Disclaimer

The information contained in this report is of a general nature only and should

not be regarded as a substitute for a reference to the legislation or legal advice.

© Copyright State of Victoria 2006

This publication is copyright. No part may be reproduced by any process

on how to reproduce any material from this publication contact Consumer

Affairs Victoria.

Published by Consumer Affairs Victoria,

452 Flinders Street, Melbourne, Victoria, 3000.

Authorised by the Victorian Government,

452 Flinders Street, Melbourne, Victoria, 3000.

Consumer Affairs Victoria is a division of the Department of Justice.

Printed by Impact Digital, 32 Syme Street, Brunswick, 3056.

In July 2005, the Minister for Consumer Affairs in

Victoria, the Hon Marsha Thomson MLC, initiated a

major review of Credit, led by James Merlino MP with

assistance from Consumer Affairs Victoria.

To assist this Review, the National Institute for

Economic and Industry Research was commissioned to

Dr Ian Manning and Ms Alice de Jonge, are their own

and not necessarily shared by Consumer Affairs

Victoria or Mr Merlino.

The Review is especially concerned to ensure

vulnerable and disadvantage consumers have access to

‘safe’ and affordable credit and that effective controls

exist to prevent predatory finance practices.

There are a range of market and regulatory measures

in place now to deal with these matters. The Paper

examines the rationale, background and operation

of a number of these measures including the use of

interest rate caps. Whether interest rate caps continue

to be relevant in today’s more competitive

environment is a matter for the Review to consider.

Consumer Affairs Victoria welcomes feedback on the

Discussion Paper, especially while the Consumer

Credit Review is taking place. Feedback can be

forwarded via creditreview@justice.vic.gov.au or to the

Consumer Credit Review, GPO Box 123A, Melbourne,

VIC 3000.

1.1 Finance and justice 3

1.2 From the invention of private property

to the first deregulation of credit 5

1.3 The re-regulation of credit after 1854 8

1.4 The second deregulation and its aftermath 12

1.5 History and the future 14

1.6 Conclusion 15

2.1 Interest rate caps – the NSW experience 17

2.2 Payday loans 18

4.1 Two approaches to credit price caps 23

4.2 The services