU level, and has reported on all existing research, which it and its stakeholders are

aware of at the national level. Despite some informative research on national details on

the subject in countries such as France, Poland, the Netherlands Slovakia or the UK,

there is an evident shortage of existing research on the issue of interest rate restrictions

from the rest of the EU Member States. While the subject has featured in much literature

on price controls on credit in the US and in the context of the microfinance agenda in the

area of development economics, there is as of yet no such report for the EU. One

significant concern with the state of the research on this issue, is that one particular

forerunner for investigation of the issues, with very affirmative statements and

conclusions with regard to the effects of IRR generally and in several specific EU

countries, has managed to contaminate the objectivity of subsequent research. Empirical

studies have a lot of clout and policy makers should base their decisions on arguments

backed by evidence and scientific research. However, the danger of too little such

research is that the results of the few existing ones are increasingly referred to without

questioning the validity of the results. The prior absence of such studies at the national

level makes the current research more difficult, however, our survey of stakeholders in

all Member States has provided us with valuable material to assess the impact of IRR on

credit markets and stakeholder groups.

A variety of different data sources has been used, first to comprehend the idiosyncrasies

of each Member State’s credit market and to assess the possible impact IRR may have

had and the potential effects these restrictions could have when applied to other Member

State markets. We have restricted the selection of data to reliable industry and public

sources that allow for comparison across the Member States and have also incorporated

some national sources for specific phenomena which are measurable.

We have also reviewed the wider literature. This is primarily focused on the historical

tradition of usury laws, and there is also a significant literature on the operation of usury

ceilings within the U.S, where there has been a long running debate over the past 30

years on this issue. Material found from the various scientific and regulatory websites has

been downloaded and collated within a bespoke research blog.

Telephone interviews with certain stakeholders were conducted and a number of face-toface

meetings were organised with certain specific stakeholders. While stakeholders vary

as to the usefulness and knowledge of the subject matter that they have been willing to

contribute to the research, stakeholder involvement with the issue of IRR and its

potential usefulness and limitations etc.. are a worthwhile endeavour in itself. In addition

to the bilateral communication with stakeholders, part of the research team has

presented the study to the Forum of users experts in the area of financial services (FINUSE)

set up by the European Commission in order to alert them to the study and seek

their involvement. This meeting was held on 4 May and attended by Sebastien Clerc-

Renaud from iff and by Damon Gibbons from the UK’s Centre for Responsible Credit, who

is one of the national experts supporting the study. Both explained the nature and details

of the study and FIN-USE members were encouraged to participate by contributing a

survey response for themselves.

iff/ZEW – Final Report on interest rate restrictions in the EU 345

Methodology

In the first phase of the study the research team, which comprises both an economic

team and a legal team, has:

Conducted a search of, and examined the, existing literature and data sources,

Undertaken initial interviews and discussions with specialists, which has helped to

determine the relevant areas for further investigation and helped to define the

categories of credit and forms of IRR that lie within the scope of the project,

Designed a survey for the collection of legal detail from the selected legal experts

in all Member States,

Developed a number of hypotheses concerning the economic and social impacts of

IRR,

Designed surveys for stakeholders to obtain their views on these hypotheses and

to seek any further evidence that they may have on these issues.

The core of the empirical evidence in this project is therefore derived from an

international survey based on questionnaires sent to the following stakeholders:

Individual credit providers (Provider Questionnaire - PQ),

Credit provider trade associations, regulators and supervisors, consumer

organisations and other interest groups (Stakeholder Questionnaire - SQ),

and Legal expert questionnaire.

A particular focus of the initial work was on understanding and collecting the details of

the legal and regulatory underpinnings, techniques, and mechanisms behind the different

forms that IRR can take. Subsequently full research attention was dedicated to collecting

the views of stakeholders through the designed survey and telephone conversations.

Individuals and organisations interviewed

Legal experts

Overseeing the legal work on this project were Prof. Udo Reifner together with attorney

at law Michael Knobloch with the collaboration of Prof. Iain Ramsay from the University of

Kent. In addition, additional national legal experts participated in the study. These have

been selected on the basis of their prior contributions to national research on consumer

credit issues, and their efficient collaboration in previous European research projects.292

Individual providers

A Provider Questionnaire was designed and sent to credit providers in order to assist with

the economic analysis in this project. The questionnaire was designed to collect insights

about the experience of individual players in the consumer credit market with regard to

the effects of the interest rate restrictions, or lack of thereof, on the different segments

of the market.

292 The list of legal experts that have contributed their expertise on their country is available in the Annex.

346 iff/ZEW – Final Report on interest rate restrictions in the EU

The questionnaire was first sent to providers in the six chosen case-study countries, and

was subsequently distributed to other providers from different Member States through

the main EU level provider associations who disseminated the Provider Questionnaire to

their own national association members, who past these on to their own members, the

individual providers themselves. The introduction of the questionnaire was tailored to the

country in which the providers were operating and the Provider Questionnaire can be

found in the Annex.

Provider associations and consumer organisations

In order to collect more general provider opinions on IRR, the research team has also

approached trade associations for the different credit market providers through the use

of the Stakeholder Questionnaire.

We have asked national as well as European associations for their views. At the European

level, we have asked the European Mortgage Federation (EMF) as well as Eurofinas to

complete a questionnaire on behalf of their members, and the research team met with

the EBIC Consumer Credit Working Group on the 3rd of June to explain the aim of the

study and the reasons why provider involvement was important. The members of the

working group agreed to encourage their members (provider associations themselves) to

complete a Stakeholder Questionnaire, as well as asking them to forward the Provider

Questionnaire to their own members.

In addition, the Stakeholder Questionnaire was also sent to consumer associations and

other stakeholders associated with the user side, and, as mentioned previously, we have

also engaged FIN-USE in the study.

Public authorities - Government and regulators

We have sent the Stakeholder Questionnaire to all identified public authorities in each

Member State. There is significant differentiation in the national regulatory structures

across the EU, although there is now also a clear trend towards the consolidation of

supervisory authorities293. As a result, we have had to take into account three main

systems of regulation:

The widespread ‘sectoral model’ is used in 11 Member States (Greece, Spain, Cyprus,

Lithuania, Slovenia, Bulgaria, Romania and with some variations in France, Portugal,

Finland and Luxembourg) and is based on a separate authority for each sector (banking,

securities and insurance).

The so-called ‘single regulator model’, preferred by the UK and some of the smaller

new EU Member States. The single regulator may, however vary. For example, all the

financial supervision functions can be transferred to the national central bank, as in the

case of the Czech Republic and Slovakia, or a new single authority separate from the

central bank can be created, as in the UK, Estonia, Latvia, Hungary, Malta, and now

Poland). In fact, although the term ‘single regulator’ is used, this can be misleading. For

example, the UK, which is most associated with the term, still separate banking,

insurance, and mortgage regulation (Financial Services Authority) from consumer credit

regulation (Office of Fair Trading).

In between these two models, the ‘twin peaks model’ (practiced in the Netherlands

and to some extent in Italy) allocates responsibilities according to supervisory objectives

(ie. prudential supervision and conduct of business regulation are the responsibility of

293 European Central Bank, Recent Developments in Supervisory Structures in EU and acceding Countries 2006;

ibid. EU Banking structures: The impact of ageing on EU banks 2006.

iff/ZEW – Final Report on interest rate restrictions in the EU 347

two different authorities). Elements of this model are also present in the French and

Portuguese supervisory structures.

Questionnaires

We present a short section on the process, which may assist in understanding the

questionnaires.

Legal Expert Survey

In order to obtain a legal inventory of existing IRR in the EU, iff designed a questionnaire

for its legal experts. The Legal Questionnaire (LQ), not reproduced in the Annex to avoid

confusion with the Stakeholder Questionnaire, contained questions broadly structured by

IRR starting with direct restrictions on the interest rate (Sections A to C), indirect

restrictions on other credit parameters with an equivalent effect to IRR (Sections D to E)

ending with a set of questions on sanctions and control mechanisms (Sections F to G).

In more detail, the introductory questions concerned the economic (Q1-5), political and

social background (Q6-16); Section A) restrictions regarding contractual interest rates

through direct ceilings and strict "usury ceilings"/“rate caps” on borrowing rate or Annual

Percentage Rate of Charge (APR) (Q17-34); B) restrictions on default interest and

penalties (Q35-44); C) limitations to the variability of rates (Q45-48); D) restrictions

regarding other cost elements emerging in connection with the credit contract (Q49-54);

E) indirect ceilings concerning other contractual conditions, calculation methods, size,

time, amortisation of instalments and other credit terms (Q55-77); F) sanctions and

control mechanisms (Q78-85); G) restrictions affecting the SME and micro-lending

market beyond the core consumer markets reported on until then and summary views

(Q86-92).

The legal results were subsequently sent to the public authorities for verification and

some minor adjustments and corrections were received from them.

The Stakeholder Questionnaire

The results of these surveys were received at different times and the deadline given to

respondents was extended twice in order to allow for the maximum of stakeholder

responses to be taken into consideration in the analysis of IRR for this report. The

approach behind the design of the questionnaires and details of how the different

questions from the survey relate to the hypotheses of interest to the study were

explained to the European Commission at the interim stage of the research.

By conducting the stakeholder survey, we were able to obtain insights about the practical

effects of IRR, in both the formal rules and regulations and also other less formal

mechanisms that moderate consumer credit prices (for example moral consensus or

ethical practice). The targeted stakeholders include associations of providers of consumer

credit at the national and European level, consumer protection organisations and public

authorities including government officials and regulatory bodies. The respondents were

asked both closed and open questions. In the case of the latter they were given an

opportunity to write their detailed responses in the entry fields provided (reproduced in

part in the following annexes). Responses also helped the study by providing references

and links to information and data sources.

The Stakeholder Questionnaire was structured in four sections. The first one starts with

questions concerning details and views on the forms and features of IRR (Part A: Q1.1-

1.36). This section aimed at strengthening our understanding of the regulatory

framework of those markets. It also asked questions which are targeted at the

effectiveness of interest rate restrictions of both prevailing rates as well as hypothetical

legal rates.

348 iff/ZEW – Final Report on interest rate restrictions in the EU

The second section dealt with issues related to the extent of private over-indebtedness

(Part B: Q2.1-2.5). While some questions asked about determinants of overindebtedness

in more detail, other questions targeted current trends in the respective

country. The latter questions being useful in a cross-sectional comparison across

countries with different legal situations on interest rate restrictions.

The third section addressed access to consumer credit (Part C: Q3.1.-3.6). Several

aspects centred on the level of credit access (this could then be compared in a cross

section of countries) and evidence of possible credit exclusion due to interest rate

restrictions.

The fourth section included a set of questions on the characteristics of the different credit

markets (Part D: Q4.1-4.14). It allowed us to better understand whether there are

different market structures, including international integration, competition as well as

market vulnerability in times of crisis in a cross-country comparison in our case study

countries.

The Provider Questionnaire

Since filling out such questionnaires can be very burdensome for providers, there was a

general risk that more information will originate from those with a political interest in EU

policies. The letter from the Commission accompanying the questionnaire demonstrated

the political importance of responding. More details on the survey of providers including

the questionnaire itself is available in Annex XII: Provider Questionnaire.

Development of the questionnaires and testing

The questionnaire was developed along the following steps:

1. In the first stage, IRR were defined theoretically and the key research questions

were drawn up.

2. In a second step, hypotheses concerning the possible outcome of this research in

relation to dissemination, barriers, incentives, features, risks, etc., were discussed

with economists, sociologists and lawyers based at or cooperating with iff.

3. These hypotheses were operationalised into research dimensions and subdimensions.

The questionnaire was tested first in Germany and then discussed with experts from the

UK and the Netherlands.

The chart, below, provides an overview of the methodology for the study.

iff/ZEW – Final Report on interest rate restrictions in the EU 349

Figure 146: Process of the surveys for the empirical work

Legal Survey

Aim: Obtaining existing information and details of legal and regulatory situation

Target: Law experts, public authorities, financial regulators, trade associations

“Stakeholder Questionnaire”

“Countries with/without significant Restrictions”

Public authorities, regulators, provider associations,

consumer associations

Expert Interviews and on-site stakeholder meetings

Public authorities, law experts, provider and consumer associations

Study on Interest Rate Restrictions

Telephone interviews, mobilisation of stakeholders, collection of existing data and

studies etc…

Follow-up Questionnaire to complete

inventory of IRR and their equivalent,

and provide information and data for

calculation of economic and social impact

“Provider

Questionnaire”

Providers

350 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex IV: Stakeholders contacted

Country Regulator, Authorities, Government agencies

Austria Oesterreichische Nationalbank; Bundesministerium für Finanzen; Finanzmarktaufsicht

Belgium Banque Nationale de Belgique; Commission Bancaire, Financière et des Assurances; Fonts de

Traitement du Surendettement

Bulgaria Bulgarian National Bank; Ministry of Finance

Cyprus Central Bank of Cyprus; Ministry of Finance; Authority for the Supervision and Development of

Cooperative Societies; Ministry of Commerce, Industry and Tourism; Authority for the Supervision

and Development of Cooperative Societies

Czech

Republic

Ministry of Industry and Trade; Ministry of Finance; Czech National Bank

Denmark Danish Financial Authority (Finanstilsynet); Danish Mortgage Credit Complaint Board; National

Bank of Denmark; Ministry of Justice

Estonia Bank of Estonia; Consumer Protection Board of Estonia; Financial Supervision Authority of

Estonia; Ministry of Finance; Ministry of Justice

Finland Ministry of Justice; Financial Supervisory Authority; Bank of Finland

France Ministry of Justice; French Treasury and Economic Policy Directorate General; Banque de France;

Commission bancaire; Comité consultatif du secteur financier; Direction générale de la

concurrence, de la consommation et de la répression des fraudes

Germany Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin); Deutsche Bundesbank;

Bundesministerium der Justiz; Wettbewerbszentrale

Greece Bank of Greece

Hungary National Bank of Hungary; Hungarian Financial Supervisory Authority; Ministry of Finance

Ireland Irish Department of Finance; Irish Financial Services Regulatory Authority

Italy Ministry of the Economy and Finance; Banca d'Italia

Latvia Ministry of Economy; Bank of Latvia; Financial and Capital Market Commission

Lithuania Bank of Lithuania; State Consumer Rights Protection Authority; Ministry of Finance

Luxembourg Banque de Luxembourg; Ministère des Finances

Malta Central Bank of Malta; Ministry of Finance & Economic Affairs; Malta Financial Services Authority

(MFSA)

Netherlands Netherlands Authority for the Financial Markets (AFM); Ministry of Finance; De Nederlandsche

Bank

Poland Office of Competition & Consumer Protection (OCCP); National Bank of Poland; Ministry of Finance

Portugal Banco de Portugal

Romania National Bank of Romania; National Authority for Consumers Protection

Slovakia National Bank of Slovakia; Ministry of Finance

Slovenia Ministry of Economy of Slovenia; Bank of Slovenia; Ministry of Finance

Spain Banco de Espana

Sweden Swedish Financial Supervisory Authority; Ministry of Integration and Gender Equality; Ministry of

Finance; Kronofogden; Sveriges Riksbank

United

Kingdom

Financial Services Authority (FSA); Office of Fair Trading; Financial Services Authority (FSA);

Department for Business, Innovation & Skills; HM Treasury; Competition Commission UK; Bank of

England; Scottish Government

iff/ZEW – Final Report on interest rate restrictions in the EU 351

Country Provider Associations

Austria Verband Österreichischer Banken & Bankiers; Verband der österreichischen Landes-

Hypothekenbanken; Österreichischer Genossenschaftsverband (ÖGV)

Belgium Febelfin - Fédération belge du secteur financier; Union Professionnelle du Crédit (UPC); PBA -

Private Bankers Association Belgium

Bulgaria Association of Banks of Bulgaria; Bulgarian Association of Leasing

Cyprus Association of Cyprus Commercial Banks; Cyprus Banks Association

Czech Rep. ACSS - Asociace českých stavebních spořitelen; Czech Banking Association (CBA); Czech Leasing

and Finance Association (CLFA)

Denmark Realkreditraadet (Association of Danish Mortgage Banks); Association of Danish Finance Houses

(Finans og Leasing); Realkreditraadet (Association of Danish Mortgage Banks); Danish Bankers

Association

Estonia Estonian Banking Association

Finland Federation of Finnish Financial Services

France Association française des sociétés financières (ASF); Fédération Bancaire Française (FBF); Caisse

nationale des Caisses d´Epargne (CNCE)

Germany Verband deutscher Pfandbriefbanken (vdp) e.V.; Bundesverband der Deutschen Volksbanken und

Raiffeisenbanken (BVR); Verband der Privaten Bausparkassen e.V.; Bankenfachverband e.V.;

Bundesverband Öffentlicher Banken Deutschlands e.V. - VÖB; KfW Bankengruppe; Verband

deutscher Pfandbriefbanken (vdp) e.V.; Gesamtverband der Deutschen Versicherungswirtschaft

e.V. (GDV); Rheinischer Sparkassen- und Giroverband; Bankenfachverband e.V.; KfW

Bankengruppe; Bankenfachverband e.V.; Deutscher Sparkassen- und Giroverband (DSGV);

Bundesverband Deutscher Banken; Verband der privaten Bausparkassen (vdpb); Zentraler

Kreditausschuss (ZKA); Zentralverband des Deutschen Pfandkreditgewerbes e.V.; Deutscher

Sparkassen- und Giroverband (DSGV)

Greece Hellenic Bank Association (HBA)

Hungary Association of Hungarian Mortgage Banks; Hungarian Banking Association; Hungarian Real Estate

Association

Ireland Irish Banking Federation; Irish Mortgage Council (IMC); Irish Finance Houses Association; Irish

League of Credit Unions

Italy Italian Association for Consumer Credit and Mortgages (ASSOFIN); Associazione Bancaria Italiana

(ABI); Associazione Italiana del Factoring - ASSIFACT; Unione finanziarie italiane (UFI);

Associazione Italiana del Leasing – ASSILEA; Unione Nazionale Imprese Recupero crediti e

informazioni Commerciali

Latvia Ombudsmen of the Association of Latvian Commercial Banks; Association of Commercial Banks

Lithuania Lithuanian Bankers’ Association; Association of Lithuanian Banks; Lithuanian Credit Unions

Luxembourg Luxembourg Bankers’ Association (ABBL)

Malta Malta Commercial Banks Association (MCBA); Finance Malta

Netherlands Nederlandse Vereniging van Banken (NVB); Nederlandse Vereniging von Volkskrediet (NVVK);

Dutch Banking Association; Vereniging van Financieringsondernemingen in Nederland (VFN)

Poland Conference of Financial Companies in Poland (KPF); Polish Bank Association; Mortgage Credit

Foundation; Conference of Financial Companies in Poland (KPF); Polish Cooperative Bank

Association

Portugal Associacao Portuguesa de Bancos; ASFAC; Associacao Portuguesa de Bancos

Romania Romanian Banking Association; ALB

Slovakia Slovak Banking Association

Slovenia Bank Association of Slovenia

Spain €uroMontepio; Asociación Nacional de Establecimientos Financieros de Crédito (ASNEF);

Asociación Hipotecaria Española (AHE); Confederación Espanola de Cajas de Ahorros; Associacion

Espanola de Banca

Sweden Swedish Bankers' Association (Svenska Bankföreningen) AFINA; Swedish pawnbrokers

association; Finansbolagens Förening; Swedish Bankers' Association (Svenska Bankföreningen)

AFINA

UK The UK Cards Association Limited; British Cheque Cashers Association (BCCA); British Bankers'

Association (BBA); Consumer Finance Association (CFA); Building Societies Association; Council of

Mortgage Lenders; Consumer Credit Trade Association (CCTA); Consumer Credit Association UK;

Finance & Leasing Association (FLA); ABCUL (The Association of British Credit Unions Limited);

National Pawnbrokers Association (NPA)

EU European Savings Banks Group; Eurofinas (The European Federation of Finance House

Associations); European Banking Federation; EBIC European Banking Industry Committee;

Austrian Savings Banks Association; European Federation of Building Societies; European

Mortgage Federation; European Financial Services Round Table (EFR)

352 iff/ZEW – Final Report on interest rate restrictions in the EU

Country Consumers Associations

Austria Arbeiterkammer Wien; Verein für Konsumenteninformation (VKI)

Belgium Test Achats; Centre de recherche et d'information des organisations de consommateurs (CRIOC);

Verbruikersateljee vzw - Beweging voor de kleine consument

Bulgaria Bulgarian National Consumer Association; Consumer Center for Information and Research;

Federation of Consumers of Bulgaria

Cyprus Cyprus Consumer Association

Czech

Republic

Consumers' Defence Organisation of the Czech Republic; Czech Consumers Association (SCS);

Debt Advisory Center

Denmark Forbrugerradet (Danish Consumer Council)

Estonia Estonian Consumers' Union

Finland Kuluttajaliitto; Kuluttajavirasto Konsumentverket; National Consumer Research Center

France UFC-Que Choisir; Association Léo Lagrange pour la Défense des Consommateurs (ALLDC);

Organisation Générale des Consommateurs - OR.GE.CO; Association française des usagers des

banques (AFUB); ANIL (Association Nationale pour l'Information sur le Logement);

Consommation, Logement et Cadre de vie (CLCV); Fédération des Associations Indépendantes de

défense des Epargnants pour la retraite (FAIDER); SOS Familles Emmaüs; Association de

consommateurs Union féminie civique et sociale (UFCS); Secours Catholique; INDECOSA CGT;

Confédération Syndicale des Familles (CSF); Institut National de la Consommation (INC)

Germany Verbraucherzentrale Bundesverband (vzbv); Verbraucherzentrale Hamburg e.V.; Stiftung

Warentest

Greece EKPIZO Consumers' Association "Quality of Life"; Hellenic Ombudsman for Banking-Investment;

KEPKA - Consumers Protection Centre

Hungary National Association of Consumer Protection in Hungary; Association For The Hungarian

Consumers (AHC/FOME)

Ireland Financial services ombudsman; Consumers' Association of Ireland (CAI); Free Legal Advice

Centres Ltd. (FLAC)

Italy Altroconsumo; Adiconsum; ADUSBEF; Associazione Conciliatore Bancario Italia; Consumatori

Associati;

Latvia Consumer Rights Protection Center; Latvian National Consumers' Association

Lithuania Lithuanian Consumer Institute (LCI); Lithuanian National Consumer Federation

Luxembourg Union Luxembourgeoise des Consommateurs (ULC)

Malta Ghaqda tal-Konsumaturi; Maltese consumer association

Netherlands Consumentenbond; NIBUD (National Instituut voor Budgetvoorlichting)

Poland Polish Consumer Federation National Council; Association of Polish Consumers

Portugal DECO - Ass.Portuguesa para a Defesa do Consumidor; Centro de Arbitragem De Conflitos de

Consumo de Lisoboa; Associaçao Portuguesa de Direito do Consumo (APDC)

Romania Association for the Protection of Consumers (APC)

Slovakia Association of Slovak Consumers (Zdruzenie slovenskych spotrebitelov)

Slovenia Slovene Consumers Association

Spain Organización de Consumidores y Usuarios (OCU); Spanish Independent Financial Advisors’

Association; Confederación de Consumidores y Usuarios (CECU); ADICAE

Sweden Swedish Consumers' Association; Swedish Consumer Agency (Konsumentverket)

United

Kingdom

Community Development Financal Association; Advice UK; Consumer Credit Counselling Services;

Consumer Focus; Citizens Advice Scotland; Citizens Advice; Credit Action; Money Advice Trust;

The Financial Inclusion Centre; Money Advice Scotland; Financial Services Consumer Panel;

Community Finance Solutions; Which?

EU BEUC; ANEC

iff/ZEW – Final Report on interest rate restrictions in the EU 353

Country Other stakeholders

Finland Guarantee Foundation

France

CSC Commission de la Securite des Consommateurs; CLERSÉ; Association d'économie financière;

Conseil économique et social (CES); Association Française des Intermédiaires Bancaires (AFIB);

Audencia Nantes - School of Management

Germany Zentrum für Europäische Rechtspolitik an der Universität Bremen (ZERP); Creditreform; SCHUFA

Holding AG; Finpolconsult; GP Forschungsgruppe

Ireland Law Society of Ireland

Italy Unione Nazionale Imprese Recupero crediti e informazioni Commerciali

Luxembourg Ligue Medico-Sociale

Netherlands Dutch government; BKR (Dutch Bureau of Credit Registration)

Portugal Observatório do Endividamento dos Consumidores (OEC); Universidade de Coimbra (FEUC)

Romania Credit Bureau

Spain FinanzKontor; Instituto de Crédito Oficial (ICO)

Sweden Skatteverket

United

Kingdom

Fair Finance Ltd; Personal Finance Research Centre; Confederation of British Industry (CBI); Unite

the union; ACCA The Association of Chartered Certified Accountants; Toynbee Hall; New Local

Government Network (NLGN); NIACE (National Institute for Adult Continuing Education); Lending

Standards Board; London Rebuilding Society; UK Financial Inclusion Taskforce; UK National

Housing Federation; National Association of Credit Union Workers; Bassac; Barnardos; New

Economics Foundation; ACEVO - Association of Chief Executives of Voluntary Organisations

EU Association of Consumer Credit Information Suppliers (Accis); European Trade Union

Confederation (ETUC); CEPS (Centre for European Policy Studies)

354 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex V: Legal Experts

Country Legal Expert

Austria Prof. Dr. Michael Gruber (University Salzburg)

Belgium Prof. Dr. Evelyne Terryn (Univ. Leuven)

Bulgaria Ass. Prof. Ivan Mangatchev (New Bulgarian University)

Cyprus Ass. Prof. Nikitas Hatzimihail (University of Cyprus)

Czech Republic Petr Müller (Lawyer in Prague)

Denmark Susanne Nielsen (Danish Mortage Credit Complaint Board)

Estonia Mark Butzmann (Lawyer in Tallin)

Finland Riitta Kokko-Herrala (Kuluttajavirasto)

France Masset Denevre (Institut National de la Consommation)

Germany RA Michael Knobloch (iff)

Greece Melina Mouzouraki (Lawyer in Athens)

Hungary Csongor Buzády (Lawyer in Budapest)

Ireland Mel Cousins (Lawyer in Dublin)

Italy Prof. Diana Cerini (Milan University)

Latvia Theis Klauberg (Lawyer in Riga)

Lithuania Frank Heemann (Lawyer in Vilnius)

Luxembourg Elise Poillot (Université du Luxembourg)

Malta Dr. Grazio Mercieca (Consumer Affairs Council Malta)

Netherlands Tamara Madern (NIBUD), Prof. Dr. Nick Huls (Univ. Erasmus)

Poland Prof. Dr. Wlodimierz Szpringer (Warsaw School of Economics)

Portugal Joao Espanha (Espanha associados)

Romania Dr. Rodica Diana Apan (North University Baia Mare)

Slovakia Prof. Maria Patakyova (Comenius University Bratislava)

Slovenia Dr. Meta Ahtik (Ljubljana University)

Spain Dr. Elena Perez Carrillo (Universidade Santiago de Compostela)

Sweden Ann-Sofie Henrikson (Umea University)

United Kingdom Prof. G.A. Williams (University of Kent)

iff/ZEW – Final Report on interest rate restrictions in the EU 355

Annex VI: ECRI (2009) Dataset main definitions

The following definitions need to be mentioned in relation to the ECRI dataset (ECRI,

2009).

1. Credit to the non-financial business and household sector. Credit to the non-financial

business and household sector comprises all loans to households and NPISH (nonprofit

institutions serving households) and credit to non-financial corporations.

2. Total credit to households and NPISH. According to the European System of Accounts

1995, the household sector covers individuals or groups of individuals as consumers,

but also as entrepreneurs (ie. sole proprietorships and partnerships). Non-profit

institutions serving households are a separate institutional sector, although often

reported together with households. Unless stated otherwise, loans to NPISH are

included in the various categories of loans (consumer credit, housing loans and other

loans).

3. Consumer credit. Consumer credit corresponds to the outstanding amounts (stocks)

of loans at the end of the year granted by the resident MFI sector to resident

households and NPISHs for consumption purposes. Consumer credit notably includes

loans related to credit cards as well as overdrafts. Consumer credit by lender or by

type is not necessarily comparable across countries, as various country-specific

classifications and definitions apply.

4. Housing loans. Housing loans correspond to the outstanding amounts of loans at the

end of the year granted by the resident MFI sector to resident households and

NPISHs for housing purposes. The data series comprise loans on dwellings both

secured and unsecured. Breakdowns by maturity and currency are provided where

available.

5. Gross domestic product (GDP). Gross domestic product is the final result of the

production activity of resident producer units.

6. Final consumption expenditure. The final consumption expenditures of the household

sector correspond to the value of the consumption goods and services acquired

through purchases (social transfers in kind receivable from other institutional units

are excluded). According to the international principles of national accounting, final

consumption expenditure may take place on the domestic territory or abroad.

7. Gross disposable income. The gross disposable income of the household sector is

intended to measure the monetary income which can be used for consumption and

saving. It is a balancing item of the current income in the secondary distribution

income account of households and NPISH and derived from the balance of primary

incomes by adding or subtracting all social transfers (except social transfers in kind).

356 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex VII: Databases used in the analyses of over-indebtedness

In this Annex, data bases which cover indicators for over-indebtedness on the European

level are described more closely. It refrains from using national data sources because of

various statistical problems. These problems include: statistical categories and definitions

are not the same across countries; the methodology applied varies and the frequency of

data collection varies as well. These are strong arguments for using an international EU

survey to collect comparable statistics. Two surveys provide rare sources of comparable

information about levels of over-indebtedness across Europe.

EU SILC

The EU SILC is a pan-European panel survey collecting data on living conditions,

including social exclusion and poverty on an annual basis. The collected indicators play a

major role in tracking developments in individual Member States as well as in the fight

against poverty. Some of the questions in the EU-SILC survey can be classified as

objective questions, such as whether the household has been in arrears over the past 12

months, whereas others are based on households’ own assessments of their situation,

such as their ability to make ends meet. It mainly focuses on income and detailed income

components are collected mainly at personal level although a few components are

included in the household part.

Data collection for EU-SILC started in 2004. The survey then covered twelve EU 15

countries (Belgium, Denmark, Greece, Spain, France, Ireland, Italy, Luxembourg,

Austria, Portugal, Finland and Sweden) as well as Estonia, Norway and Iceland. The

remaining EU-25 countries joined the EU SILC in 2005. Bulgaria, Romania, Turkey and

Switzerland have launched SILC in 2006.

As EU SILC is based on the idea of a common “framework” and not a common “survey”

as was the case for its predecessor - the European Community Household Panel (ECHP) –

there are harmonised lists of annual variables, which are transmitted to Eurostat. This

setup is aimed at maximising comparability of the information produced. This way

problems with harmonisation of national data are largely avoided.

Measures that can be drawn from EU SILC to evaluate the over-indebtedness of

households are numerous: they comprise arrears, debt ratios as well as subjective

measures Data from EU SILC will be used at an aggregate level for the analysis of

probable impacts of regulatory decisions on over-indebtedness. Relying on an EU-wide

survey such as the EU SILC has the advantage that all countries are covered, the data

survey is run with the same frequency and that results are harmonised, allowing for a

high comparability of the results across countries. The indicator reflects repayment

problems and is generally judged to be good in terms of information content and

reliability.

The Survey on Household Finance and Consumption, discussed in chapter 4.5.2 might

provide further valuable insights.

Eurobarometer

The Eurobarometer survey is collected on behalf of the European Commission. Relying on

this cross-national survey avoids incomparability emanating from lacking harmonisation.

The Standard Eurobarometer survey series is a longitudinal study, designed to compare

and gauge trends within Member States of the European Union. Each survey consists of

approximately 1000 face-to-face interviews per Member State. Besides, special

irregularly repeated modules investigate topics such as environment, gender roles,

family, or financial services in a European perspective. Such a special Eurobarometer

iff/ZEW – Final Report on interest rate restrictions in the EU 357

module from Eurobarometer 60.2 and 63.2 will be used in the following to analyse credit

access.

The Eurobarometer 60.2 was conducted in fall 2003, the 63.2 in spring 2005. This round

diverged from the standard measures and queried, among others, the use of and

attitudes towards financial services. Respondents were asked about the type of bank

accounts they personally had. Demographic and other background information provided

include respondent’s nationality and household income.

358 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex VIII: Measures of over-indebtedness

For the analysis of the hypotheses it was necessary to determine an appropriate measure

for over-indebtedness. The following paragraphs discuss and classify possible measures

from which two indicators on the aggregate level– a subjective burden indicator

(households ability to make ends meet) and an objective over-indebtedness measure

(households in arrears with their payment obligations) – were chosen for the analysis.

Data

Data generally used to identify situations of financial difficulty can be classified into two

groups:294 Aggregate data provide indications regarding the pervasiveness of financial

difficulties amongst a particular population and are usually collected by national central

banks. Aggregate indicators of over-indebtedness quantify the size of the phenomenon

and can thus be helpful in temporal or cross-country comparisons. The disadvantage of

these data is that they do not always correctly identify situations of over-indebtedness

and that they are not available for all countries or can be derived consistently across

different countries. Furthermore, aggregate data have the problem of not capturing well

the individual circumstances of consumers:295 To determine the sustainable amount of

debt a household can bear, lending institutions draw on numerous characteristics of the

future borrower. In contrast, aggregate indicators establish thresholds to determine

over-indebtedness for all consumers. Such general measures fail to allow for the different

needs of consumers. People of different ages have different requirements for borrowing

depending on their time preferences and possibilities to consume out of current income

or accumulated wealth. Since an aggregate consumption/saving pattern cannot be

stipulated, Betti et al. (2007) conclude that there is “no simple aggregate measure of

“normal” or “excessive” consumer indebtedness for the economy as a whole. A simple

aggregate measure of consumer indebtedness that covers all age groups does not

contain much absolute information about how much consumers as a whole are

indebted.296

Individual data enable the identification of socio-demographic and economic profiles of

individuals or households experiencing situations of financial difficulties and can give

useful insights into the most frequent causes of over-indebtedness. Furthermore,

individual data may serve for the evaluation of measures undertaken to prevent and

manage over-indebtedness. Individual data is collected via surveys, conducted by

national statistic institutes, central banks or research institutes. These surveys can

include questions referring to objective as well as subjective indicators of overindebtedness.

Especially answers to subjective questions can be influenced by

respondents’ socio-cultural context and their perception of economic and financial

fragility and difficulty in a certain moment. Again, unfortunately, individual data is usually

not homogeneous for different countries.

Measures determining over-indebtedness

Specific measures of indebtedness can again be grouped into three categories:

administrative, objective and subjective measures. An overview of these measures is

given in Table 63, they are discussed in more detail in the following.

294 See Vandone (2009), p. 71.

295 See Betti et al. (2007), p. 143.

296 Betti et al. (2007), p. 143.

iff/ZEW – Final Report on interest rate restrictions in the EU 359

Table 63: Overview of measures determining over-indebtedness

Administrative

measures

Objective measures Subjective measures

Court arranged solutions Debt ratios: Inability to make ends

meet.

- Consumption/income

ratio

Cases of approaches to

counselling agencies

- Debt/asset ratio

Perception of heavy

payment burden

Debt write-offs by banks - Debt/disposable income

ratio

Mortgage repossessions - Debt-servicing ratio

Arrears

Financial coping strategies:

- Overdraft use

- Extended credit card

facilities

- New credit to pay off

debts

Administrative measures

Administrative measures comprise the number of court-arranged solutions to debt (legal

measure), cases of debt problems reported to debt advice agencies or debt write-offs by

lending institutions. They are a by-product of official functions and take into account only

those cases where payment difficulties have been registered officially in some way.

Legal measures depend on the legislation of the respective countries, which complicates

international comparison. A further obstacle to using such indicators is that some

European countries (Bulgaria, Greece, Italy and Lithuania) do not even have a law on

personal bankruptcy.297 More fundamentally, another drawback of legal measures is that

they include only the severe cases in which indebtedness has already led to consumer

bankruptcy. The indicator thus grasps only a small fraction of over-indebted households

and disregards cases of severe disruption to the established consumption pattern. Thus,

legal debt settlements can at best paint a picture of the outcome rather than the

situation of indebtedness. Keeping in mind that debt settlement procedures are hardly

comparable between different countries, they serve at best as a partial measure of

indebtedness.

297 For a comparison of the insolvency statutes of countries in the EU see Brennecke, Otépková (2009).

360 iff/ZEW – Final Report on interest rate restrictions in the EU

The drawbacks of legal measures apply even more to data regarding households

approaching debt agencies. Recorded numbers of people seeking advice will depend on

the pervasiveness of counselling agencies, the size of their staff and hence on the

funding available to them. Paradoxically, rising numbers of over-indebtedness may

therefore be caused by an increased commitment of debt agencies instead of a change in

the advice-seeking consumers’ initial situation.298 Thus, the informational content of

figures from debt agencies is low.

Data on debt write-offs by lending institutions is collected only in some countries. Still, if

the information is recorded by credit reporting agencies it is usually not publicly

accessible. In countries where data on write-offs is collected by central banks, the data is

aggregated, which leads to a low informational content for mapping over-indebtedness.

Besides the accessibility of debt write-off data, it is difficult to assess what causes might

induce changes in the write-off rates: write-off rates are linked to the policies of

individual borrowing institutions, which may vary over time. So a rising level of write-off

rates is not necessarily a symptom of growing excessive debt loads of households.299

Indicators depicting mortgage possessions share the same drawbacks as debt write-offs.

There are even two more reasons why information relating to properties taken into

possession may not be an accurate measure of households’ debt burden:300 first, there

are potentially significant time lags involved in the possessions process (up to a year),

such that properties taken into possession today may result from actions related to

repayment problems in the past. Secondly, actual repossession may not be

accomplished, as the involved parties may agree on a different arrangement. Therefore,

only part of the consumers who experience problems paying their mortgages will end up

having their property repossessed.

Altogether, administrative measures tend to indicate problems with a delay, they capture

only a fraction of over-indebted households and they are hardly comparable in crosscountry

analysis due to different national frameworks.

Objective measures

Most of the existing objective indicators of over-indebtedness are based on the notion of

unsustainable consumption behaviour of the households.301 To determine when a

household’s debt becomes unaffordable, commonly applied indicators are concerned with

a household’s spending behaviour, level of debt, or ability to service outstanding debt.

High debt levels do not necessarily lead to over-indebtedness, as long as households

manage to meet their obligations. However, it puts them into a potentially vulnerable

position. Adverse shocks, such as unemployment, price increases or changes in interest

rates can push such vulnerable households into serious financial difficulties. The abovementioned

indicators thus help to form views about the extent to which households

exposed themselves to potential economic shocks because of their credit obligations.

One way to assess unsustainable spending behaviour is by looking at the consumptionto-

income ratio. High values of the ratio (eg. above 100 percent) can be perceived as an

indication of over-indebtedness. However, drawing conclusions from the permanent

298 See EUC (2008), p. 44.

299 See BERR (2007), p. 18.

300 See BERR (2009), p. 24.

301 See Betti et al. (2007), p.142.

iff/ZEW – Final Report on interest rate restrictions in the EU 361

income hypotheses, younger households who have to acquire durable goods and retirees

who deplete their savings are expected to exhibit high consumption to income ratios.302

The level of household debt can be measured by the debt to asset ratio. The ratio

compares the overall stock of debt to the household’s total stock of assets. It can

indicate how well consumers are potentially able to cover debt out of their stock of

assets, while this ability is likely to depend on the liquidity of the asset compared to the

payment date of debts.303 However, the ratio gives no indication of this relative liquidity

of assets owned compared to debts and potential difficulties of households in servicing

the debt in each period using their available resources at the time.

The aggregate household debt to disposable income ratio is not as such a measure of

over-indebtedness. However, comparing debt to income facilitates an assessment of

households’ real exposure to debt, which is not possible by judging from levels of debt

only.304 Moreover, the debt to income ratio can indicate whether income readily covers

current obligations. A clear advantage of the measure is that it can be used for crosscountry

comparisons within the EU, as the data is either collected by the national central

banks or statistical authorities, which use a harmonised methodology.305

However, not all possible financial commitments are covered by the debt-income ratio as

only credit related to the financial sector is reported in official statistics in all countries.

The credit to income ratio cannot directly be used to infer the burden of debt for the

households as only part of the outstanding amount is payable in the near term. Rather,

the monthly percentage of income devoted to financing commitments can be deemed an

adequate indicator of the current payment burden.

The ratio of debt payments to disposable income (after tax) can be regarded as a

measure of the debt-servicing burden to households. Credit-service ratios are not

collected in all Member States and sometimes only raw data series are provided (such as

disposable income and credit service) from which the ratio must be calculated. But where

the data is collected, the definition is consistent among countries in the EU. As usually

only aggregate information is available, the information content regarding the burden to

individual households is low and hence the usefulness of the indicator with regard to

over-indebtedness limited. An even more important drawback of the indicator is that it

typically includes only debt payments (on mortgages and consumer credit) and no other

monthly payment burdens such as utility bills or rent payments.

Deducting indications of over-indebtedness from the measures described above requires

defining their critical levels.306 However, there is no established methodology for

determining the critical level of these ratios beyond which a household can be regarded

to be over-indebted. According to the permanent income hypotheses, thresholds would

vary with households’ characteristics, as has already been discussed in the example of a

debt-to-income ratio, which varies with age. It might therefore be sensible to incorporate

a household’s absolute income level in order to decide whether a high debt-servicing

ratio leaves the household with enough money to cover its basic needs.

302 On the contrary, Betti et al. (2007) argue that over-indebtedness is more likely to be observable by a low

ex-post consumption-to-income ratio than a high one ex-ante, p. 140 f.

303 See BERR (2009), p. 13.

304 See BERR (2007), p. 12.

305 See EUC (2008), p. 45.

306 Critical levels have been defined in major empirical studies on over-indebtedness, such as Oxera (2004),

MORI (2005), ECRI and PfCR(2008). The combination of objective indicators and subjective measures

suggested by MORI (2005) have become the benchmark and a convention for subsequent evaluations of

‘over-indebtedness’ in the UK. A more exhaustive compilation of concepts of over-indebtedness can be

found in Anderloni and Vandone (2008).

362 iff/ZEW – Final Report on interest rate restrictions in the EU

An indicator not depending on a certain threshold refers to the financial arrears of

households. Indicators on arrears capture all forms of debt and household bills for which

a household is behind in payments for longer than a specified period of time. This cut-off

time period is important, otherwise households which simply forgot to pay a bill or debt

once would be considered over-indebted. Data on financial arrears is collected in all

countries. Most data (and often more precise data relating to individuals) is collected by

private credit bureaus and is not in the public domain. In some countries it is primarily

the central bank accumulating the data through a public credit register. It is also

common that both types of institutions gather the information.

Another group of objective indicators deals with strategies households may adopt to cope

with financial stress. Financial coping strategies307 comprise households constantly in

overdraft, the use of extended facilities on credit cards to pay for everyday living

expenses and to draw cash, and the need to take out additional credit in order to pay off

debts.

Subjective measures

Subjective measures also try to grasp the extent of how stressed households are because

of their financial obligations. Unlike for the other measures, in this approach individual

households are considered to be the best judges of their own financial well-being. All

households who claim to be unable to repay their debts without substantially lowering

their standard of living are regarded to be over-indebted. There is no doubt that

subjective measures also need to be considered with some caution: as the data is

survey-based, it is necessary that consumers be honest in the report of their financial

situation. Survey responses may also differ across households due to different

perceptions and reactions, although the financial circumstances may be identical.308

Furthermore, respondents may feel urged to report at least some kind of ‘financial

difficulties’ in an interview concerning over-indebtedness, especially when credit and debt

problems are continually discussed in the media. It has to be presumed that the more

vague the question concerning ‘financial difficulties’, is the greater the likelihood that the

interviewer is capturing self-reflective attitudes and responses rather than the underlying

debt position of the household. Despite these limitations, the important advantage of

subjective measures is that they give a direct indication of over-indebtedness by

capturing not only those households that are in arrears but also those that are currently

managing their credit commitments but are vulnerable to falling behind.

Conclusion

Because of the lack of models from economic theory309, lack of data on household income

and asset, and lack of robustness in the results from an “objective” approach, we decided

to rely in this study on a broadly undisputed objective measure – the fraction of

households in payment arrears as well as a subjective indicator – households being

unable to make ends meet.310

307 See ECRI and PfRC (2008), pp. 6 ff. for studies drawing on financial coping strategies.

308 Compare Disney et al. (2008), pp. 19 and 51 for the mentioned drawbacks.

309 Studies from the UK constantly identify a lack of overlap between the indicators. According to MORI “this

demonstrates the need to consider several measures when conducting analyses in this area” while Disney et

al (2008) conclude that by considering multiple indicators it is rather probable that the sub-set of ‘overindebted’

households will not be identified correctly as almost all families might satisfy one or more criteria.

They suggest considering “a more ‘structural’ and life cycle-based approach to the issue of ‘overindebtedness’,

while acknowledging that specific indicators and self-reported perceptions are important

components of the over-indebtedness phenomenon.”

310 As some recent studies focus more on a subjective approach to determine and analyse over-indebtedness;

eg. Betti et al. (2007).

iff/ZEW – Final Report on interest rate restrictions in the EU 363

Annex IX: Incidence of short-term credit facilities

The following Venn-diagrams adumbrate the France

pervasiveness of different forms of shortterm

credit facilities measured by the

percentages of households indicating to

hold none, one or more of the categories

overdraft facility, credit card or other card.

However, households holding credit cards

and other cards with pure debit functions

cannot be distinguished and are therefore

considered having acces to short-term

credit. Therefore the complement, the

fraction of households without access to any

of these forms of short-term credit (credit

excluded households) shown by the

categorie “none” underestimates the true

prevalence of households without access to

short term credit.

Germany

The Netherlands

Sweden

UK

Source: Eurobarometer 60.2, Variable Q11 (“D o you personally have…?”)

364 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex X: Stakeholder Questionnaire

Your details as Respondent

0.1 Your country Country A-K Country L-Z

0.2 Your activity Please select from the dropdown menu

Other:

0.3 Your name:

Name of your institution:

Address, telephone and email:

PART A: Interest Rate Restrictions (IRR)

1.1 Has the role of interest rate

restrictions been discussed among

regulators, providers or consumer

organisations in your country in the

past five years?

Select

Please give details:

1.2 Are you aware of studies that

have been undertaken regarding

interest rate restrictions? Are there

any plans in your country to do

so?

Yes No Don’t

know

Please give details:

1.3 Do any of the following terms

appear in a legal context and how are

they defined: ”mainstream credit”,

“fringe lending”, “money-lenders”,

“last resort loan”, “prime”, “subprime”,

“payday loans”, “usurious

rates” etc..? [please give translated

word in your own language]

1.4 What does “usury” (ie. the

respective word in your language)

refer to (generally and, where

applicable, in your legislation)?

iff/ZEW – Final Report on interest rate restrictions in the EU 365

1.5 Which forms of interest rate

restrictions currently exist?

Absolute or relative contractual interest

rate ceilings (fixed administratively by

statute)

Absolute or relative contractual interest

rate ceilings (fixed by court rulings)

Capped default interest rates

Laws designed to prevent exploitation

and unfair competition with effects on credit

cost

Restrictions on the compounding of

interest

Restrictions on the variability of variable

interest rates

Other forms of restrictions to the level or

rate of interest including moral consensus

Anti-Trust regulation or laws designed to

improve levels of competition

Regulations concerning early repayment

fees

Please give detail:

1.6 To what extent are interest rates

regulated?

Scale: 1 (not at all regulated) to 5

(very significantly regulated)

Select

Please give detail:

If IRR in the form of ceilings exist in your country, please comment on the following:

1.7 What were the reasons for

introducing controls on the cost of

credit?

1.8 To what extent are those reasons still

valid?

1.9 What were the reasons for choosing

the particular form of the restriction

and the level at which the ceiling was

set?

If IRR in the form of ceilings do NOT exist in your country, please comment on the

following:

1.10 What policy concerns have

contributed to the decision not to

introduce ceilings and what

alternative controls on the cost of

credit exist?

366 iff/ZEW – Final Report on interest rate restrictions in the EU

1.11 Are there any plans to review the

position on the introduction of

ceilings?

Yes No Don’t

know

Please mention any circumstances or

criteria identified, which if satisfied, could

lead to a change in policy:

If IRR do NOT exist in your country, please skip the next 6 questions and [go to

Question 1.18]

1.12 Have consumers benefited from

the introduction/retention of interest

rate ceilings?

Yes No Don’t

know

Please give details and mention whether

consumers are sufficiently aware of the

existence of the IRR:

1.13 How easily have credit providers

circumvented the interest rate

restriction rules mentioned?

Scale: 1 (not at all) to 5 (very easily)

Select

Please give detail (eg. on compliance in

case of IRR in the form of a ceiling):

1.14 To what extent are the interest rate

restrictions effective?

Scale: 1 (not at all) to 5 (very

significantly effective)

Select

Please give detail:

1.15 What are the compliance costs of the

different forms of IRR? Has a

regulatory impact assessment been

conducted?

1.16 To what extent is compliance with

the IRR monitored by the regulator?

Scale: 1 (not at all) to 5 (very closely

monitored)

Select

Please give detail (eg. on steps that

regulatory agencies take to monitor

compliance and/or to enforce the IRR):

1.17 If IRR in the form of ceilings, has

experience with the ceiling

demonstrated any loopholes or

other practical problems with the

legislation? What steps have been

taken to address these issues?

1.18 What changes to the current IRR

regulatory apparatus would you

suggest to improve the

effectiveness of existing IRR?

iff/ZEW – Final Report on interest rate restrictions in the EU 367

1.19 What have you learnt from

experience in other EU Member

States and other countries that have

introduced IRR (caps or other

controls)?

Imagine a country with no IRR at all. If interest rate restrictions were introduced

and effective and set at…

….a level that was double the average of the interest rates offered on the entire

credit market (ie. set at 200% above the average calculated price for all credit

products in a previous period)…

for average

consumers?

for the lowincome

consumers?

….access to credit would Select Select

….cost of credit would Select

Select

….over-indebtedness would Select

Select

….variety of products would Select

Select

1.20

…. the number of individuals

resorting to the illegal credit market

would

Select

Select

….a level that was fixed at 30% APR for all credits…

for average

consumers?

for the lowincome

consumers?

….access to credit would Select Select

….cost of credit would Select

Select

….over-indebtedness would Select

Select

….variety of products would Select

Select

…. the number of individuals

resorting to the illegal credit market

would

Select

Select

1.21

Please elaborate on each of your answers here:

Which credit forms/types/products would especially be affected by interest rate

restrictions in that hypothetical country?

for average

consumers?

for the lowincome

consumers?

Mortgages (for property purchase)

1.22

2nd charge mortgages (for

consumption)

368 iff/ZEW – Final Report on interest rate restrictions in the EU

Auto/vehicle loans

Other personal loans

Other point-of-sale/mail-order loans

Overdraft facilities

Credit card credit

Pawnbroker loans

Home (collected) loans

SMS loans contracted at a distance

Payday loans

Other:

Other:

Please elaborate on each of your answers here:

Which of the following regulatory activities would have the most pronounced

effects on the four desired outcomes shown below? [Scale: 0 (the opposite effect)

to 5 (very strong effect]

Reduced

level of

overindebtedn

ess

Reduced

cost of

credit

Improve

d credit

access

Wider

variety of

credit

products

Interest rate

restrictions

Select Select Select Select

Tighter responsible

lending requirements

(in general)

Select Select Select Select

Regulations to limit

rolling

over/consolidating

existing credit

Select Select Select Select

Strengthening of

disclosure obligations

Select Select Select Select

1.23

Regulation on personal

bankruptcy (in general)

Select Select Select Select

iff/ZEW – Final Report on interest rate restrictions in the EU 369

More specifically: to

make insolvency easier

Select Select Select Select

More specifically: to

enable earlier

discharge

Select Select Select Select

Please elaborate on each of your answers here:

How would you describe the adequacy of the actual level of regulation of IRR

faced by the following different credit forms/types/products in your country?

Mortgages (for property purchase) Select

2nd charge mortgages (for

consumption)

Select

Auto/vehicle loans Select

Other personal loans Select

Other point-of-sale/mail-order loans Select

Overdraft facilities Select

Credit card credit Select

Pawnbroker loans Select

Home (collected) loans Select

SMS loans contracted at a distance Select

Payday loans Select

Others: Select

Others: Select

1.24

Please elaborate and identify specific products here:

How would you describe the adequacy of the actual level of regulation of IRR

faced by the following types of credit institutions?

Banks Select

1.25

Non-bank lenders Select

370 iff/ZEW – Final Report on interest rate restrictions in the EU

Finance companies/mortgage

specialists

Select

Moneylenders and payday lenders Select

Home credit (collected at home) Select

Others: Select

Please elaborate:

With regard to IRR in your country, are you in favour of IRR? How important is this

regulation?

Interest rates should be capped. Yes, the regulation is Select.

No, the deregulation is Select.

Don’t know.

Please give reasons why/why not:

Variability of interest rates in variable

rate credit contracts should be

regulated.

Yes, the regulation is Select.

No, the deregulation is Select.

Don’t know.

Please give reasons why/why not:

Default interest should be capped. Yes, the regulation is Select.

No, the deregulation is Select.

Don’t know.

Please give reasons why/why not:

1.26

Level of fees and charges should be

regulated.

Yes, the regulation is Select.

No, the deregulation is Select.

Don’t know.

Please give reasons why/why not:

1.27 If controls are to be introduced or

retained, what do you think would

make for an effective regime? Eg.

what form should such controls take

and at what level should ceilings be

set?

iff/ZEW – Final Report on interest rate restrictions in the EU 371

1.28 Should caps be fixed or floating? Fixed ceiling Floating ceiling

Neither

Please explain:

1.29 Should caps differ according to

type of loan?

Different ceiling levels One unique

level

Please explain:

1.30 To what extent should fees and

charges be included in the cap?

Borrowing rate APR Other

Please explain:

1.31 How have regulations of the unfair

terms in consumer contract (as per

Directive 2005/29/EC on Unfair

Commercial Practices) been used to

control high cost credit and default

charges?

[Please mention if there have been any

problems]

1.32 Where controls exist in court based

rules concerning unfairness more

generally, are the rules sufficiently

clear and how have courts

implemented these? Is there

evidence that courts have been

willing to intervene to limit the cost

of credit?

[Please provide citations for key cases

where possible]

1.33 Are there any regulatory actions

you can think of which would

increase levels of access to credit for

low income consumers?

1.34 What are the regulatory measures

currently being planned with

regards to consumer credit

regulation?

1.35 Are there any regulatory actions

you can think of which would limit

future levels of over-indebtedness?

1.36 Micro credit: Are there special

regulations with regard to microlending

which set interest rate

restrictions or gives relief from it?

(please give details)

PART B: Over- Indebtedness

2.1 Do you think that private overindebtedness

is a problem in your

country?

[See definition on page 2]

Yes, it is a Select problem.

No Don’t know

Please explain:

372 iff/ZEW – Final Report on interest rate restrictions in the EU

2.2 Do you think that this problem has

improved or worsened over the

last five years?

Select

Please explain:

2.3 Are there any statistics regarding

over-indebtedness of private

households?

Yes No Don’t

know

If yes, please provide a link to such data:

2.4 In your view, what are the main

driving forces behind overindebtedness?

Which of the following changes have occurred within the last three years? How have

they influenced over-indebtedness?

Aspects Evolution Influence on

overindebtedness

Situation of the labour market Select Select

Level of financial literacy/competence Select

Select

Standards of “responsible lending” Select

Select

Access to credit for vulnerable

consumers

Select

Select

Prices of real estate property Select

Select

Entry of new providers in the credit

market

Select

Select

New insolvency legislation Select

Select

Consumer propensity of demand Select

Select

Level of restrictiveness of IRR rules Select

Select

2.5

Please elaborate on each of your answers here:

PART C: Access

How would you describe the following aspects of the level of credit access

Access to… …for average

consumers?

…for low-income

consumers?

…. banking services in general Select Select

…. mortgage credit (for property

purchase) Select Select

…. mortgage credit (for consumption)

Select Select

…. non-mortgage consumer credit

Select Select

…. credit card credit

Select Select

3.1

…. overdraft credit

Select Select

iff/ZEW – Final Report on interest rate restrictions in the EU 373

…. credit provided by non-banks

(retail, leasing, moneylenders,

pawnbrokers etc.)

Select Select

Please elaborate on each of your answers here:

3.2 In your view, should low-income

borrowers be given any greater

access to credit than they are given

at present?

Yes No Don’t know

Please give reasons why/why not:

3.3 If yes, to which forms of credit? Mortgages (for property purchase)

2nd charge mortgages (for consumption)

Auto/vehicle loans Other personal

loans

Other point-of-sale/mail-order loans

Overdraft facilities Credit card credit

Pawnbroker loans Home collected

loans

SMS loans Payday loans

Others:

3.4 If IRR in the form of a ceiling exists

in your country, is there evidence

that low-income consumers have

been excluded from the credit

market following the introduction

of this ceiling? (If so, please tell us

whether or not this is a deliberate

policy intention or an unintended

effect)

When answering, please give details on the

availability of credit options and extent to

which consumers have turned to illegal

lenders:

3.5 What kind of political and legal

discussions are presently ongoing

to make credit more affordable?

3.6 Is affordability a concept in your

legislative or regulatory frameworks?

Yes No Don’t

know

Please give details:

PART D: Credit Markets

374 iff/ZEW – Final Report on interest rate restrictions in the EU

Please indicate which of the following forms of consumer credit exist and to what

extent they are used:

non-existent uncommon widespread

1 2 3 4 5

Mortgages (for property purchase)

2nd charge mortgages (consumption)

Auto/vehicle loans

Other personal loans

Other point-of-sale/mail-order loans

Overdraft facilities

Credit Card Credit

Pawnbroker loans

Home (collected) loans

SMS loans contracted at a distance

Payday Loans

Others:

Please specify:

4.1

Please indicate data sources and elaborate, where possible, on the purposes for which

the different types and forms of consumer credit are mainly used:

Please indicate the value for the typical interest rates and loan sizes of the

following types of consumer credit: [We ask regulators to kindly provide all

relevant data and sources available, and provider associations to answer for

the types of credit that apply to the activities of their association]

Typical

interest

rate

(APR)

Range

of APRs

(%,

minmax)

Typical

fees (in

%

annualis

ed)

Typical

loan

size

(EUR)

Typical

min.

loan

size

(EUR)

Market

size

(outstandi

ng

volume,

EUR)

Mortgages (for

property

purchase, for

LTV loan of

<75%)

2nd charge

mortgages (for

consumption)

4.2

Auto/vehicle

loans

iff/ZEW – Final Report on interest rate restrictions in the EU 375

Other personal

loans

Other point-ofsale/

mail-order

loans

Overdraft

facilities

Credit card credit

Pawnbroker loans

Home (collected)

loans

SMS loans

Payday loans

Others:

Others:

Please elaborate on each of your answers here:

4.3 How would you assess the different credit forms/types/products with regard to

the attributes/market features below: [From a scale of 1 (insignificant) to 5

(very significant)]

Incidence of high

cost credit

1 2 3 4 5

Incidence of

consumer debt

problems

1 2 3 4 5

Transparency of

pricing

1 2 3 4 5

Mortgages (for

property

purchase)

2nd charge

mortgages

(consumption)

Auto/vehicle

loans

Other personal

loans

Other point-ofsale/

mail-order

loans

376 iff/ZEW – Final Report on interest rate restrictions in the EU

Overdraft

facilities

Credit card credit

Pawnbroker

loans

Home (collected)

loans

SMS loans

contracted at a

distance

Payday loans

Others:

Others:

Please elaborate and identify specific products here:

4.4 Is there a need for controls on the

cost of credit to protect low income

and vulnerable consumers?

Yes No Don’t

know

Please give reasons why/why not:

4.5 Which type of credit cards are

commonly held by consumers?

[See definitions on page 2]

Mostly single function cards (eg. debit

cards)

Mostly double function cards (ie. used as

a payment device and as a source of credit)

Both types of credit cards are equally

common (single and double function credit

cards)

Credit cards are not popular at all

Don’t know

Comments:

4.6 Do you see a trend towards using

credit cards as a regular source of

credit (ie. increased use made of the

credit facility of the double function

card)?

Select

Please explain:

4.7 How would you describe the presence

of an illegal market in lending

money to low-income households?

Select

Comments:

iff/ZEW – Final Report on interest rate restrictions in the EU 377

4.8 What is the importance of the

different credit markets? (in

EUR/local currency and % of total

lending activities if possible)

[See definitions on page 2; From a

scale of 0 (non-existent) to 5 (very

significant)]

Mainstream lending (typical bank loans

and non-bank credit): Select

Alternative lending (higher cost small

credit): Select

Community/informal lending (family,

social banks, welfare): Select

Illegal lending: Select

Please give detail and specify source or

basis for any figures or estimates:

4.9 In general would you say that your

country’s credit markets are price

competitive?

Scale: 1 (not competitive) to 5 (very

highly competitive):

Select

Please give details (see also next two

questions):

4.10 Have you observed problems in

specific sub-markets?

Yes, No. Don’t

know.

Please give details:

4.11 Have you observed examples of

sales practices which appear

designed to make it more difficult for

price competition to operate?

Yes, No. Don’t

know.

Please give details:

Did you notice a trend in consumer credit business with respect to any of the

following areas in the period 2002 to 2007 and since 2008?

… before the

financial crisis

(2002–2007)

… after the

financial crisis

(from 2008)

Level of interest rates Select Select

Use of fees outside the scope of the APR

calculation Select Select

Use of default charges

Select Select

Cross selling of ancillary products (eg.

insurance) Select Select

Productivity within firms

Select Select

Costs of operation, including capital

Select Select

Volume of consumer credit (incl. mortgages,

and from all sources) Select Select

4.12

Refusal rates for consumer credit

applications Select Select

378 iff/ZEW – Final Report on interest rate restrictions in the EU

Credit defaults by private households

Select Select

Please elaborate and explain what you think were the main causes of the trends

mentioned:

How would you describe the level of the following:

…for

average

consumers?

…for lowincome

consumers?

Understanding of consumer credit products Select Select

Level of financial literacy

Select Select

Credit options available to choose from

Select Select

4.13

Please elaborate here:

In your view, how important are the factors below as barriers to the cross-border

provision of consumer credit in the EU? [From a scale from 1 (not at all) to 5 (very

significant)]

Different language Select

Different currency

Select

Different approach to consumer protection

Select

Lack of information about the customer

Select

Consumer preference for local creditors

Select

Different legal frameworks for debt recovery

Select

Interest rate restrictions (incl. restrictions on

fees) Select

Extra costs of doing business abroad

unrelated to those above Select

Other factor: Select

4.14

Please elaborate here:

iff/ZEW – Final Report on interest rate restrictions in the EU 379

Annex XI: Tables of stakeholder responses

The tables with stakeholder answers to some of the survey questions are provided as

illustrative examples of stakeholder responses. Not all answers have been reproduced.

Selection 1: “Has the role of IRR been discussed among regulators, providers or

consumer organisations in your country in the past five years?” and “Are you

aware of studies that have been undertaken regarding IRR?” (SQ 1.1 and 1.2)

MS Level of

discussion

Details Studies on IRR exist

AT Somewhat

discussed

Currently concerns regulation of

default interest rates only.

Yes but restricted to default interest.

BE Somewhat

discussed

By parliament (Conseil de la

Consommation) and at several

meetings with representatives of

the financial institutions, the

government and the consumer

organisations.

Révision du mode de fixation des taux annuels

effectifs globaux en matière de crédit à la

consommation (2006) at

http://statbel.fgov.be/fr/binaries/354_tcm326-

41864.pdf.

BG Somewhat

discussed

After the financial crisis the banks

raised the rates and the

consumers were not happy as

they had to pay more, and some

media reports on usury within the

illegal sector.

Yes

CY Somewhat

discussed

A usury bill (not involving ceilings)

has been drafted and IRR was

discussed ahead of the CCD

transposition set for October

2010.

Yes (a central bank study compared ceilings in

different EU Member States incl. Malta)

CZ Intensively

discussed

A draft aimed at introducing IRR

was raised by a group of MPs in

the lower Chamber of the Czech

Parliament ahead of the elections

but proposals were not put to

vote. The draft was not supported

by the Government. Drafting of

CCD implementation brought the

matter into discussions.

Ministry of the Interior of the Czech Republic -

Prevention of indebtedness study. Some

criticise that the draft laws were proposed

without any data based studies. A report was

also published by the Liberalni Institut

(Schwartz, 2007) see http://www.responsiblecredit.

net/index.php?id=1980&viewid=45753.

DK Intensively

discussed

Discussed in connection with

implementation of the CCD 2008.

A proposal from the Danish

opposition parties of a ceiling of

aprox. 17% APRC was rejected.

The Ministry of Economy and Commerce are

currently undertaking a study of the market

for consumer loans as a part of the discussion

of IRR. Such a cap has been proposed by

different political parties in parliament, but so

far without a majority of MEP's behind it. As a

result the government established a ministerial

working group in the beginning of 2009 that

were to analyze the effect of APR-ceilings in

other countries and publish a report by the

end of 2009. The report has not yet been

published and is expected in the Summer of

2010, but was influential in implementation of

the CCD 2008 as many problems were

postponed to the publication of the report.

380 iff/ZEW – Final Report on interest rate restrictions in the EU

EE Somewhat

discussed

The discussions have mostly

focused on problems with micro

credit (especially SMS-loans) with

an extremely high APRC.

No study, but a legal article "Protection of

Consumer Rights in SMS Loan Agreements" (in

English) was published in the European Review

of Private Law 2010 (18) by K.Saare, K.Sein,

M.A. Simovart. See http://www.responsiblecredit.

net/index.php?id=1980&viewid=45754

FI Somewhat

discussed

The introduction of IRR has been

discussed due to problems

relating to increase of SMS loans

and the issue has been raised by

the consumer ombudsman.

However, at least for the time

being other regulatory options to

reduce problems have been

chosen.

No. Though reports in the 1980ies on the

subject were made.

FR Intensively

discussed

During the 2 years of debates on

the bill concerning the credit

reform, the role of IRR has been

strongly discussed. Principalement

à deux occasions : 1) Les

difficultés éprouvées par les

accédants ayant souscrit des prêts

au logement à taux variables dont

iols n'avaient pas compris les

clauses de révision 2) Pour une

réforme des taux d'usure en crédit

à la consommation. Although the

modalities of fixing the ceiling and

the credit categories were

discussed (with changes made by

parliament in 2010), the principle

of IRR was never questioned. This

subjet has been discussed in 2010

during household overdebtedness

bill and a government report

about bank charges which is going

to be published may also mention

this subject.

Rapport sur les modalités de fixation du taux

de l'usure", General Inspectorate of Finance &

General Inspectorate of Social Affairs,

Februray 2009. See http://www.responsiblecredit.

net/index.php?id=1980&viewid=45728

DE Somewhat

discussed

In the context of statistical

changes to reported data due the

change in competence from the

Bundesbank to the European

Central Bank

No, at least not specifically. Only the statistics

problems discussed on the new ceiling of usury

after court chosen indicator ceased to be

published by Bundesbank. See the report by

the Bundesbank concerning the change in

statistics and the Report on its implications by

Hartmann-Wendels/Spörk in 2006.

See:http://www.responsiblecredit.

net/index.php?id=1980&viewid=45755

GR Not at all

discussed

n.a. No

HU Somewhat

discussed

Interested in experience of EU

peers.

Yes

IE Not at all

discussed

n.a. No

IT Somewhat

discussed

Debate has been going on in

connection with the reform of

Italian Usury Law

No

LV Not at all

discussed

n.a. No

iff/ZEW – Final Report on interest rate restrictions in the EU 381

LT Somewhat

discussed

The question of IRR is usually

raised during various discussions.

No

LU Not at all

discussed

n.a. No

MT Not at all

discussed

n.a. No

NL Somewhat

discussed

The maximum interest rate has

been lowered in 2006 and 2009.

The Netherlands has IRR on

consumer credit only (not on

mortgage credit). The level has

been subject of discussion. Loans

with a duration shorter than 3

months had initially been exempt,

but will be covered by the

restrictions later in 2010.

A study on SMS-credit and pawn-brokers was

done in 2009. See Report from "Research voor

beleid'. http://www.responsiblecredit.

net/index.php?id=1980&viewid=45753.

PL Somewhat

discussed

IRR were introduced in Poland in

2006 following an intensive

debate. Before the maximum

percentage, tests were conducted

mainly by representatives of the

banks. Since the adoption of the

regulation the issue has not been

widely discussed, although the

government had plans to amend

the present regulations in 2010.

KPF commissioned a study evaluating the

impact IRR have had on the market. The

report was published in March 2009 (English

translation available). A barometer Rynku

Consumer Finance (BAROMETR KPF-IRG) has

also been published (see

http://kpf.pl/raporty/barometr-kpfirg/)

PT Intensively

discussed

A legal document has been

adopted (Decreto-Lei nº

133/2009, de 2 de Junho) which

establishes a ceiling for interest

rates in consumer credit (personal

loans; auto loans; credit cards;

etc.). There has been some legal

uncertainty regarding the way the

ceilings were introduced, some

debate is expected.

No

RO Not at all

discussed

n.a. No

SK Intensively

discussed

Public discussion is high due to

recent introduction and debate

following this.

Ministry of Finance internal analyses on credit

market in 2008. A report was also published

by the Hayek Foundation see

http://www.responsiblecredit.

net/index.php?id=1980&viewid=45753

SI Somewhat

discussed

Draft Act amending the Consumer

Credit Act (issued in the Slovenian

Official Gazette No. 77/2004 and

111/2007). IRR for non-banking

credit providers have been

introduced in Slovenia before. The

discussions in the last 5 years

have focused around how to make

the IRR more effective and

prevent evasion.

No

ES Somewhat

discussed

Mostly in the context of floors to

variable mortgage credit, but also

due to planned changes in

existing IRR on overdrafts.

No

382 iff/ZEW – Final Report on interest rate restrictions in the EU

SE Somewhat

discussed

Interest rates on SMS-loans have

led to some discussions.

No

UK Intensively

discussed

Previously considered by the

government and ruled out in

2004. The discussions are

focussed on the consumer credit

industry and there hasn't been an

active discussion in the context of

mortgages. There has been

consistent lobbying for interest

rate controls over many years and

opponents of IRR and the

Government have dismissed the

views that consumer benefits

could outweigh consumer

detriment.

Most have been Government sponsored

studies. The Policis study "Interest Rate

Controls in other Countries" for the

Department of Trade and Industry in 2004 and

the study on the effects of interest rates have

been disseminated world-wide. More recently,

the issue was referenced within the Report by

the Financial Inclusion Taskforce on Access to

affordable credit (2010) and the OFT study on

High Cost Consumer Credit (2010). A further

study was conducted in 2010 by the New

Economics Foundation (Doorstep Robbery)

which provided counterbalancing findings

compared to the 2004 report mandated by the

Government. We are not aware of studies in

regard to mortgages, or of any plans to

undertake such studies.

Note: Discussions at the EU level on these issues were reported by several respondents with regard to the EU

level and the Study on mortgage policy options which although focused more on mortgage markets in general,

nevertheless concerned in parts the issue of IRR and was discussed it some detail. Regarding the fourth column

on IRR studies, though not all empirically based, studies on interest rate restrictions have been made in 11

Member States and these have been taken into account in this report for the European Commission. Of these

11 Member States, the Czech Republic and the UK appear to be the only countries without an interest rate

ceiling that have done some research into IRR. Italy is the Member State that particularly stands out by the fact

that it is a major country with ceilings in place which does not appear to have produced a report on the issue

(made publicly available in any case).

Selection 2: “To what extent are interest rates regulated?” (SQ 1.6)

MS Extent of IRR

regulation

Comments and distinguishing features

Belgium 4 - significantly Ceilings but also limits to the authorised changes in interest

rates sold at variable rates

Cyprus 2 - very little Only the compounding of interest rate is regulated. The Cyprus

legal environment is characterized by liberalization of interest

rates and usury is not qualified as a criminal offence. A usury bill

has been drafted to change this leading up to the transposition

of the CCD which is expected to be transposed into national Law

in October 2010.

Czech

Republic

2 - very little Judgements by court rulings based on breach of good morals

cover all types of interest rates. Though the Supreme Court ruled

that the interest rate 4 times higher than usual interest rate

offered by banks is in violation of good morals, court rulings are

not generally binding and as such regulation through the

consideration by courts is not seen to be exercising significant

control. Interest rates are regulated in some specific areas, such

as building savings where a relative ceiling exists that limits

loans from exceeding the maximum rate on deposits by 3

percentage points. The only way of direct regulation is related to

the late payments or default interest rates.

iff/ZEW – Final Report on interest rate restrictions in the EU 383

Denmark 2 - very little Current provisions have no effect. Loans with APR's of 40-50-

60% and SMS-loans with APR's of several thousand percent

exist. Regarding the interest all credit agreements are covered

by the Danish criminal code (the usury restrictions). Apart from

short-term credits (less than 3 months) and certain credit buys

of securities all consumer credit agreements have to comply with

the consumer credit agreement act. Accordingly the lenders have

to give certain information on the credit. (For credits lower than

EUR 200 (DKK 1,500), there is no legal obligation for the credit

provider to give such credit information). Furthermore any

agreement (credit or otherwise) has to comply with the

regulation in the Danish agreement act, which contains

provisions regarding unfairness etc. If the interest rate is

considered flagrantly unfair (by the courts or by the complaint

boards) it can be overridden. Distance selling and agreements

entered outside the usual place of business of the supplier are

covered by the Danish act on certain consumer agreements.

Estonia 3 - partly According to Estonian Civil Law Act § 86(4) the consumer credit

contract is void when cost of the credit is more than three times

higher than average credit cost, published periodically by central

bank. Restrictions mainly apply to consumer credit. (But we

would like to point out that the scope of consumer credit

regulation is broader than the scope provided in art 2 of

Directive 2008/48 and art 2 of Directive 87/102. For example

consumer credit regulation applies also to mortgage credit and

micro credit.)

Finland 3 - partly Default interest rates are regulated in detail, contractual interest

rates are not.

France 4 - significantly The legal usury rate is not an absolute ceiling but varies with the

average APR measured from the market. Therefore, the level of

interest rates is only regulated in so far as a money-lender would

like to exceed the market average by more than 33%. Also, it

reduces the speed of an increase in interest rates, but does not

entirely prevent it. Also significant are the penal sanctions

involved and the long experience with administration of ceilings

by the central bank. The rest of the regulations that affect

interest rates indirectly are mainly designed to protect

consumers from insufficient information and understanding of

credit products. All in all, the strongest restriction to the level of

interest rates is the competition between banks, whose strategy

is to use mortgage loans as a loss leader. The legal usury rate is

probably more of a constraint for consumption credit.

Germany 4 - significantly A number of strict regulations apply as well as further rules on

missing or inadequate statement of credit costs.

Greece 3 - partly Pricing of the banking services and products are not regulated.

Ireland 2 - very little Although there are some ceilings in place (Credit Unions and

money lenders) these are not extensive.

Italy 4 - significantly The ceiling applies to the APR and is comprehensive of all

commissions and expenses (excluding taxes). The ceiling is

based on provider offered rates on the market and not regulated

through exogenous sources. Some respondents argue that Italy

has only partial regulation because the regulator only sets the

threshold beyond which rates are considered usurious and does

not control them as such.

384 iff/ZEW – Final Report on interest rate restrictions in the EU

Latvia 2 - very little Interest according to Art.1753 of the Civil Law is defined as

compensation to be given for granting use of, or for lateness

relating to a sum of money or other fungible property,

proportionate to the amount and the duration of use thereof.

Therefore it is assumed that interest should be proportionate to

the amount of money borrowed and time of contract. In addition

to this, Cabinet regulations No.692 Regulations Regarding

Consumer Credit Agreement state, that a consumer has a duty

to pay only the payments indicated in the credit agreement as

well as consumer has a duty to pay interest and other charges

only for the time period up to which the consumer has settled his

or her credit obligations. This limits percentage rates to the

amounts that are stated in the credit agreement.

Malta 2 - very little Effectively the ceilings that exist do not apply and the regulation

is thus very weak.

Netherlands 4 - significantly The maximum interest rate concerns all consumer credit but not

mortgage credit. However the exogenous method of fixation of

the level means that regulation can be constraining for a

sustained period of time.

Poland 4 - significantly The ceiling based on an exogenous source affects all consumer

credits and is accompanied by regulation of other costs of

consumer credit (excluding the interest rate, collateral and

premium for insurance coverage) which cannot be higher than

5% of the credit amount.

Portugal 4 - significantly The law sets ceilings for specific credit contracts differentiating

by type (normal, car or credit card). Definitions based on the

purpose of the credit could regulate the market even more

significantly.

Slovakia 3 - partly Ceilings affect consumer credit only.

Slovenia 4 - significantly Verbraucherkreditgesetz: Der effektive Jahreszinssatz in den

zwischen den Verbrauchern und Nichtbanken abgeschlossenen

Kreditvertraegen darf nicht 200% des zuletzt seitens der

Slowenischen Zentralbank veroeffentlichten durchschnittlichen

effektiven Jahreszinssatzes fuer Verbraucherkredite

ueberschreiten. Der durchschnittliche effektive Jahreszinssatz

fuer Verbraucherkredite wird seitens der Slowenischen

Zentralbank ermittelt und zweimal jaehrlich veroeffentlicht.

Spain 2 - very little Only overdrafts face ceilings limited to 2.5 times the legal

interest rate, this may change as a result of transpoisition of the

CCD. In addition, there are some court ruling applying the antiusury

law and others which apply this 2.5 times limit to other

forms of credit by analogy, therefore IRR is somewhat regulated.

UK 2 - very little Although credit unions may face a ceiling and store cards may be

subject to a ceiling of some form in the future, it has not been

deemed necessary to regulate interest rates in the UK. Mortgage

regulation especially does not directly address the level of the

interest rate charged. The CCA does define the total cost of the

credit to the consumer which includes all mandatory fees. It also

governs the calculations of APRs as does the CCD 2008. The UK

consumer credit market is highly competitive which in turn

generates intense competition on interest rates charged.

Consumer behaviour demonstrates that customers are willing to

shop around on a regular basis to benefit from lower interest

rates charged by other lenders. Although rates per se are not

specifically regulated there are provisions to deal with unfair

relationships, unfair terms and competition failures.

Note: Scale for answers ranged from 1 (not at all regulated) to 5 (very significantly regulated).

iff/ZEW – Final Report on interest rate restrictions in the EU 385

Selection 3: Some answers to the question: “What policy concerns have

contributed to the decision not to introduce ceilings and what alternative

controls on the cost of credit exist?” (SQ 1.10)

Member

State

Stakeholder Reasons for not introducing ceilings

MS with no plans to review IRR position

Bulgaria Public

Authority

The main policy concern was the potential distortion of the free

market principles in pricing credit products by banks. The risk

premium over the market interest rates, charged by banks, is

calculated on the basis of the current economic situation and

reflects the credit quality of the customers. If there is IR ceiling,

quite significant amount of bad" customers would be charged with

administrative rate, which is lower than the rate they will pay if

no IR ceiling existed. This would eventually lead to unfair

treatment of good customers, which repay their debts but still are

charged with interest rate closer to the administrative one.

Another major policy concern was, that the practise evidences

that banks will transfer the final effect of the decreased rates to

customers in form of additional hidden taxes and penalty fares.

Cyprus Public

Authority

The Law Regulating the Liberalisation of the Interest Rate and

Related Matters was issued in 1999 in order to abolish the

existing ceiling on lending interest rate with the aim to promote

free competition.

Finland Public

Authority

Reasons not to introduce ceilings relate mainly to fear that

vulnerable consumers would fall outside the normal credit market

and a black market would emerge.

Germany Consumer

Organisation

A main objection is the general attitude that price are to be

influenced by the market not the law. Rules and court rules on

fees can always refer to inadequate and one-sided influence on

pricing, which market rules do not do or do not adequately apply.

Hungary Public

Authority

Professional viewpoint was not to introduce IRR because the

interest rate would have gathered around the maximum amount

(as is observed with capping of early repayment fee).

Latvia Public

Authority

Latvia has so far stuck to approach of market self-regulated

interest rates accompanied with requirements to give clear

information on rates to consumer in advertising and contracts.

Until present no need to control interest rates by stating ceilings

has been discovered. Cost of credit is not presently controlled,

but information on costs has to be clearly provided to consumer.

Lithuania Other The loan market is comparably new in Lithuania so it is poorly

regulated. There was no common consensus to impose IRR.

Luxembourg Public

Authority

Loans are granted to consumers almost exclusively by a limited

number of well-known banks. The problem of charging excessive

interest rates has not materialised in these cases. Therefore the

need was not felt to introduce IRR.

Malta Public

Authority

The decision to lift IRR on credit and financial institutions was

implemented by way of a provision in the central bank legislation

in 2002 and while this was subsequently removed, it has been

inserted by way of secondary legislation in 2009 through the

Interest Rate (Exemption) Regulations L.N. 142 of 2009.

Romania Provider

Association

Ceilings have not been introduced in commercial agreements

because they could restrict access to credit and financial

inclusion. There are also concerns that such introduction will

damage the broader economy by limiting consumer spending, as

has happened in Japan.

Spain Public

Authority

Credit institutions should compete freely under transparent

market conditions.

386 iff/ZEW – Final Report on interest rate restrictions in the EU

MS with plans to review IRR position

Czech

Republic

Provider

Associations

Establishment of ceilings will not lead to usury extermination.

Legal bans do not prevent subjects/entities willing to behave in

such a way from usury and any other kind of immoral behaviour

in general. Effective supervision over the entire credit market and

appropriate enforcement of existing regulation on the national

level would do the trick. Restrictions of access to credit, financial

inclusion. Concerns that it will damage broader economy by

limiting consumer spending, as has happened in Japan. Drive up

illegal lending, loan sharks, personal security issues, credit into

non taxable environments.

Czech

Republic

Consumer

Organisation

There is lack of interest of the lawmakers. Recently a bill aimed

at setting the ceilings was proposed, but the Parliament did not

manage to discuss it before the elections.

Czech

Republic

Public

Authority

Mainly contractual freedom - interest rate represents the risk a

credit provider is taking when granting a credit. A well informed

debtor could be thus prevented from being granted a credit, even

when the interest rate would be appropriate to the risk. Ceilings

also give way to black economy. Credit providers usually find

other means to maintain the level of profit which is not less

burdening to debtor than a high interest rate. We have evidence

that the regulation in Slovakia is evaded.

Czech

Republic

Other Czech society used to be relatively low-stratified. There was not

strong need of introduction of instruments of this kind. The

current attempts to introduce the ceilings are related to

significant increase of over-indebtedness and debt traps.

Denmark Provider

Association

The consumer is all ready protected to a very large scale eg. precontractual

information requirements, right of withdrawal, the

concept of responsible lending where the creditor should asses

the consumers ability to pay down the loan. Interest rate ceilings

will not ensure that consumers are not becoming over-indebted

but can instead lead to consumers taken bigger/longer running

loans than they need so that APRC is legal. Interest rate ceilings

might lead to exclusion of some consumers from the regulated

lending market and may push them out on the non-regulated

lending market. Interest rate ceiling can be counterproductive for

the competition among lenders - the ceiling can be seemed as a

signal to the market to use a rate just below the fixed rate.

Denmark Consumer

Organisation

Two main arguments were used by opponents to APR-ceilings: 1)

An APR-ceiling will become the price-norm and thus drive up

APR's on cheaper loans 2) an APR-ceiling will lead to financial

exclusion - poor people will no longer be able to get loans when a

high risk premium can no longer be included in the interest rate.

In effect there are no alternative controls on the cost of credit.

iff/ZEW – Final Report on interest rate restrictions in the EU 387

United

Kingdom

Consumer

Organisation

Decision not to back calls from some consumer groups for an

interest rate ceiling in the UK include: - Interest rate ceilings

would possibly lead to a reduction in access to credit for some

borrowers. There is some evidence to suggest that this could

create a welfare loss as consumers face reduced options for

income smoothing/absorbing income shocks; - There is also a

concern that a rate ceiling would merely lead lower income

consumers to migrate to other forms of credit or para credit (mail

order goods, buy back stores) or take more credit requiring

security, which may not be suitable for needs and which creates

a different risk exposure; - At the extreme is the fear that some

consumers may resort to illegal lenders (loan sharks); -

Arguments that rate caps would have any significant effect on

over-indebtedness in the UK are not convincing. The sort of high

cost credit likely to be covered by a rate cap only accounts for

around 5-10 per cent of the credit debts. Instead, the significant

challenge for tackling over-indebtedness in the UK is to ensure

responsible lending practices by mainstream lenders whose

agreements make up the large bulk of the debt problems; - UK

rate cap advocates have argued that such a policy is needed to

protect the poorest and most vulnerable consumers from falling

into a cycle of debt dependency using high cost credit. While this

is a problem, a rate cap alone may not resolve the issue, given

the underlying problem of poverty and the issues raised above on

money management and migration to possibly more expensive

alternatives. Evidence that the supply of alternative low-cost

third sector credit (credit unions) is currently sufficient to pick up

the slack is not convincing. Any rate cap policy would have to be

accompanied by a policy of ensuring low income households have

better access to credit at mainstream (or near mainstream) rates

- this is after all the main hoped for policy outcome of a rate cap.

There is scepticism that a rate cap that focuses on APR's will be

effective, given that these are a poor cost comparator. Equally a

rate cap that focuses on borrowing rates is not likely to be

effective unless default charges are taken into account. For

instance overdraft credit can be by far the most expensive credit

product in the UK because of the structure of default charges, but

the borrowing rates are relatively low compared to possible

substitutes (such as payday lending). As a result a rate cap

policy is not seen as the most effective and pressing solution for

the problems in UK credit markets.

United

Kingdom

Provider

Association

Lack of consumer choice, lack of consumer access to relevant

credit products, financial exclusion, restriction of competition and

the opening up of unregulated markets. The UK Office of Fair

Trading notes [A report by Europe Economics for the Office of

Fair Trading International research: Case studies on Ireland,

Germany and the United States December 2009 OFT 1150b] that

in Germany it difficult for anyone without a Schufa rating to

obtain a loan. This has led to the development of small scale

lending from outside Germany (mostly based in Switzerland)

which is not regulated (Kredite ohne Schufa). This illustrates a

feature which is also apparent in the USA. Where there is

demand for credit but regulations seek to control the supply,

whether through interest rate caps or behavioural controls,

market operators will seek to find ways round the regulations in

order to supply the unmet demand.

United

Kingdom

Provider

Association

Concerns that the introduction of ceilings would impact on

consumer choice, lead to an increased risk of financial exclusion

and limit market competition. Existing consumer credit legislation

and protection is considered to provide a sufficiently robust

framework.

388 iff/ZEW – Final Report on interest rate restrictions in the EU

United

Kingdom

Public

Authority

Relevant considerations include: 1) the availability of other

regulatory tools; 2) the concern that imposing a single figure

would adversely impact on the wide variety of lending forms

currently available; and 3) the risk rates in the market will

migrate towards any ceiling.

United

Kingdom

Provider

Association

The UK approach is to allow prices to be set by the market

(which is competitive and open). There are a series of policy

concerns that sit behind this:

(a) research (eg. POLICIS [2004])indicates that price caps can

operate to exclude certain groups of borrowers from access to

commercial credit market. The characteristics of these groups are

(i) they only want (and can only afford) small amounts of credit

and/or (ii) they may be somewhat higher-risk because they are

having to work to tighter budgets;

b) research (eg. POLICIS [2006]) has shown that consumers

excluded in this way are much more likely to use illegal,

unregulated sources of credit. Such illegal markets are not

benign and debt recovery is routinely based on fear, intimidation

and control.

(c) rate caps restrict the diversity of credit products available to

consumers (eg. Staten and Johnson [1995]; eg. POLICIS

[2004]). This means that consumers may have to use products

that are not suited to their needs (because they are the only

ones on offer). For instance, a consumer having a pressing need

for a £300 loan may have instead to borrow £1500 because that

is the minimum loan available on the (rate-capped) market. In

this type of case, borrowing more than is needed creates a

greater risk of default (and the problems associated with it) for

that person.

(d) it is well-documented that rate caps cause the displacement

of charges, to the detriment of consumers (eg. Staten and

Johnson [1995]). So, for instance, the levels of (non-interest)

charges payable if payments are missed tend to increase. The

result is that consumers lose price transparency and consumers

who (through no fault of their own) miss payments, can find that

they are more heavily penalised and suffer increased detriment

on top of the problems they already face.

(e) rate caps also distort markets in another way by forcing more

credit to be tied to the purchase of goods (eg. Staten and

Johnson [1995]). If the credit charge (or part of it) is built into

the cash cost of the goods, the price of those goods rises

(coloured pricing). The eventual outcome (observed in US

studies) is a two-tier retail system, which is, in effect, another

form of price displacement. Consumers constrained by the price

cap are forced to use those (higher-priced) stores or retailers

willing to extend them credit on purchase of goods. In the

meantime, those consumers who can access normal cash credit

are able to drive much better bargains from a far wider range of

stores.

Note: the term “Public Authority” has been used to assemble banking authority, financial regulator and

government officials together.

Selection 4: Some answers to the question: “What have you learnt from

experience in other EU Member States and other countries that have introduced

IRR?” (SQ 1.19)

MS Stakeholder Other EU lessons

Belgium Provider

Association

We learned that, compared to neighbouring countries, since the

beginning of the nineties the Belgian legislator provided for a

whole series of IRR which are stricter than in these countries. As

a consequence, Belgian providers are suffering from competitive

iff/ZEW – Final Report on interest rate restrictions in the EU 389

disadvantages compared to providers operating from a country

with a less restrictive regulation.

Bulgaria Public

Authority

The main thing learnt is that the situation in every country is

unique and it is not possible to directly copy&apply IRR policy of

other EU member states, without taking into account the different

macro- and micro-economic factors

Czech

Republic

Provider

Association

Such measures are always based on a variety of reasons -

cultural, historical, economic, legal... Therefore the main lesson is

never to assume partial knowledge without knowing the context.

Regulating through the licensing regime was seen by provider

respondents as appropriate, drawing on the experience of the UK.

Another respondent said that other EU experience has provided

evidence that the regulation is evaded (eg. in Slovakia). A further

respondent mentioned that without social / municipal / guided

lending there is an increased likelihood that a ceiling would cause

consumers to turn to illegality without any redress.

Denmark Consumer

Organisation

Subprime lending is reduced, or non-existent. SMS-loans are not

possible in a number of Member States. In Denmark, with no

effective controls, there are major problems with high-cost

consumer credit.

Estonia Public

Authority

While drafting the regulation, practices from Germany, France,

Sweden, Netherlands etc. were scrutinized.

France Other Apparently the countries that have no regulations do not seem to

complain.

Germany Consumer

Organisation

We have actually been asked by colleagues and member-states

on our usury rules a couple of times. We have furthermore

learned that price caps are more eagerly applied in other

countries than here eg. with early repayment fees. We see that

European law is only introducing the concept now.

Hungary Consumer

Organisation

In countries where there is a welfare state, IRR works, but in

case of an emerging country, introducing caps or other controls

meets difficulties.

Latvia Public

Authority

So far Latvia has not evaluated experience of other EU member

states in relation to IRR, since there have been no plans to

introduce IRR in Latvia. If decision will be taken to do so, other

states experience will be evaluated in search for best practice

that could be incorporated in the Latvian market.

Netherlands Consumer

Organisation

Most states have not these strict rules, which leads to room for

different interpretations about what is “too high” for credit prices.

Poland Provider

Association

Regardless of their form IRR seem to cause a decrease in credit

availability for particular consumer groups. The effectiveness of

the ceiling much depends on the supervisory authority in charge.

Portugal Consumer

Organisation

The ceilings released show that the Portuguese pay interest on

consumer credit three or four points above the average of other

European countries. Take the example of France where the

maximum rate of the cards in the third quarter of 2009 was set at

21.4% (in Portugal was 32.8% in January).

Romania Provider

Association

EU experiences have only been look at for early repayment. In

countries where there is not an organisation with direct oversight,

a licensing regime would be appropriate. An example of good

practice would be Great Britain which chose not to introduce such

IRR

Slovakia Provider

Association

Licensing regime is the best way for all - consumers, regulators

and providers. It sets up clear rules how to do the business, avoid

mis-practices on the market and strongly limit illegal lending.

Spain Consumer

Organisation

EU studies from other countries constitute a good starting point.

We have taken notice that in Belgium, mortgages are usually

390 iff/ZEW – Final Report on interest rate restrictions in the EU

capped at the top and at the bottom.

UK Provider

Association

IRR are an old-fashioned idea and have been tried in many forms

for many centuries. This includes the UK, where an Act of 1713

fixed the legally permitted maximum at 5% for both business and

consumer loans. Crowther observed that 'there is a fair amount of

evidence that during the eighteenth century the limitation was

observed, although of course various evasion devices existed,

such as the issue of annuities at 10%...and both evasion and

avoidance became more commonplace in the fifty years before

repeal [in 1854]'. So 'work around' techniques, described in our

answer to Q1.10 above, are not new. These distort markets so

that, as Robert Reich put it in 1979: 'Ironically, usury ceilings are

most harmful to citizens they were apparently designed to protect

- relatively poor credit risks.' There is a rapidly-growing body of

data that is starting to capture the modern detail of the market

distortions that price caps create. Looking, for example, at illegal

lending, PFRC/POLICIS [2006] estimated from surveys that

165,000 households in the UK (with no price caps) were using

illegal lenders. POLICIS [2004] estimated from surveys that

Germany and France (both countries with price controls) had

markedly higher illegal lending markets. The German illegal

sector was two and a half times bigger (c.400,000) and the

French illegal sector three times bigger (c.500,000). Japan is an

example of a market where it is now generally accepted that

credit price controls are creating serious societal harm, with

growing criminalisation of illegal lending (see, for example, HSBC

analysis of Japanese market 17 March 2003). Looking at

displacement effects (for explanation, see our answer to Q1.10),

OFT [2010] describes some of these at play in the German

market. German consumers unable to access the mainstream

market often borrow from Swiss and Luxembourg lenders (ie.

geographic displacement, 'Schufafrei') and also use mail order

credit (ie. price displacement into cost of goods). POLICIS [2004]

notes that one side effect of French rate caps is that higher-risk

borrowers tend to use mainstream lending models that they

struggle to manage, and suffer default consequences as a result.

Pawnbroking systems are often overlooked in these discussions,

but are highly relevant to the debate. Commercial pawnbroking

(as for instance in the UK) can only operate with fairly high APRs.

A methodology found in continental Europe (where pawnbroking

has been state-run) is to systematically under-value the pawned

goods, so as to achieve a much lower headline rate of charge. In

summary, there are many examples from across the world of the

ways in which rate caps distort markets, to the detriment of

consumers.

iff/ZEW – Final Report on interest rate restrictions in the EU 391

UK Public

Authority

Studies carried out for the Department for Business suggest that

some countries that operate a cap on interest rates have

increased levels of illegal/unlicensed lending. The FSA's Mortgage

Market Review looks to draw on experience from mortgage

markets around the world. What has been noted is the view,

particularly from the industry, that the approach to IRR can have

a marked effect on the diversity of products/choice in the market.

We have no direct experience but believe that the introduction of

IRR limits product offerings including the availability of shortterm

credit or credit facilities designed for those with thin credit

files. Studies from US markets (in many ways similar to UK credit

markets) present a mixed picture of the outcomes from rate cap

policies and raise the 'winners and losers' question. A clear theme

is many of these studies is the need to actively develop better

lower cost products for lower income households and that rate

caps are not necessarily a good 'protective' substitute for this.

Where IRR exist, consumers do not have access to small sum

short-term loans and have to use inappropriate long-term loans

with high default charges and other fees and penalties. They not

only are denied the type of credit they require, but they have to

pay a very high price for the inappropriate products.

Theoretically, the APR calculation within the EU is standardised.

But in practice this is not the case. The APR in Ireland does not

include collection charges for home credit loans, but in the UK it

does. Purely because of this difference, the de facto interest rate

ceiling of 200%APR allows home credit to operate in Ireland, but

prevents all other small sum short term lenders from so doing.

The only result is that consumer choice is limited. Not one

consumer is actually advantaged by this policy. Despite an agreed

APR formula throughout the EU, pawnbrokers in Germany are

allowed to charge additional fees for administration, safe keeping

and insurance of pawned objects, which do not appear to be

included in interest rate calculations as they would be in the UK.

Consequently pawnbrokers enjoy an almost complete monopoly

in Germany in offering small loans to the exclusion of other

lenders and the detriment of consumer choice. They are

increasingly lending to middle and higher income earners.

Standing and membership charges are not included in the APR in

Germany even if membership is a condition of the loan.

Consequently, in EU states where fees and charges can be

excluded from the APR interest rate restrictions will have little or

no effect on the actual cost paid by consumers for credit. But in

those EU states where such charges have to be included in the

APR calculation, interest rate restrictions will preclude small sum

short term loans being offered at all. Customer choice and access

to relevant credit products would be removed. See the Policis

2004 report.

Note: the term “Public Authority” has been used to assemble banking authority, financial regulator and

government officials together.

Selection 5: “If controls are to be introduced or retained, what do you think

would make for an effective regime?” (SQ 1.27)

MS Stakeholder Comments from respondents on how to make for an

effective regime for setting ceilings

Austria Public

Authority

You always have to take care that ceilings do not extent the

demanded prices. We had to experience that with the ceiling of

default interest rates. They are now always 5% over the agreed

interest rates. Controls only are effective if a supervisory board

has enough resources and the power of sanctions.

Belgium Provider Controls are already in place.

392 iff/ZEW – Final Report on interest rate restrictions in the EU

Association

Cyprus Public

Authority

A regime would be effective if it was principle based.

Czech

Republic

Provider

Association

Any form of regulation always allows providers to go around it

(and thus made the system less transparent to the client).

Regulation should not focus on the economic parameters (the

rates of their structure, fees, etc.), but effective control will enable

clients to easily determine the price at which they can buy (as in

ordinary supermarket). The decision to purchase is already on the

client.

Czech

Republic

Other 1. ceiling should be set probably at the relative levels (ie. certain

multiples (2 - 4) or excess over the market basic rate); 2. another

crucial issue is - control of the creditors and instruments and

practices they use.

Czech

Republic

Public

Authority

In our opinion on-site inspections would be effective. Moreover, if

ceiling level should be introduced, we would prefer a certain

percentage of an average rate set for different segments of the

credit market. But generally we are not in favour of any ceiling.

Denmark Provider

Association

A general clause about unfair interest rates is enough and gives a

good protection for the consumer together with the other

consumer protections rules in the CCD 2008.

Denmark Consumer

Organisation

First of all we need to acknowledge that there are different credit

markets and that they should be regulated differently. Ceilings

should take into account the providers cost of funding (the market

interest rate at a particular point in time), and the security of the

loan. For example, in Denmark mortgage lending is very secure as

you can evict your customers from their homes and sell it at

current market price. Therefore it is possible to provide mortgage

credit at the market rate (funding cost) + 0,5% (administration

fee). In terms of an APR-ceiling for consumer credit it should

probably be combined with a cap on the total cost of the credit in

relation to the amount of the credit to avoid providers prolonging

the repayment period in order to stay below the APR-ceiling. If a

one-cap-fits-all solution is chosen the Danish Consumer Council

have officially advocated that the ceiling be set at the Danish

National Banks interest rate + 15%.

Estonia Other Strict rate caps and supervision authorities for all credit providers

(also for non-banks) regulated by law.

Finland Other Should be in legislation. IRR level could be eg. compared to those

in recovery proceedings.

Finland Public

Authority

Regulations should be written in the law. We have no comment on

pricing level.

France Consumer

Organisation

The controls already exist through the French Prudential

Supervisory Authority (autorité de contrôle prudentiel)

France Other The issue is very important. The answer is YES but there is so

much difference between the development of the theory and

practice that means one should be realistic.

Hungary Consumer

Organisation

The best solution would be: Limit to variability of interest rates in

variable rate credit contracts, and the level of fees and charges

should be regulated as well.

Hungary Public

Authority

An impact assessment is needed.

iff/ZEW – Final Report on interest rate restrictions in the EU 393

Latvia Public

Authority

Effective regime would be a mixture of free market regulation for

mainstream lending with some restrictions for loans that present

risk to low-income consumers. However, this might be achieved

also by restricting certain credit products instead of setting

ceilings. It might be difficult to set one set of restrictions for all

EU, since it is formed of countries with different economic

situation.

Netherlands Public

Authority

Form: supervisory control; Level: market interest plus a ceiling.

Netherlands Consumer

Organisation

Some licensing system and punishments to withdraw the license.

Poland Public

Authority

Present regulations could be considered as appropriate.

Portugal Consumer

Organisation

More attention should be given to the non-observation of IRR; the

existing sanctions do not seem sufficient or inhibiting.

Romania Provider

Association

We do not believe that these controls can produce an effective

regime.

Slovakia Provider

Association

IRR or other price controls are not delivering the results they are

aimed for. They do not help customers, they increase costs for

legal lenders and do not limit illegal lending.

Slovenia Public

Authority

- Verbraucherkreditgesetz:Ueberwachung des Anzeigeninhalts, der

Kreditangebote und -vertraege, der Gesamtkosten- und der APRAusrechnung

sowie der Beruecksichtigung der fuer

Kreditvermittler vorgegebenen Geschaeftsbediengungen;

- Meldewesen (Regelmaessigkeit).

Spain Consumer

Organisation

A possible formula would be to proceed with the action raised,

over the next two years, where the difference between the

reference rate that is being applied (whether the Euribor or other)

and the Bank rate European Central exceed 15% of the value of

Euribor (in the case of EURIBOR) or 30% of the value of the index

in question (in the case of other indices such as the ECSC).

Sweden Provider

Association

No further controls are needed.

United

Kingdom

Consumer

Organisation

A brief answer to this might be: Controls need to be focused at

the detriment; Controls need to be part of a package of measures

aimed at reducing over-indebtedness and increasing access to

alternative forms of suitable credit for lower income consumers;

Controls need to properly consider the likely effects on different

groups of consumers - will there be winners and losers and if so

who?; Controls should avoid a priori assumptions about any

ceiling level.

United

Kingdom

Provider

Association

An effective regime results from a free, competitive market

without restrictive barriers to entry. However, if a ceiling were to

be set it would need to allow for products to continue to be offered

ie. short term, low level lending is expensive to offer and the

interest rate has to reflect this, otherwise creditors will withdraw

and consumers will be pushed to less reputable forms of

borrowing?

United

Kingdom

Provider

Association

As the UK has demonstrated, the key is transparency and clarity,

ensuring consumer choice. Other approaches would stifle

competition to the detriment of the consumer.

United

Kingdom

Provider

Association

Consumers benefit from transparency, simplicity and fairness

which is provided through a combination of statutory regulation

and self-regulation. This is far more important than imposing

market controls.

394 iff/ZEW – Final Report on interest rate restrictions in the EU

United

Kingdom

Provider

Association

Simplicity, transparency and fairness are of much greater benefit

to consumers than simplistic rate or amount restrictions. Those

three principles allow innovation, increased competition, increased

choice and increased access for consumers that are otherwise

stifled by rate and amount restrictions.

United

Kingdom

Provider

Association

We do not believe that controls should be introduced in the UK

apart from those that already exist and which we have described.

Note: the term “Public Authority” has been used to assemble banking authority, financial regulator and

government officials together.

Selection 6: “Do you think that private over-indebtedness is a problem in your

country?” (SQ 2.1)

Member States where Over-indebtedness is a small problem

There is robust, statistically valid research (eg. Kempson [2002] and CCCS Statistical Yearbooks)

which make very clear that, in the context of unsecured credit, the phenomenon of private overindebtedness

cannot be described as 'a problem'. There is a relatively small percentage of

consumers who struggle to repay their credit debts, mainly as a result of unexpected life events.

However, that percentage has remained fairly stable and may even have dropped in the last year.

In this context, it is worth stressing that unsecured credit in the UK accounts for only about 15% of

all credit outstanding. Also, assets and equity far outweigh levels of personal debt. (UK)

Until 2008, the number of default was very low and even decreasing since 5 years. Of course, due

to the financial crisis, there was an increase during 2009. Over-indebtedness remains however well

monitored. (Belgium)

Due to a lack of a consistent definition measuring the scale of private over-indebtedness is difficult.

Different studies show different figures but all contain a conclusion that over-indebtedness is not as

big an issue as in other EU Member States. (Poland)

There is generally very low level of debt in Slovakia. However, there are a few communities -

especially Roma communities, in which the over-indebtedness is serious problem due to illegal

lending practices and local usurers. (Slovakia)

Very strong self regulation which limits the maximum credit amount which can be offered based on

income and financial obligations of the borrower. Although over-indebtedness concerns relatively

few households, the problem for those households can be enormous. The group that is confronted

with over-indebtedness is small but the problems for this group are big. (Netherlands)

In general, the levels of household indebtedness are low in the Czech Republic compared to

Western Europe (around 49 % of gross disposable income compared to 93.2 % of disposable

income in the Euro zone). However, the pace of growth has been fast due to credit availability,

higher living standards and very low base indebtedness levels (since 2000 household indebtedness

increased 8-fold) - between 2000 and 2008 the annual increase of household loans over 30 %.

Current risks are relatively low - levels of mortgage loans comparable to levels of term deposits.

Due to the aggressive marketing (performed especially by some non-banking credit providers, who

do not assess creditworthiness sufficiently, and offer sharking loans in some cases) in combination

with insufficient financial literacy some groups in the Czech society already encountered overindebtedness

problems. The volumes of non-banking credit (based on approximations) have been

going down since early 2009, levels of NPLs have been steadily rising - hitting 13 % of all nonbanking

credits (end 2009). Due to the crisis the consumer non-performing credits share hit 9.35

% (end March 2010) of all bank consumer credits and the volume up by 39 % y-o-y. Also the

household insolvency started to rise (around 100 cases in January 2009 and 600 cases in January

2010) It becomes more and more important issue (currently not severe, nevertheless some data:

eg. the total volume of household loans from banks in 2000 was CZK40bln, in 2009 it was CZK897

bln (source Czech Stat. Office report- bellow link). Hand in hand - the indebtedness in rapidly

increasing, eg. according to Czech Credit Bureau was the total amount of personal declared

bankruptcies in 2009 was 2500, while in 2008 the amount was 718. (Czech Republic)

It would appear that to date the incidence in non-performing loans in the personal sector ie.

Households and Individuals, is still relatively low in comparison with other economic sectors. This

said, there will always be people who will enter into loan commitments that would be difficult to

repay according to agreed terms. Banks are however being very cautious in their lending process

and vet customers' ability to repay very carefully prior to agreeing to sanction any lending. (Malta)

iff/ZEW – Final Report on interest rate restrictions in the EU 395

Structural over-indebtedness is an issue for a relatively small number of consumers. The UK has

developed sophisticated debt relief and management structures, both statutory and voluntary

(supported by the financial services sector) that provide beneficial assistance to all consumers who

encounter structural debt problems. The UK is a sophisticated and well developed credit market

and it is inevitable that there will be some level of over-indebtedness. However as studies have

evidence this tends to be more as a result of unexpected lifestyle changes (eg. unemployment;

divorce; children; etc.) and is therefore difficult to predict. (UK)

Member States where Over-indebtedness is a severe problem

According to the data published by the National Bank of Romania in December 2009, the

percentage of overdue credit is 3,96%. However the Government announced that the budgetary

staff (civil servants, teachers, physicians, police etc.) will have their salaries diminished by 25%.

Problems with paying the instalments will expand. (Romania)

Because of historically low interest rates which encourage too big loans. (Finland)

15,2% of all lenders registered in Credit Register hold by Bank of Latvia have had at least one

violation or delay of payment. (Latvia)

According to our studies 5-10% of the population are overindebted/in the risk of being

overindebted. (Sweden)

The estimates of the severity of private indebtedness have changed. Based on a survey conducted

by BNB almost all of the 5 largest banks (having a market share of 57.4%) estimate private

indebtedness as high. (Bulgaria)

Member States where Over-indebtedness is a very severe problem

Over-indebtedness is regarded as a social risk, reflected on the known archetype indebtedness

triangle risk," translated into 3 topics: i) serious employment problems, ii) multi-overindebtedness

and iii) cost savings and / or weak social frameworks. These 3 vertices confess this phenomenon as

a complex problem (from the standpoint of financial, social and psychological issues which involves

the entire family): - It is costly to the financial system and to the public policy (justice, welfare,

housing and health), -Social stigma and psychological difficulties in the financial recovery of the

household, loss of social skills, decline in job performance, isolation and shame, physical and

psychic disturbances - high risk of social exclusion and poverty - Urgent: how much longer it takes

to be treated, the greater the breach of contract, plus interest, minus assets - a family can not

close as a company. (Portugal)

Overindebtedness is a serious problem in the Swedish society of today. According to The Swedish

Enforcement Authority there are 400 000 people that are over-indebted.(Report from 2008 about

the causes and consequences of over-indebtedness). The report 2003:04 Överskuldsättning -

omfattning, orsaker och förslag till åtgärder" (Over-indebtedness - extent, causes and measures

proposals) content almost 150 pages about over-indebtedness. Some reasons for overindebtedness

that are mentioned are divorces, bankruptcies, sickness and unemployment. The

reason why a person becomes over-indebt is however not often caused by one single reason. The

report distinguishes debts related to the individual problems (such as sickness) and to the credit

market (such as loans with high interest on a loan) and debts from the public sector (such as

government subsidies or student grants). 17 proposals to decrease people in over-indebt are

mentioned in the report. They are not evaluated yet. The two earlier mentioned national

associations for people in debt have also recently published proposals to solve the problems with

people in debt. "Eternity debtors" are a common word in the debate. Conclusion: overindebtedness

is an actual debate theme in Sweden, especially in media. Interesting is that fast loan

with a high interest on the loan (because no security) is increasing in Sweden. SMS-loans (from

companies that offer loans in minutes, simply by sending a text message) are a rather new

occurrence (and problem) on the Swedish market. (Sweden)

The continuation of bad credits, "restructured credit", "delay in payments" grows. In April 2010,

loan loss provisions in the banking sector overall increased by 2.3%, or 35.4 million lats (in March -

by 11.9% or 166 million lats), totalling 1.6 billion lats, or 10.6% of total banking portfolio (at end-

March - 10.3%). (Latvia)

396 iff/ZEW – Final Report on interest rate restrictions in the EU

The debt overhang is the result either of bad debt is an accident of life that causes a decrease in

revenues. If the accident of life is unpredictable it is not the same for the bad debt. But nobody

(state, consumer groups and especially not the commission’s debt) does address the issue of

restructuring. The problem of overindebtedness of individuals emerged in France two decades ago.

A law was passed at the end of 1989 in order to provide solutions for individuals who were no

longer able to meet their loan repayments. This initial mechanism has been amended three times ,

in 1995, 1998 and 2003. However, the number of over-indebted people remains, till today,

relatively important. So a new Act has just been adopted by the French Parliament in order to

improve the prevention and the resolution of such difficulties. According to the Banque de France

barometer of the over-indebtedness of December 2009, 954,000 files were submitted to the

commissions of over-indebtedness between January 2005 and December 2009, that is to say an

average of 190,986 files per annum. It's a 15% growth in only one year. (France)

The household debt, whose main cause is the mortgage credit places us at the top of the

international debt ranking family and has reached such severity that the threat of foreclosure

housing usual payroll, property and other assets of families, hangs over hundreds of thousands of

consumers. The latest analysis and existing data are already talking about more than 100,000

families in judicial mortgage enforcement procedures, figures that can bend in coming months,

according to most analysts. Not surprisingly, the General Council of the Judiciary has expressed its

concerns and has published recently that the number of foreclosures in the Spanish courts in the

period between January and September 2008 was already 53.696 and that this figure would

amount to 86.681 foreclosures in 2009 (three times in 2007) and even arrive in 2010 at 121.006

executions. (Spain)

A debt counselling service for low-income citizens has recently been established. Unfortunately

resources are insufficient to help middle- or high-income citizens that are over-indebted, but the

demand is there. The burst of the price-bubble on the real-estate market has left many unable to

meet their contractual financial commitments because they have had to sell their home at a prize

much lower than what they owe on their mortgage (due to divorce, unemployment or other

reasons. (Denmark)

We have seen the level for debt problems brought to Citizens Advice Bureaux in England and Wales

increase for over a decade. Credit debt problems were at a high level even before the recession

and have now grown further. Personal insolvency levels are high and growing. (UK)

Selection 7: Some answers to the question “Do you think that this problem

[private over-indebtedness] has improved or worsened over the last five

years?” (SQ 2.2)

The table below shows extracts of a few responses for countries where the trend is seen

as not having worsened significantly for consumers.

Czech

Republic

It worsened for certain groups in the society, especially due to the influence of

economic crisis (growth of unemployment) and low level of financial literacy (within

2009 the number of persons with delays with repayments increased by 16 %).

Nonetheless, the new insolvency legislation improved the situation of households in

the Czech Republic significantly. Also, banks, even more than before, run very

thorough creditworthiness assessment before granting any credit. Recently most of

the households became very cautious with getting new credits and/or increasing

indebtedness.

Finland In general, the trend is slightly negative. However, regarding small and short term

credits, typically SMS loans, the situation has worsened significantly during the past

4 years.

Germany There was a certain levelling, things are yet worsening, there are too few

counselling services financed to meet the requirements given the number of people

affected.

Hungary More people have taken out loans which they might not be able pay back. Many

people have even taken out more than one loan. The number of 'late payers' is

significantly growing from day to day.

iff/ZEW – Final Report on interest rate restrictions in the EU 397

Malta The problem of over-indebtedness has worsened marginally mainly due to the

current local economic conditions brought about by the on-going world financial

crisis, which resulted in increased unemployment and consequently somewhat

increased defaults on borrowers' obligations.

Netherlands Credit was tightened because of new regulations, but the economic crisis caused

loss of income, resulting in overindebtedness. It is more depending to the social

structure of the community then (acceptance of having debts is a social aspect).

Problems occur in difficult circumstances, for example unemployment. The

Netherlands have a high level of mortgage debt compared to other countries.

Actual payment problems and forced sales of homes are relatively limited, but have

increased slightly due to the economic crisis.

Poland Most people already over-indebted have a problem now, so we assume that the

problem is slightly worse. The financial crisis has affected Poland in the minimal

compared with other EU countries.

Romania Credit was introduced quite recently in Romania, just about 5 years ago. People

had no experience to deal with the advertisements. Overindebtedness worsened for

a small proportion of the population who were effected by the foreign currency

lending issue. There has clearly been a swing to higher levels of saving since the

'credit crunch'.

United

Kingdom

As a result of deteriorating economic conditions generally. However, there is a

greater awareness of debt issues and the availability of support and assistance for

those who find themselves in financial difficulties. As remedial assistance developed

in the UK over the past 15 years, statistics concerning overindebtedness suggested

a rising problem. However, it is most likely that greater awareness of remedies led

to greater reporting of problems, not actually a significant increase in the problems

themselves. Also, in March 2010, BIS published a report containing an assessment

of over-indebtedness in Britain, based on a set of indicators

(http://www.bis.gov.uk/assets/biscore/consumer-issues/docs/10-830-overindebtedness-

second-report.pdf), which attempted to compare evidence with

previous analysis in this area. Although the data were not directly comparable in

every case, it suggested a slight increase in the proportion of people who could be

described as 'over-indebted'.

398 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex XII: Provider Questionnaire

iff/ZEW – Final Report on interest rate restrictions in the EU 399

400 iff/ZEW – Final Report on interest rate restrictions in the EU

iff/ZEW – Final Report on interest rate restrictions in the EU 401

402 iff/ZEW – Final Report on interest rate restrictions in the EU

iff/ZEW – Final Report on interest rate restrictions in the EU 403

404 iff/ZEW – Final Report on interest rate restrictions in the EU

iff/ZEW – Final Report on interest rate restrictions in the EU 405

406 iff/ZEW – Final Report on interest rate restrictions in the EU

iff/ZEW – Final Report on interest rate restrictions in the EU 407

408 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex XIII: Provider Questionnaire - Methodology and feedback

The survey consists of national providers of both secured and unsecured consumer

credit. The list of contacted providers include commercial banks (mainstream), non-bank

mortgage lenders, non-bank point-of-sale lenders, credit card companies, moneylenders

such as payday lenders and door-to-door lenders, and pawnbrokers.

The questionnaire has been sent to individual providers directly in March and April after

an initial invitation call/email and has been filled in by providers from 4 of the selected

countries (UK, Germany, Netherlands and Sweden). In addition to that providers from

Czech Republic and Norway have submitted a written comment on the study hypotheses

and a personal interview has been conducted with a major home credit provider from

Poland. Providers who have explicitly refused participation have been additionally asked

to indicate their reasons. Unfortunately, the input we received for the different countries

is not comprehensive enough to conduct quantitative analysis on the European data.

However, it allows presenting some important insights on the hypotheses.

From a total of 137 credit provider companies contacted we have received a total of 18

responses, 3 submitted written comments on the hypotheses of the study and 31 explicit

refusals. Thereof 14 providers have already refused when first being contacted by phone

(before receiving the questionnaire) and 17 providers have refused to participate after

having received the questionnaire. The response rate of 13% is comparably high by

industry standards.

The providers in Germany and UK have filed most of the explicit refusals. In UK we have

therefore also obtained a relatively high participation rate 33% in contrast to Germany

where the response rate amounts to only 5%. All the request to French providers have

been left unanswered and despite of repeated reminders sent by email (in French) and

repeated phone calls we have not received any indication of reasons for choosing not to

participate. In Sweden and Netherlands the response rate was respectively 7% and 12%.

Frequent reasons to refuse:

Country

Returned

Question

naires # Refusals Branch

Survey

contents

does not

apply to

company

Not

authorized

Not

interested

aire takes

too much

time /They

are too

busy

Hang up/

forward to

nowhere

Language

Barriers Other

Do not

participate

in studies

in general No Reason

FR 0 4 commercial bank

mortgage specialists - 2

pawnbroker

1 1 2

DE 1 9

commercial banks - 2

point of sale lenders - 3

mortgage specialists - 3

pawnbroker

2 1 1 2 3

NL 1 3 non-bank mortgage

specialist pawnbroker

point of sale lender

2 1

PL 0 5

commercial bank

pawnbrokers - 3

non-bank mortgage

specialist

1 1 1 2

SE 1 3 point-of-sale lenders - 3

commercial bank

1 1 1 1

UK 11 8 commercial banks - 2

credit card companies - 3

pawnbrokers - 2

4 1 1 2 2

14 32 7 1 4 2 4 1 5 3 8

Most of the providers who have explicitly refused participation indicate that IRR are

either not relevant for them or consumer credit is not their main activity. In particular, as

indicated in the responses to the Stakeholder Questionnaire, in some countries, including

Germany, as well as in particular segments there is a little debate on this issue and thus

little interest in participating in the survey.

iff/ZEW – Final Report on interest rate restrictions in the EU 409

Comparably large number of providers have indicated that as a part of their corporate

policy they do not participate in surveys or do not provide prognosis-related numbers.

A further group of refusals is based on lack of interest or time (considering the length of

the questionnaire). It appears to us that many providers do not see their advantage in

taking the time to participate in a survey, as they are affiliated with associations which

are supposed to do this kind of communication for them.

Further feedback on the PQ:

Concern has been raised regarding the ability of the survey to provide a meaningful

comparison among different European countries and different market segments.

Furthermore, the definition of different lender categories is considered too general. In

particular, specifications of sub-groups of credits from the category “moneylenders” were

omitted. As a result, respondents from the sector claim that the indicators use were

either irrelevant or misleading. For instance the applied interest rate measure (APR) is

considered to be a misleading indicator regarding short term/low value debt. Compared

to other market segments, APR seems to adequately reflect the costs of credit in the

home credit industry as it includes all fees associated with doorstep loans. In contrast, in

other market segments administration fees and other fees associated with the loan are

included in the APRC. We take account of this difference when interpreting the survey

results. The critique on the use of APR can also be linked to the discussion on which

interest rate should a possible restriction be applied. From the responses of home credit

providers, it can be concluded that an interest rate restriction on APR instead of on

interest only (as in the case of Poland) would result in an offering of longer-term home

loans only due to the APR decreasing with maturity. In this regard the TTC (total costs of

credit) has been suggested as an alternative measure.

Respondent from the home credit sector have been reluctant to answering questions

regarding “high-risk” borrowers as they argue that the risk is a function of credit amount,

not an individual. Furthermore, the request to provide an indication of average market

rate for markets for which there is no such value publically available, such as the home

credit sector, was met with criticism. The lack of publically available value makes it even

more plausible to address the question on the average market rate exactly to the

providers as they have the best insight on the market segment in which they are acting.

Finally, participants seem reluctant to give indication on the case Portugal as they do not

have sufficient knowledge and experience on the market. Furthermore, respondents have

reported that the question has raised confusion and led to initial misunderstanding of the

aim of the study. Similar to the previous point this has been a reason for several

respondents to initially ignore our request to fill in the survey questionnaire.

410 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex XIV: Provider general feedback on IRR

Pawnbroker, UK, denied participation:

On the market mechanism:

My general opinion on interest rate caps is that they are a clumsy tool to address an issue which is

already managed in a competitive market. Issues surrounding competition and unfair / misleading

practices are already enshrined in law and protect the consumer.

There have been various studies that demonstrate internationally that rate caps have distorted the

market and led to increased in unregulated credit and further financial exclusion. The OFT in the

UK has already issued comments to this effect as part of its recent High Cost Credit Review.

Credit card issuer, UK, filled in PQ:

Any simple link between average rates in the credit card market and a fixed ceiling will restrict the

ability for the industry to apply risk based pricing correctly and therefore cut off credit to the

underserved market. In effect it will increase the population of the UK who will not be able to

access credit through reputable lenders. This is inconsistent with UK policy.

Commercial bank, UK, filled in PQ:

On the market mechanism:

The UK has a wide, varied and sophisticated credit market which is highly competitive. Access to

credit is determined based on the customer’s and lender’s risk profile ie. ability to pay and appetite

to lend. X believes that charging interest rates based on risk is a feature of prudential lending. The

transparency of interest rate and charge structures offered by lenders and the ability of debtors to

switch from lender to lender is a far better control on any possibility of unacceptable behaviour by

lenders than IRR.

On costs of credits and illegal lending:

X believes that interest rate caps do not work and could lead more consumers into debt or force

them to borrow from unscrupulous lenders who operate on or outside the fringes of regulation. In

other countries which impose a maximum interest rate, the cap becomes the normal price for

credit and rates that are lower than the cap all converge towards it. If the limit is set too high,

consumers end up paying more, and if the limit is set too low, lenders will be forced to exit the

market leaving vulnerable customers prey to loan sharks.

On the impact on supply:

The most likely impact of an IRR would be a reduction in supply and therefore consumer choice.

An IRR would be likely to result in adverse selection as the demand for credit from borrowers that

would have been priced-out of the un-capped market increases. These borrowers are more likely to

have been higher-risk borrowers and as such it is likely that levels of defaults would rise. As a

result of this, lenders are likely to offset their additional losses by increasing the average interest

rate for all customers. Both theory and evidence suggests that as the average interest rates rise,

lower risk borrowers would borrow less while individuals of lower credit quality will either remain

unaffected or increase their demand for credit. This will lower the average credit quality of a

lenders book. At the extreme, only poor credit quality households would borrow, leading to market

failure. To avoid adverse selection/deterioration in credit quality, lenders will tighten credit

restrictions, for example by tightening score cards. This is likely to lead to a reduction in the total

supply of credit.

Home credit provider, UK, filled in PQ:

On illegal lending:

The very real risk is that a price cap would cause this (illegal lending) to increase. The research by

POLICIS for the UK Government said that France and Germany (with price caps) had higher levels

of illegal lending than the UK.

iff/ZEW – Final Report on interest rate restrictions in the EU 411

Home credit provider, UK, filled in PQ:

Most groups realise that introducing a rate cap would cause companies to be more selective & offer

longer term products which would make credit difficult to obtain for anyone wanting a short term

product to help them with an unforeseen short term cash flow problem. The problem for small

business is the difficulty of borrowing from the banks. A rate cap would reduce the choice

customers have for short term products. Many customers do not want to take out a loan for say 2

years, as the APR may be lower but the total cost of credit could be higher, when they could take

out a loan for 16 or 22 weeks with us & have it paid off in a short space of time.

On illegal lending:

The result of too much regulation could result in lenders pulling out of the market & increasing the

market share of illegal moneylenders.

Home credit provider, UK, filled in PQ:

On the impact on demand:

In our view, rate caps do not affect demand for credit (see for instance POLICIS [2004]). Instead,

they affect supply of legal credit.

On illegal lending:

The result of too much regulation could result in lenders pulling out of the market & increasing the

market share of illegal moneylenders…In fact, the reality is that if a price cap suppresses supply,

that will mean more illegal lending.

On the impact on administration fees and other fees:

So, for instance, if the availability of legal cash credit is suppressed by means of the cap, one

would expect to see more retail credit (where the credit charge is built into the cash price). When

the credit charge is built into the cash price of the goods it is easy to circumvent any price cap

(which will only apply to an explicit credit charge).

Credit card issuer, UK, filled in PQ:

On international competition:

During the past 14 years we have also tested our business model in Spain, Italy and France

(through a partnership agreement with a local Spanish bank). In each case we have decided not to

pursue a consumer lending business. A number of reasons have contributed to this decision- they

vary from the limitations on availability of information about consumers (data sharing is essential

to our business model so that we can make responsible decisions about whether to grant credit to

an individual); to restrictions on interest rates; to restrictions on our ability to collect on debts; to

other regulatory requirements that make the business opportunity unattractive.

On credit access:

It is better to include rather than exclude consumers, who may otherwise have to resort to other,

possibly unregulated, sources of credit.

Home credit provider, PL, in a personal interview:

IRR increase the intransparency in the charges and fees. IRR do not reduce the costs of credit

because of hidden costs. IRR lead to reduced credit volume and from a macroeconomic perspective

reduction on growth.

IRR harm competition. X considered a market entry in Germany but decided to refrain from

entering because of the existing interest rate cap. Even if yet not binding an existing cap adds

uncertainty to the providers as the cap level can be changed in future to a more restrictive one.

The level of over-indebtedness in a country is influenced by intercultural differences in the

subjective perception of debt. In Poland there is a general reluctance to debt in contrast to UK.

Interest rate regulation should focus on stress testing the banking system.

412 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex XV: Stakeholder general feedback on IRR

(Government official, Austria)

On the one hand a regulation concerning the level of fees and charges is welcome but on the other

hand a regulation may minimize the competition. From consumer's view it would be important to

regulate a maximum level of fees and charges.

(Provider Association, Czech Republic)

Regulation by the cap (usually relative) makes sense in some areas. Generally does not make

sense.

In case of delay is necessary to ensure compliance with the conditions on which the debtor and the

creditor agreed contractually. The solution is not a cap.

Level of fees and charges should not be regulated. Client must be able to easily obtain information

about them.

The market is just the difference between the offered goods. Unification leads to a reduction in

diversity of supply. The same applies to credit caps.

(Provider Association, United Kingdom)

There are no consumer benefits, only detriment, caused by regulated interest rates.

IRR lead to significant consumer detriment, lack of choice, lack of competition, lack of access and

unregulated lending.

Caps are detrimental to consumers and, in NO jurisdiction where they are used, have resulted in

cheaper, accessible loans. On the contrary they have led to circumvention, higher prices,

inappropriate products and greater consumer indebtedness.

There should be no caps. If there are, any fee or charge that is excluded will simply be used to

circumvent the cap. No jurisdiction has a cap that cannot be circumvented.

(Other Activity, Finland)

Because we don’t have IRR at all in Finland, we have SMS loans with very high IR. IRR would

vanish those.

If it were easier to get out of debts it would make it easier to have a fresh start. On the other side

access to credit would be more difficult.

(Provider Association, Germany)

IRR exclude consumers with a lower creditworthiness from obtaining credit. This is because banks

are not able to cover the actual costs of credit for those consumers due to IRR. If insolvency would

be made easier, banks would have to face higher losses and would therefore have to compensate

higher costs in their credit prices.

There should be different ceiling levels, since there are different credit products with different risks.

Eg. a car credit has a lower risk than an unsecured loan.

Access to credit always depends on the individual creditworthiness of a customer that has to be

thoroughly assessed when applying for a credit. The availability of credit products for low-income

consumers increases, when using risk-based pricing. IRR at a low level are generally minimizing

access to credit for low-income consumers, since banks are then not able to adjust their pricing to

the risk of the borrower. Risk-based pricing generally raises credit access for low-income

customers. IRR at a high level (eg. average times 3) are sufficient in order to avoid overindebtedness.

Limited access to credit is a vital mean in order to avoid over-indebtedness and an

expression of responsible lending.

(Provider Association, Italy)

There have been phenomena of credit rationing and no direct benefits in term of fighting usury;

consumers are aware credit providers simply abide to rules: NO CIRCUMVENTION.

iff/ZEW – Final Report on interest rate restrictions in the EU 413

Compliance costs are significant: no impact assessment has been conducted.

IRR are not effective to fight usury.

There is no evidence of a positive effect of rate restriction in the control of over-indebtedness. No

understanding of responsible lending requirements and disclosure.

(Provider Association, United Kingdom)

Free market competition and a realization that imposition of IRR would restrict the market and

reduce the availability of credit.

Research indicates that IRR results in less product diversity; and access to credit becomes

restricted; and lenders withdraw from the market; and pricing is increased elsewhere eg. fees.

IRR would make it uneconomic to lend to some customers. This would impact on the pricing and

availability of credit for the rest of the market.

Any regulatory action that restricts the ability to lend will limit future over-indebtedness.

Alternatively, making consumers liable for borrowing irresponsibly.

(Provider Association, United Kingdom)

Rate caps are shown to create a series of adverse unintended outcomes and tend to harm those

they are intended to protect.

(Provider Association, Romania)

In countries where there is not an organisation with direct oversight, a licensing regime would be

appropriate. An example of good practice would be Great Britain which chose not to introduce such

IRR.

IRR by their nature are an inappropriate and inadequate form of regulation.

Free market competition should establish prices. Improve access to credit. However, the underlying

principles behind treating customers fairly should be applied in particular around transparency, fair

market practice and clear customer communications.

All caps are a bad idea.

We do not agree with caps however there are clearly different sectors and different products within

the market which would require different types of regulation.

(Banking Authority, Malta)

The capping of interest rates would spell the end for certain local licensed institutions.

(Provider Association, Slovakia)

Licensing regime is the best way for all - consumers, regulators and providers. It sets up clear

rules how to do the business, avoid malpractices on the market and strongly limit illegal lending.

IRR generate big costs for legitimate lenders as they would need to take into account the added

compliance costs. We do not believe that any IRR can provide effects, which are displayed above.

IRR are inappropriate as they do not respect different risks of different customers. It is highly

ineffective and do not provide customers any value.

IRR or other price controls are not delivering the results they are aimed for. They do not help

customers, they increase costs for legal lenders and do not limit illegal lending.

Caps should not exist at all.

(Other Activity, Estonia)

These restrictions will have effect only if higher interest rates and/or cost of credit are prohibited

and punishable by law.

(Other Activity, France)

414 iff/ZEW – Final Report on interest rate restrictions in the EU

As stated above, we are rather for the removal of regulation on the rate of wear.

It takes freedom.

(Provider Association, Czech Republic)

Any form of interest rate restriction would have a substantial impact across all sectors of the

market.

Free market should establish the level of interest rates.

An effective regime of controls was already introduced by CCD (Act. No. 145/2010 Coll.). No new

controls would help to make the market more effective.

(Provider Association, Denmark)

A general clause about unfair interest rates is enough and gives a good protection for the consumer

together with the other consumer protection rules in CCD 2008.

(Provider Association, United Kingdom)

Where they do not exist, this is primarily because they will: reduce access to credit; reduce

consumer choice; increase financial exclusion; increase the cost of credit; force consumers into the

unregulated credit market; they do not benefit consumers; it will have a number of unintended

consequences.

Consumers benefit from transparency, simplicity and fairness which are provided through a

combination of statutory regulation and self-regulation. This is far more important than imposing

market controls.

(Consumer Organisation, United Kingdom)

We are open to the argument but remain to be convinced that a rate cap would, on balance,

benefit to UK consumers. We are also unconvinced that rate caps are the most pressing credit

issue for UK policy makers.

We believe that there is a need for a package of policy intervention, although we are not sure what

role (if any) interest rate caps should pay in this.

(Consumer Organisation, Portugal)

The law only sets ceiling on IRR and for certain credit contracts; this rates encourages institutions

to extend credit risk and even mass advertising, encouraging consumers to unnecessary

consumption starting a spiral of indebtedness.

(Government official, Latvia)

Regulation on default interest rates is important to prevent over-indebtedness of consumers.

However, it should be general enough stating overall rules and boundaries and not numeral caps.

Fees and charges should be regulated by market to offer consumer competitive products for

adequate price. However those charges should be adequate and justified.

If ceilings are to be set, they should be flexible enough to promote creating of new products and

allowing consumers access to credit.

Caps should differ according to type of loan.

(Provider Association, United Kingdom)

It has not been deemed necessary to regulate interest rates. Rate restrictions would not enhance

the UK's existing legislation but would potentially restrict the market, constrain competition and be

to the detriment of the consumer.

The use of the IRR model does not exist in the UK. We believe that the focus should be on

transparency and clarity ensuring that consumers can make informed choices and understand how

the product works and the cost attached to it.

iff/ZEW – Final Report on interest rate restrictions in the EU 415

The introduction of IRR would distort the market and lenders would gravitate towards the cap.

As we have seen within the UK, increased regulatory burden results in additional cost for the credit

market and this is, inevitably, reflected in the cost to the consumer and a reduction in competition.

As has already been stated, we do not believe that IRR deliver sufficient consumer benefit when

viewed against the impact on the competition, product availability, and consumer choice.

(Provider Association, Czech Republic)

The main impact of the caps would be worsening of the access to credits. A rejection rate in the

banks will increase just because of the individual cost of risk (of worse-profile clients) will not cover

maximum rate.

Regulation of variability does not bring anything else than restricting to variety. It certainly would

not protect consumers in any aspects; it would only limit them in their choices.

Regulation of prices does not mean that consumers are going to pay less, mostly quite contrary. It

means neither wider access to credits nor wider range of products offered, again both quite

contrary.

(Provider Association, Belgium)

A too restrictive regulation has counterproductive affects: closure of the market, less innovative

products…

The caps must vary in relation with reference indexes of the market (either mortgage or consumer

credit).

(Provider Association, Netherlands)

The effect of IRR is that potential interest margin for suppliers will be limited.

The limitation of interest rates will lead to a worsening of the profitability of small loans in

particular. Micro Finance will thus be negatively affected by restrictions.

(Public Authority, Bulgaria)

The main thing learnt is that the situation in every country is unique and it is not possible to

directly copy and apply IRR policy of other EU member states, without taking into account the

different macro- and micro- economic factors.

Level of the interest rates depends on risk - a possible cap would stop credits towards risky but

potentially profitable investments. Moreover entrepreneurs with no credit history would not be able

to borrow as the bank would not lend at higher rates.

If IRR regime is introduced, it should be constantly monitored and changed according the pace of

the development of economic cycle, in order not to restrict the economic growth.

(Government official, United Kingdom)

Concerns that introducing ceilings on interest rates could lead to a decrease in access to legitimate

sources of credit and an increase in use of unlicensed sources of credit.

(Government official, Finland)

Reasons not to introduce ceilings relate mainly to fear that vulnerable consumers would fall outside

the normal credit market and a black market would emerge.

Without any restrictions on default interest rates, these interest rates would be higher and thereby

cause more problems to vulnerable consumers. Rules on default interest rates are fairly well-known

among consumers.

If fees and charges are not included in the cap, rules will probably be very ineffective and easy to

circumvent.

(Consumer Organisation, Germany)

416 iff/ZEW – Final Report on interest rate restrictions in the EU

A main objection is the general attitude that prices are to be influenced by the market, not the law.

Rules und court rules on fees may yet always refer to inadequate and one-sided influence on

pricing as market rules do not or do not adequately apply.

(Consumer Organisation, Denmark)

In effect there are no alternative controls on the cost of credit.

In relation to consumer credit there are major problems due to the lack of IRR. We have three

distinct markets for consumer credit: 1) consumer credit obtained through banks are usually

reasonably priced and the consumers ability to repay the loan is assessed 2) consumer credit

obtained in retail stores or via the internet have APR's of 20-60% and the consumers ability to

repay the loan is not assessed at all or not assessed in a responsible manner (hence we have many

over-indebted low-income consumers) and 3) SMS-loans and Web-loans with APR's of several

thousand percent (typically you borrow 1000 DKK and repay 1300/1400 DKK after 30 days). A cap

of 30% APR would not have an impact on the first market where APR's are already below the cap

and price competition puts (moderate) downwards pressure on prices. On the second market

lenders would have to rethink their business models. Lower risk premiums on the individual loans

means that they no longer could afford the same number of customers defaulting on their loans

and (better) assessments of the consumers ability to repay the loans would have to be made. This

means that prices would be lower and that some low-income consumers would be denied access to

credit. This could be a good thing as 30% is still very high and anybody with a reasonable chance

of paying back the credit would be able to borrow at this rate. In terms of the third market, a cap

of 30% APR would be the end of SMS-loans and Web-loans in Denmark as the business model is

premised on the ability of lenders to absorb a large number of defaults through high interest rates.

The common solution to reducing over-indebtedness relating to lending from the three distinct

credit markets is to increase the providers financial risk when engaging in irresponsible lending. On

the first market (as well as the mortgage market) more consumer friendly 'regulation on personal

bankruptcy' and 'tighter responsible lending requirements' would have the most pronounced

effects. On the second market (which contributes the most to over-indebtedness) interest rate

restrictions would have the most pronounced effects. On the third market the loans are relatively

small and are not as such causes of over-indebtedness.

First of all we need to acknowledge that there are different credit markets and that they should be

regulated differently. Ceilings should take into account the providers cost of funding (the market

interest rate at a particular point in time), and the security of the loan.

In terms of an APR-ceiling for consumer credit it should probably be combined with a cap on the

total cost of the credit in relation to the amount of the credit to avoid providers prolonging the

repayment period in order to stay below the APR-ceiling.

A sophisticated model for the setting of ceiling levels generally, could be that the National Banks of

the member states regularly calculate 3 caps based on, for example, the market interest rates of

3-, 7- and 10-year repayment periods plus say 5% (to cover administration, reasonable profit and

defaults/risk). The exact level could be calculated by analyzing average administration costs,

reasonable profit and a risk premium/default rate that is politically acceptable.

(Financial Regulator, Netherlands)

Consumer credit interest rate should be capped to protect consumers. Mortgages don't need

capping because there is enough other regulation for protection purposes.

(Consumer Organisation, Romania)

Why regulate the prices in the financial market and not also in the vegetables market?! Other types

of regulations are needed for banks, that is when investing and speculating deposits; but not for

the interest rate for credit. Better is to educate consumers not to spend money they haven't

earned yet; save first, spend later. Of course, a special case is the mortgage credit. Here we have

a misbalance in the present time: consumers had obtained a credit for a house whose price now

has diminished.

If low-income people will be given greater access to credit (that is by imposing low interest rates),

it will be unfair towards other people, with medium-income.

(Government official, France)

iff/ZEW – Final Report on interest rate restrictions in the EU 417

The credit market is much more competitive now than it was when the legal usury rate was

introduced. Still, the financial crisis led the French government to consider that IRR are an effective

protection against excessive risk-exposure by money lenders and borrowers. Indeed, sub-primes

and near-primes do not exist in France. The interest rate of revolving credit is relatively lower than

in other countries, though it nears the usury rate with an average APR around 16-17%.

All in all, consumers have benefited from the interest rate ceilings. On the one hand, this ensures

that money-lenders will not use their market power to enforce abusive interest rates. On the other

hand, it does prevent the riskiest consumers from accessing the credit market. The general

consensus in France is that lesser access to credit is preferable to an increased risk of overindebtedness

for low-income consumers. The consequences of the development of sub-prime

markets in other countries in the past years have strengthened this consensus.

Overall, the French government is satisfied with the IRR system.

(Banking Authority, Spain)

We do not favour introducing IRR, either caps or other types of controls.

418 iff/ZEW – Final Report on interest rate restrictions in the EU

Annex XVI: Observations on SME and microenterprise lending

As explained in the introduction and elsewhere (see Chapter 2.1 and subsection 2.1.5 on

page 157), the study is primarily concerned with credit markets to consumers as IRR is

primarily designed for reasons of consumer protection rather than for prudential reasons.

Because the general purpose behind IRR to protect borrowers from the securing of

excessive or abusive economic rent by the lender is also valid for the lender-borrower

relationship for business loans to small and medium sized enterprises (SME), this Annex

will outline some definitions with regard to business lending, market description for

microcredit, and summarise some recent legislative changes with regard to IRR that have

affected credit markets for businesses in France.

Small and medium size enterprises (SMEs) and microenterprise lending

SMEs are usually classified using methods based on the number of employees

(headcount), annual sales (turnover) and value of assets and net profit (balance sheet).

According to the European Commission’s definition from its Recommendation of

6 May 2003 concerning the definition of micro, small and medium-sized enterprises311,

enterprises qualify as micro, small and medium-sized enterprises if they fulfil the criteria

shown in the graph below.

Source: European Commission Recommendation 2003/361/EC.

Though small enterprises, defined as enterprises which employ fewer than 50 persons

and whose annual turnover or annual balance sheet total does not exceed €10 million,

will also be affected by IRR, it becomes increasingly unlikely that firms will be

significantly affected by these the larger they are. One reason is because the financing

regimes become structurally very different as the firm’s reliance on banking

intermediation as a source of finance decreases and capital markets offer increasing

financing possibilities. This is in strict contrast to small enterprises who are compelled to

rely on bank financing and the existence of a sustainable commercial relationship with

their banker. One reason this relational banking is so important for SMEs is because

there is not the large quantity of public information available that comes with being a

larger business, and as such bankers need to be well acquainted with the needs and

appreciation of risks of its client’s business ventures. Furthermore, statistically, firms

with larger turnovers become less reliant on only one bank and are far more likely to

311 See http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/.

iff/ZEW – Final Report on interest rate restrictions in the EU 419

have multi-banking channels for loans. SME lending is also distinguished from corporate

lending to larger firms by bank regulators in their assessment of risk.312

A further distinction can be made between small businesses313 (whether privately owned

corporations, partnerships, or sole proprietorships) and micro-enterprises, this later

category covering the smallest businesses, often located in private homes, single-family

operated business with few (or no) employees other than the owners. There is also a

notable trend to further segment different-sized microenterprises; for instance, the term

Very Small Business is now being used to refer to businesses that are the smallest of the

smallest, such as those operated completely by one person or by 1-3 employees. This

attempt to define a microenterprise reflects recent EU entrepreneurial and employment

developments and testifies the aims of the European Commission to encourage the

adoption of measures addressing the specific problems micro enterprises face, especially

during the start-up phase (see microcredit section on the following pages).

A differentiation between lending to SMEs and consumers is nevertheless common in the

definition of banking based on the object of the lending activity ie. between those

conducted by a commercial bank (also known as a business bank) which takes deposits

and gives loans to businesses or corporations, and those by a retail bank which deals

with individual citizens or consumers as opposed to companies or other banks (primarily

handling savings and checking accounts, mortgages, and personal loans).314 Banking

provided to individuals, referred to as retail banking is thus distinct from business

banking where loans are provided to SMEs.

IRR and its effects on credit markets for SME and microenterprises

The first observation is that EU SME financing is not subject to IRR to the same extent as

consumer credit markets. Ceilings only exist in Italy and France (for overdrafts only). The

study thus focused primarily on IRR with respect to consumer credit (mortgage and nonmortgage).

Because similarities between SMEs and consumers exist, such as their

dependency on the banks for their financing (unlike larger firms that can access capital

markets) and vulnerability to tightening credit exposures by the banks as a consequence

of stronger bank capital requirements as at the current time, the findings of the report

can be extrapolated and are generally applicable to an assessment of IRR effects on

credit markets to professional credit (business loans for commercial purposes).

Nevertheless, some of the differentiating factors between the two distinct users of

finance for economic activity are outlined below:

Overdraft facilities or credit lines can be far more substantial: This facility

with a stated maximum amount that an enterprise is entitled to borrow from a

bank at any given time is of far greater relevance and necessity for businesses

that have larger inflows and outflows affecting their liquidity or cash flow position.

312 The so-called Basel II rules where bank appraisal of ex-ante individual counterpart risk and credit risk in

their portfolios, and where retail credit and loans to SMEs receive a different treatment to corporate loans

which require more regulatory capital for given default probabilities. The main reason for this differential

treatment is that small business loans and retail credit are generally found to be less sensitive to systematic

risk. Their risk of default is thought to be largely of an idiosyncratic nature and, as a result, default

probabilities are assumed to be more weakly correlated when compared with corporate loans. Another

reason for the preferential treatment of retail credit lies in a technical assumption by the Basel Committee

that maturities are shorter. See “Credit risk versus capital requirements under Basel II: are SME loans and

retail credit really different?” BIS, Tor Jacobson, Jesper Lindé, Kasper Roszbach, August 2005.

313 Examples of small businesses include: convenience stores and other small shops, hairdressers, tradesmen,

lawyers, accountants, restaurants, guest houses, photographers, small-scale manufacturing etc.

314 Note however that there is another common classification of credit banking based on the distinction

between commercial banking (generally accessible to anyone with banking needs) and merchant banking

(serving mainly large companies and very wealthy individuals), which include SME and household finance in

both definitions.

420 iff/ZEW – Final Report on interest rate restrictions in the EU

Personal credit cards may be used for SME source of financing: The more

severe consequences of liquidity constraints, mean that entrepreneurs tend to

resort to credit cards for financing when other sources of credit are unavailable.

Credit cards offer the self-employed a convenient payment mechanism and easily

accessible way to borrow. Whereas large firms are more likely to use business

credit cards, the smaller firms will often only have their personal credit cards.

Credit standards may differ: These internal guidelines or criteria reflect a

bank’s lending policy and written or not written define the types of loan a bank

considers desirable and undesirable, its designated geographical priorities,

collateral deemed acceptable or unacceptable, etc.

Credit terms and conditions are more varied: The specific obligations agreed

upon by the lender and the borrower may consist of more than the direct price or

maximum size of the loan, as SME lending is also subject to other terms and

conditions in the form of non-interest rate charges (ie. fees), collateral

requirements (including compensating balances), and loan covenants.

Debt restructuring may be more common: As more rational economic agents

with repeated use of credit, enterprises may be more prone to use debt

restructuring than households ie. to avoid defaulting on existing debt or to take

advantage of lower interest rates or lower interest rate expectations.

Greater variety of non-interest rate charges: Various kinds of fees that can

form part of the pricing of a loan for SMEs are less relevant for credit to

households, such as commitment fees on revolving loans, administration fees (eg.

document preparation costs), and charges for enquiries, guarantees and credit

insurance.

Greater product and service offering: Among the additional services provided

to businesses are more treasury services, revolving credit, merchant credit, cash

management, group insurance, corporate cards and secure internet banking.

Differing profitability potential: SMEs may be extended credit at lower rates of

interest than households, since SMEs may bring in more money to the bank as

compared to retail banking customers who tend to need more maintenance

compared to their deposit sizes.

More alternative sources of financing: Other than a bank start-up loan, an

entrepreneur wanting to start a business can obtain financing from either their

own personal savings, friends and family (with lower returns expectations), loans

guaranteed by specific entities, hire purchase, factoring, and the possibility to

obtain equity investments from venture capital firms or other investors if

operating a more sophisticated business.

Timely interventions in times of crisis: Because of the role SMEs play in terms

of employment and investment spending for economic growth, businesses are

more likely than consumers to receive help from the authorities as was seen in

the recent financial crisis where the European Commission for example introduced

an updated Late Payments Directive in April 2010 and the European Investment

Bank made €30 billion available for SMEs.

Greater requirements for documentation: Because lending to businesses is

inherently risky (especially smaller businesses), providers will want assurances

that their funds will be repaid. The documentation which banks generally require

from an entrepreneur in order to assess a borrower’s risk is far greater than for

households. Eg. to obtain a bank loan, businesses must typically be prepared to

provide several years of financial statements, information on existing debts along

iff/ZEW – Final Report on interest rate restrictions in the EU 421

with accounts receivable and payable, lease details, projected future income

streams and signed personal financial statements. Microenterprises and potential

borrowers from guarantee funds will also need to prove their good character as

well as their expertise and commitment to business success (with expected

contributions from their own funds).

These specific factors of SME lending may have a more or less determining role in

upholding the conclusions of the report with regards to the hypotheses investigated with

regards to the effects of IRR on aspects of the credit market. However, because the

empirical work and quantitative and qualitative fieldwork was focused on credit to

consumers, any deviations in the assessment of the validity of our set of hypotheses

cannot be totally ignored.

Microcredit in the EU

Microcredit315 is the extension of very small loans (micro loans) to the unemployed, to

poor entrepreneurs and to others living in poverty that are not considered bankable.

These individuals typically lack collateral, steady employment and a verifiable credit

history, and therefore are unable to meet the minimal qualifications to gain access to

traditional credit. Entrepreneurs often find it difficult to borrow small amounts, because

many banks see microcredit as a high-risk, low-return activity, and their handling costs

are high in relation to the lent amount.

Microcredit should not be confused with short-term or small-sized loans. Although there

is some overlap, microcredit serves a specific purpose and is not associated with

consumer credit as such. Neither should it be confounded with business loans for SMEs

more generally as these will often be profitable ventures in themselves. Microcredit is

considered in the EU as a way to encourage self-employment and the development of

microenterprises. These microenterprises represent over 90% of all enterprises (of all

sizes) in the EU. Though there is still no internationally accepted definition of

“microcredit”, it is generally defined as the provision of credit in limited amounts to lower

income households and small, informal businesses. It is thus broadly viewed as a

business line that can be carried out by a wide range of institutions providing a range of

financial services, such as lending, deposit taking, insurance, payments and funds

transfers.

The European Commission has launched a number of important initiatives to promote the

development of microcredit in the EU including a definition which is that of a loan of up to

€25,000. The Commission states that increasing the supply of microcredit is important

for encouraging new businesses, stimulating economic growth (in line with the EU

growth-and-jobs strategy) and opening doors to people who would not otherwise have

such opportunities (EU policy on social inclusion). Microcredits are increasingly being

used as a tool against poverty and exclusion in the EU and is increasingly seen by the

authorities as a credible tool in promotion of social cohesion. Eg the European Union has

launched the JEREMIE initiative (Joint European Resources for Micro to Medium

Enterprises), in collaboration with the European Investment Bank (EIB) and the European

Investment Fund (EIF) aimed at improving SME access to finance and venture capital.

In general terms, micro-credit in Europe addresses two groups: “micro-enterprises”,

defined as enterprises employing less than 10 people and “disadvantaged persons”

(unemployed or inactive people, those receiving social assistance, immigrants, etc.) who

315 Microcredit is a financial innovation which originated in developing countries where it has enabled extremely

impoverished people (mostly women) to engage in self-employment projects that allow them to generate

income and to begin to build wealth and ultimately to exit poverty. These microcredit borrowers are

increasingly seen by traditional bankers as pre-bankable as microcredit gains credibility in the mainstream

finance industry.

422 iff/ZEW – Final Report on interest rate restrictions in the EU

wish to go into self-employment but do not have access to traditional banking services.

Micro-credit is of particular importance for rural areas and can play an important role in

helping to integrate ethnic minorities, and immigrants both economically and socially.

Over recent decades, the EU economy has evolved from one driven by large-scale

industrial enterprises to a greater dependence on smaller businesses, including one–

person business, mostly in services.316 In Western Europe, microloans are largely given to

individuals in poverty and to those excluded from the banking system. Given the size of

the banking system in European countries, we are nevertheless talking about a very

small percentage of the population.

As a result of EU and national initiatives, there are increasing numbers of statistics

available to measure the characteristics of EU microcredit. The number of loans

disbursed, the total value of loans outstanding, and the average size of a loan are thus

available based on a survey conducted by the European Microfinance Network (EMN)

which are reproduced below.317 Though not a survey including the complete entirety of

microcredit organisations, a total of 84,523 microloans worth €828 million were

disbursed in the EU in 2009. With 26% and 40% of these respective values attributable

to activity in the Eastern countries, the average loan value is higher in the Eastern

countries as in the Western EU countries. The greatest number of loans disbursed by

country is lead by France (28,863)318, which is also the country with the largest number

of active clients (70,252) followed by Finland, Romania and Spain. The largest number of

microcredit providers is found in Italy (via the large number of small social microcredit

initiatives and anti-usury associations), Bulgaria (via commercial banks) and Hungary

(savings cooperatives and Local Enterprise Agencies), followed by Spain (via savings

banks) and the UK (via public, private and non-profit development finance agencies,

CDFAs). The majority of these EU organisations work at a national (70%) or local level

(32%), with only 6% operating cross-border.

The average microcredit loan size for the EU was €9,641 in 2009 and varied from

€19,000 in Finland to €2,500 in Latvia with Finland, Sweden, Belgium, Hungary, Ireland

and Germany all in the group of countries with loan amounts above the average. Caution

in interpretation is nevertheless necessary because these averages do not reflect

differences in standard of living or relative poverty levels, and may hide the real situation

such as the case in France, where different organisations extend credit on different terms

including loan size, namely ADIE representing 97% of the market with an average loan

size of less than €3000. This example not only shows that the source for these figures

(the EMN mentioned above) should perhaps have used the median rather than the mean

for calculation of the average, but demonstrates that microenterprises may often be

targeting different consumer segments. This is true in France, where ADIE is

fundamentally geared towards the socially included, whereas other microlenders are

financing the development of more economically robust microenterprises or businesses.

The regulatory and supervisory practices may differ across countries and institutional

types ranging from: Banks (institutions licensed for taking deposits from the general

public and subject to banking supervision), and Non-banks which are made of up Other

deposit taking institutions (ODTIs, institutions that are authorised to collect deposits

316 See: The European initiative to develop microcredit in support of growth and employment (2007),available

at http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2007:0708:REV1:en:PDF. On p6 of this

document, it is stated that one aim of the initiative is to “….help micro-credit to become sustainable by

relaxing interest caps for micro-credit operations”.

317 See: Overview of the Microcredit Sector in the European Union (EMN 2008-2009, June 2010). Available at

http://www.money-advice.net/media.php?id=3910.

318 In France, two organisations, a not-for-profit organisation (France Initiative) and an NGO (ADIE) disbursed

14,050 and 13,997 loans respectively in 2009, with the difference that disbursement of funds by the former

was in the form of zero-interest quasi-equity loans that allow beneficiaries to access significant

complementary bank credits, as opposed to stand-alone interest bearing loans.

iff/ZEW – Final Report on interest rate restrictions in the EU 423

without necessarily being subject to the same regulation and supervision as banks eg.

financial cooperatives319 and Microcredit institutions (MCIs - institutions that lend for

limited amounts, without being authorised to collect deposits).

Microcredit has a number of distinctive features such as product design, client profile and

labour-intensive underwriting methodologies, four main features are320:

Microborrowers: A microcredit provider usually caters to low-income clients,

both the underemployed and the entrepreneur with an often informal family

business. Borrowers are typically concentrated in a limited geographic area, social

segment or entrepreneurial undertaking (eg. when socially motivated investors

have a special interest in targeting a given region or social segment, such as

women or immigrants). Loans are usually very small, short term, and unsecured,

with more frequent repayments and higher interest rates than conventional bank

loans.

Credit risk analysis: Loan documentation is generated largely by the loan officer

through visits to the borrower’s business and home. The borrower’s character and

willingness to repay is also assessed during field visits. Credit bureau data are not

always available for low-income clients or for all types of microfinance providers.

Credit scoring, when used, complements rather than supplants the more labourintensive

approaches to credit analysis.

Use of collateral: Microborrowers often lack collateral traditionally required by

banks, and what they have to pledge is of little value for the financial institution

but are highly valued by the borrower (eg TV, furniture). Where the lender does

take some sort of collateral, it is for leverage to induce payment rather than to

recover losses.

Progressively increasing lending: Customers who have limited access to other

financing are usually dependent upon ongoing access to credit. Because some

microlending uses incentive schemes to reward good borrowers with preferential

access to future, larger loans (sometimes with favourable repayment schedules

and lower interest rates), this may increase the risk of over-indebtedness.

For these reasons, it is understandable that many providers require higher interest rates

to offset higher operational costs involved in the labour-intensive microlending

methodology (eg. in the absence of collateral, underwriting depends on careful analysis

of the household’s repayment capacity and the borrower’s character). However, as

covered in the main part of the report with respect to credit provision of small sizes more

generally, high interest rates for sustainable microcredit are also the result of the fact

that a portfolio of very small loans are usually more costly than the same total value of

lending in larger amounts, as not all costs vary in direct proportion to the amount lent.

However, development of national microcredit sectors is not solely determined by the

access or non-access to finance, and will largely depend on the desirability and

prevalence of self-employment against a status of employee (whether motivated by

opportunity or necessity). Statistics in this regard show that the status of self-employed

is most preferred in countries like Italy and the UK. Italy incidentally, a country where

business loans are subject to usury ceilings, suggests that these ceilings may not actually

constitute a significant barrier to self-employment opportunities.

319 A term that includes a diverse group of member-owned financial intermediaries referred to as credit unions,

savings and credit cooperatives, cooperative banks etc.

320 See: “Microfinance activities and the core principles for effective banking supervision - final document”,

Bank of International Settlements, Aug 2010, at http://www.bis.org/publ/bcbs175.htm.

424 iff/ZEW – Final Report on interest rate restrictions in the EU

Because the issue of IRR and microcredit is a worldwide one, it may be useful to outline

some suggestions from a recent industry report on responsible pricing in microcredit

other than the most common option of using interest rate ceilings.321

Interest rate ceilings: Primary virtue of caps being their simplicity (easy to

understand, and the same standard gets applied to all financial institutions) and

their greatest weakness in the developing world is that interest rate caps apply

the same standard to all financial institutions, regardless of their location, type of

clients, loan size, or loan term (as we have seen, this is not the case in the EU).

The potential negative effects of caps should they be set unrealistically low have

been explored in main part of this report.

Margin Caps: A ceiling on the difference between the underlying costs incurred

by a financial institution and the amount it can charge its borrowers. This method

takes differences in costs between providers into account eg. a version proposed

by Nobel Prize winner Muhammad Yunus has a traffic light system based on the

difference between an MFI’s cost of funds and the interest rate it charges (Green

Zone which he calls poverty focused institutions whose interest rates are 10% or

less above their costs of funds, Yellow Zone between 10 and 15%, and a Red

Zone of money lenders and loan sharks who charge 15% or more above their cost

of funds). Again, this method has the advantage of simplicity. A person only

needs to know two numbers in order to determine the reasonableness of a price,

an institution’s cost of funds and the interest rate it charges. However, this

simplicity makes it very difficult to apply this standard across a wide range of

locations and conditions (a study showing that 75% of all MFIs would land in the

Red Zone, especially NGOs with small average loan sizes because the formula

does not take into account differences in operating costs).

Return on Equity Policies: This involves not looking at the price but to examine

whether the financial institution is receiving excessive profits from the poor people

it serves (ie. pricing structures that result in very high profits, high return on

equity (ROE), could be made more affordable without endangering institutional

sustainability). By deliberately setting an ROE target, MFIs hope to manage the

tension between institutional profitability and institutional mission, however

limiting profits would have little effect on the institutions with moderate interest

rates, and the approach could obscure inefficiency in financial institutions.

Comparative Transparency: Another approach to responsible prices avoids the

setting of ceilings or limits and instead advocates for a public listing of the prices

for all microfinance products in a country, using a common method for defining

the price (APR and effective interest rate (EIR) on these products posting this

information on a website (eg. MicroFinance Transparency has a graph depicting

each loan product plotted by its APR and average loan size, which produces a

market price curve and makes it easy to see which loan products carry prices that

are far above the market average for those sizes of products - though the graph

does not distinguish between loan purposes (education, housing, business,

consumption etc.). The idea being that the tool (common definition of price,

providing information on each loan product rather than averages, showing the

information differentiated by loan size, and collecting information on all loan

products for a single country at one time) could lead to downward pressure on

interest rates either through MFIs seeing their APRs compared to others,

regulators will be able to identify those who charge significantly above the market

for a given loan size and put pressure on them to lower prices, MFIs charging

321 See “Responsible pricing: the state of the practice”, the Smart Campaign and MicroFinance Transparency,

July 2010) available at: http://www.money-advice.net/media.php?id=3921.

iff/ZEW – Final Report on interest rate restrictions in the EU 425

lower than market rates may advertise this fact. But weaknesses are the long and

labour intensive process of collecting data, the uncertainty of frequency of

updates and accuracy over time, microfinance clients will not have ready access to

this data when purchasing, showing comparative data may not help much in

markets where there are few providers who charge similar prices.

Promoting Competition: This approach has worked somewhat so far where as

microfinance has become more popular, a greater number of providers have

entered the market putting downward pressure on average interest rates in

developing countries.

A case study – France (legal changes in 2003 and 2005)

The issue of IRR is seen as particularly relevant to very small enterprises322 presenting a

level of risk greater than the average SMEs (eg. young enterprises or those in the

creation phase). This was the case in France at the end of the 1990’s when a period of

low base rates meant that a number of these very small enterprises were not allegedly

being served by banks due to the usury ceilings in place limiting the ability of banks to

cover the costs of lending to them.

A study of the issues of IRR and discussions over the modernising and updating of the

existing mechanism featured in length in the 2000-2001 annual report by the

consultative committee of the French central bank.323 Back in 2000, the usury ceilings

applied to business credit (which have a history of change324) were seen as particularly

problematic because the economy was at a point in the cycle where average rates were

low and the formula for setting the ceiling (x1.33) was seen as too constraining in

comparison with its limiting effect when average rates are higher, and the time-lag

between the date of capture of the data and subsequent period for which the ceiling is

applied was seen as too unresponsive (ie. a delay of over 5 months could be constraining

in periods of high volatility in rates). When rates are low, providers argued that the risks

involved with small enterprises or specific phases of their development, could not be

sufficiently covered when the rate differential was so small, leading to banks refusing

these enterprises a credit. Providers also stressed that the APR definition used for the

ceilings made this problem particularly acute in times of low rates. A further point raised

by the providers was that interest rate ceilings were justified on the grounds of consumer

protection and that this need for protection was greater for a consumer than a business -

citing a European Commission communication making the distinction between an

experienced (in this case the business) and non-experienced person (the consumer).

Providers also challenged the legality of the ceiling on businesses on the grounds of the

law of 1993 which transposed the stipulations of the law of 1966 into the consumer code

(and thus outside the jurisdiction of businesses). During the debates, the enterprises

interviewed made the following observations: small enterprises with a weak bargaining

power vis-à-vis the providers justified protection; the rationing effects of the ceiling had

no quantitative basis; the ceilings only had an effect on the very small enterprises and

322 In France the term very small enterprise (très petites enterprises, TPE) is equivalent to the definition of the

EU for microenterprise (less than 10 employees) however micro-enterprises is a more restrictive term in

France. Microenterprises are almost exclusively sole-proprietorships (one-man business, called entreprises

individuelles, EI), and their status confers them a specific tax regime (tax exonerations such as from VAT).

323 See: Rapport annuel 2000-2001 du Comité consultatif (former CCSF of the Banque de France), pp. 21-37,

available at http://www.banque-france.fr/archipel/publications/ccsf_cc_ra/cc_ra_2000_2001.pdf.

324 A brief history of IRR affecting SMEs in France since the French revolution includes the law of 1807

reinstating a ceiling - but confusing the usury ceiling (taux d’usure) with the legal rate ceiling (taux légal)

when setting these at 5% for civil matters and 6% for businesses; the law of 1886 which removed the rate

ceiling for businesses (and then for civil matters as well in 1918 on economic grounds to promote

reconstruction), before being reinstated in 1935 before reaching its contemporary regime through the laws

of 28.12.1966 and 31.12.1989.

426 iff/ZEW – Final Report on interest rate restrictions in the EU

those starting-up or developing; those not being served by banks turned to other

alternative methods of financing including approaching foreign banks for loans; certain

business ventures would benefit from having a better quality of banking relation and

provision of counselling and information.

When the conservative majority reached power in 2002, the authorities proposed a law

to relaunch economic activity which included a revision of the usury ceilings applicable to

businesses.325 Although the reform in its initial version was suggesting only an increase in

the setting of the interest rate ceiling for businesses (ie. a relaxation by modifying the

fixation method to the average APR plus 15 percentage points), the resulting law of

August 2003 led to its removal altogether, albeit, not applicable to credit that was

extended in the form of an overdraft nor to those credit contracts that concerned their

non-commercial activities. While this 2003 law disappointed microcredit advocates by

only concerning credit to moral persons, a subsequent 2005 law then also extended the

exception from the usury ceilings to business credit extended to those individuals

exercising a professional activity (eg. the self-entrepreneur). The result of the Dutrueil

laws of 2003 and 2005326 has been the removal of usury ceilings for credits to businesses

and the self-employed, with the exception of overdraft credit which is still subject to the

applicable usury ceilings, however with the notable exception that the sanctions for

breaching of the ceilings set for overdrafts to businesses are no longer penal and only

civil. Below is a description of some of the views expressed during the legislative process.

In the first reading, end 2002, the members of the parliament discussed the arguments

put forward for the change in the regime of usury ceilings applicable to businesses

(Article 17), namely that these ceilings excluded the enterprises displaying the highest

level of risk from access to bank loans and that if the ceilings were to be set at a higher

level this would enable banks to accept to finance more risky project ventures. This

argument however, did not convince the parliamentary committee in charge because of

its reservations that the efficiency of the suggested arrangements were difficult to

measure and not sufficiently justified with regards to the risk of a hike in interest rates

charged to existing businesses that could result from a less restrictive setting of the

ceiling. At the time of the discussions, outstanding loan amounts to businesses were

about €54 billion of overdrafts and €110 billion of loans of maturity less than two years.

Although average interest rates for business start-up loans on the market were observed

at 6.7%, more than 2 percentage points below the interest rate ceilings for business

loans, the alleged credit rationing was subject to intense debate as was the observation

that these very small enterprises not receiving access to credit were being forced to turn

to other alternative sources of financing solutions that were more expensive than the

classic business loans (eg. factoring, leasing of real property or personal property, hirepurchase

etc..). Furthermore, those business creators unable to obtain a bank loan, were

said to have tried to finance their venture through consumer credits, a credit category

subject too much higher (looser) ceilings of over 18%. Though this resorting to consumer

credit would likely be reduced as a result of the supply side reaction to the change in

ceilings for business loans, the resulting balance between the additional loan volumes

distributed to newly served borrowers and increase in income from existing businesses

was discussed. While it was argued that the intensity of competition did not warrant fears

that provider lending margins would rise, other arguments voiced included that in reality,

325 Leading to the Loi n°2003-721 du 1 août 2003 pour l'initiative économique (1).

http://www.legifrance.gouv.fr/affichTexte.do?cidTexte=JORFTEXT000000414455&dateTexte=. The

reasoning behind the law for economic initiative was based on finding solutions to the creation, selling on

and taking over of businesses, along 5 lines, of which one entitled “measures enabling a better financing of

economic initiative” consisted of an investment fund (fonds d'investissement de proximité), fiscal

advantages (tax deduction of up to €10,000), doubling of the loss deductability threshold, and faciltating

credit access to bsuinesses through the relaxing of the fixation rules of the interest rate ceilings.

326 Loi n° 2003-721 du 1er août 2003 pour l'initiative économique and the loi n° 2005-882 du 2 août 2005 en

faveur des PME.

iff/ZEW – Final Report on interest rate restrictions in the EU 427

the conditions of a loan are primarily dependent on the guarantees which the business

can bring to the table, and that demands for small amounts of credit entailed

management costs that made bankers reluctant to lend. One suggested solution put

forward put never actually discussed involved excluding from the interest rate ceiling the

portion of costs related to the setting up of the loan.

As part of the reform, two and half years after adoption of the 2003 law, legislators

mandated the Bank of France to evaluate its effects on lending to small businesses and

to the expected spur to small business creation and development. This report327 was

published in 2006, and notwithstanding some alleged improvements along the lines of

the hypotheses put forward for reform, the overall findings were rather inconclusive,

especially in the face of some criticism of the methodology applied in the surveys used

and data interpretation.328 In the opening sections of its report, the French central bank

writes that the setting of interest rate ceilings supposes implicitly that it is difficult to

reach an economic “fair price” on credit markets, and that the assumptions of perfect

competition, perfectly accessible information and no externalities, underlying a free

market determination of rates, is particularly tenuous with regards to SMEs as opposed

to larger firms. The report claims that deregulation and removing the ceiling on loans to

businesses had the following favourable effects: It made the access to credit for SMEs

easier especially for the very small businesses as well as those who had recently been

created (with more than 20% of banks having increased the number of loans extended to

the self-employed and SME in all sectors – increase in the number of risky enterprises

served but also the amount extended - including less delays in granting of credit); Led

lenders and borrowers to give up less favourable credit solutions in favour of traditional

forms of credit; Favoured more access without increasing the default risks as it had been

feared; Helped the creditors to better select their borrowers and to better price them by

taking into account the true cost of the risks involved and reducing their reliance on

additional guarantees.

However, from the reading of the report, the reform cannot be said with any certainty to

have had the desired impact on access to credit. Firstly, the reform was actually one that

was weakly noticed by the firms themselves, with the report establishing that there was

in fact very little awareness by SMEs of either the pre-existing interest rate ceiling

regime or knowledge about the reforms having taken place in 2003 and 2005. In

addition, the reform was seen by the SMEs themselves as not having had any significant

effect on their relations with their banks. Secondly, some other factors may have played

a role in determining the effects outlined above, including heightened competition as a

result of globalisation, e-banking, new microcredit institutions (which the report

acknowledges), and unnoticed interest rate rises because the non increase in interest

rates are observed in a context of general decline in rates in practice. Lastly, seeing that

a decrease in lending was being witnessed Europe-wide and acknowledging the new

orientation in bank lending towards risk-based allocation of credit, it seems difficult to

agree with the main argument behind the political aims of deregulating interest rates at

the time, which was that the conjectural decrease in lending was due to the existence of

usury ceilings.

As seen previously, France is the most active European country for professional

microcredit. Though this has been intentionally promoted by the above mentioned

reforms, it is also due to the overwhelming success of Adie, which has been giving media

coverage to microcredit for over 20 years, and to the more discreet field work of France

Initiative. While both microfinance networks are different in their structures, their

327 See: “Les incidences de la réforme de l’usure sur les modalités de financement des petites et moyennes

enterprises” (Rapport au Parlement, Banque de France, Décembre 2006) available at

http://www.responsible-credit.net/media.php?t=media&f=file&id=2441.

328 See: http://www.responsible-credit.net/index.php?id=1980&viewid=39376.

428 iff/ZEW – Final Report on interest rate restrictions in the EU

philosophies and their targets, they nevertheless have the common feature of being notfor-

profit associations. Since then, further reform helping microcredit and business

creation includes the creation of a “self-entrepreneur” status (autoentrepreneur) in 2009,

and the development of a new form of microcredit serving a different purpose to

traditional microcredit for businesses. These personal loans rather than professional loans

(called microcrédit social or microcredit personnel) may target the same target group

with the same counselling functions, but are distinct because the funds are used for

purposes that are the same as consumer credit, namely personal project financing such

as for specific housing tasks, help with employment and skill development, mobility, and

family events. A report by the French Government on microfinance from March 2010329

underlines the social and economic utility of microcredit and explains the different models

of microcredit extension that exist in great detail, such as the bank guaranteed

microcredit.

329 See: http:/www.economie.gouv.fr/services/rap10/1003rapmicrocredit.pdf.

iff/ZEW – Final Report on interest rate restrictions in the EU 429

Annex XVII: Quantification of the implications of H1 and H2a

Our discussion of H1 and H2a reveals that an influence of interest rate restrictions on

the existence of certain credit market segments is highly plausible. In particular,

providers in high cost credit market segments such as home lending or payday lending,

which rely on an interest rate and fee structure implying high levels of (annualized)

interest rates, may not be able to operate their businesses. As a consequence, some

groups of the population (those perceived to have high risk and demanding small-amount

credit) may be excluded from regular credit access.

Given this concern, one may attempt to quantify the importance of lacking credit

opportunity in countries with interest rate restrictions. However, transferring the

experience of one country to another requires strict assumptions on the behaviour of

both demand as well as supply side in the credit markets. In particular, one implicitly

assumes that …

1) … the demand side behaves equally across countries. Relatively low levels of

credit incidence in one country (compared to another) need to be attributed to the

lack of supply, rather than to different attitudes toward credit or differing living

conditions in the respective countries, which might otherwise influence the

demand side.

2) … interest rate restrictions are the only institutional difference between several

countries. One needs to assume that, once interest rate restrictions are adapted

to the standards of another country, the entire business environment of suppliers

is perceived to be equivalent in those two countries.

3) … interest rate restrictions are fully effective in shutting down a specific

market segment. One needs to assume that lenders do not have the opportunity

to structure contracts in such a way that interest rate restrictions do not apply.330

4) … there are no alternatives to high cost credit for a subgroup of

consumers. In particular, one needs to assume that lenders do not offer any kind

of credit to some groups of the population when interest rate restrictions are in

place nor are alternative options for these groups.

The UK is frequently taken as a benchmark country without interest rate restrictions. For

the UK, OFT (2010) estimates an upper bound for the market size of home lending in

2008 of GBP 1,300 million (EUR 1,636.44 million).331 This corresponds to a per-capita

amount of home lending credit of EUR 26.79, and to a share of 0.67 percent of the total

consumption credit market. This figure also corresponds to 0.09 percent of the Gross

Domestic Product (GDP).332

Given that one is willing to accept the assumptions above, one may therefore take these

ratios to determine projections for potential market sizes in other countries where the

institutional environment (here: interest rate restrictions) still prevents high cost credit

markets. In the following, we illustrate this idea based on the four case study countries

with interest rate restrictions.

330 Note that this assumption is at odds with H7.

331 The lower bound of GBP 1,200 is not much lower.

332 The 2008 population of the UK is 61 million, the consumer credit market amounts to EUR 245,217 million

and the GDP amounts to EUR 1,812,076 million.

430 iff/ZEW – Final Report on interest rate restrictions in the EU

Table 64 demonstrates the results of this approach. Take Germany as an example of a

country in which interest rate restrictions prevent home credit lenders from operating as

they do in the UK: assuming that every citizen would, on average, take the amount of

home lending credit as it is observed in the UK (EUR 26.79), the expected market size for

home lending would be EUR 2.2 billion. The estimate is rather similar when assuming

that home lending would arise according to a similar share of the GDP as it is the case in

the UK. In contrast, assuming that, in the absence of interest rate restrictions, home

lending accounts for a comparable share of the total consumer credit market (excluding

mortgages) as observed in the UK, the market size would be EUR 1.4 billion.

Table 64: Projection of potential home (collected) lending market size

MS characteristics Projected home lending

volume in a NO ceiling

scenario (EUR million)

Population

(million)*

Consumer

Credit

(EUR

million)*

GDP

(EUR

million)*

based

on per

capita

value

based on

% of

consumer

credit

based

on % of

GDP

Germany 82 224,046 2,492,000 2,200 1,495 2,250

Netherlands 17 23,772 590,427 448 159 533

Poland 38 32,803 360,639 1,021 219 326

France 62 155,733 1,947,210 1,669 1,039 1,758

* Values for 2008, Source: ECRI Statistical Package 2009, ZEW.

The comparison of the projections on different bases reveals that there is a higher

similarity between the projections on the per capita basis and on the per GDP basis than

on the basis of the consumer credit market size for Germany, the Netherlands and

France. In contrast, the projections on the basis of per-capita borrowing and on the per

GDP basis are very different in the Poland. This illustrates that there are large differences

in GDP/per capita between the UK and Poland, while these values are more similar for

the UK, Germany, the Netherlands and France. However, the size of the consumer credit

market is again very different (on a per-capita basis) across all these countries. Recall

that the results of all these projection strategies should be interpreted with caution as

they are subject to the plausibility of the above-mentioned assumptions 1)-4).

In a similar vein, one may estimate the scope of payday lending in the four countries

with interest rate restrictions. For the UK, OFT (2010) reports payday lending to amount

to an upper bound of GBP 900 million (EUR 1,132.92 million).333 This corresponds to a

per-capita use of payday lending of EUR 18.55, and to ratios of total payday lending

amount of 0.46 percent of the total consumer credit market (without mortgages) and

0.06 percent of total UK GDP. Table 65 presents the estimates based on different ratios

for potential payday lending markets in other countries. Note again that these estimates

should be interpreted with caution as they are subject to assumptions 1)-4).

333 The lower bound is GBP 700 million.

iff/ZEW – Final Report on interest rate restrictions in the EU 431

Table 65: Projection of potential payday lending market size

MS characteristics Projected payday lending

volume in a NO ceiling scenario

(EUR million)

Population

(million)*

Consumer

Credit

(EUR

million)*

GDP

(EUR

million)*

based

on per

capita

value

based on

% of

consumer

credit

based on

% of GDP

Germany 82 224,046 2,492,000 1,523 1,035 1,558

Netherlands 17 23,772 590,427 310 110 369

Poland 38 32,803 360,639 707 152 225

France 62 155,733 1,947,210 1,155 719 1,217

* Values for 2008, Source: ECRI Statistical Package 2009, ZEW.

While these estimates would make sense if the assumptions above hold, one needs to be

aware of the fact that these assumptions are overly restrictive and do not well

reflect the nature of credit markets in European countries: as demonstrated in several

parts of Chapter 2.5 (eg. the discussion of H1 and H2a), European countries show

differing attitudes toward credit. It is therefore unlikely that demand is universal across

countries, as assumption 1) postulates. Rather, one may expect that lower incidence of

credit in some countries is not only due to a lower credit supply, but also due to a less

pronounced preference for consumer credit. As there are different institutions (eg.

consumer insolvency laws), but also different market structures, it is also questionable

whether the assumption 2) holds. If assumptions 1) and 2) are implausible, however, the

approach to making projections for other countries drawing on data from the UK would

not be valid.

Interestingly, assumption 3) is at odds with the findings with respect to H7 that

providers raise those fees which are unaffected by the specific interest rate restriction

implemented in a country. As a consequence, credit contracts are restructured in a way

that allows the continuation of certain (high cost) credit practices, but under different

formal conditions. If this is the case, the estimates above may be an estimation of the

market share (and this is only the case if assumptions 1), 2) and 4) hold), but do not

identify the effects of giving up interest rate restrictions: the targeted markets already

exist even though there are interest rate restrictions. The only difference is that

credit contracts are formulated in a different way. As a consequence, the effects of the

(removal) of the interest rate restriction would be rather low.

To illustrate that this objection is realistic, consider the case of Poland, where a market of

home lending does exist along with interest rate restrictions. Indeed, the Polish market

for home lending is approximately EUR 330 million, which is not too far from the

projection on the GDP basis made above. This example illustrates that a dismantling of

interest rate restrictions would not lead to an increase of home credit by the amounts

suggested in Table 64. Rather, one would expect the lenders to restructure typical loan

contracts to cover their costs by the means of interest payments directly.

These considerations underline that any quantification of the effects of interest rate

restrictions requires relatively rigid assumptions. We conclude that these figures

432 iff/ZEW – Final Report on interest rate restrictions in the EU

are inherently unreliable, as the necessary assumptions are unrealistic and do not

capture the specific characteristics of consumer credit markets. However, based on the

UK experience, it appears that the markets for high cost credit account for only a

marginal share of total consumer credit market (about 1 percent). For the other

countries, there is no reason to expect that the effect of abolishing interest rate

restrictions could lead to a higher increase of credit volume.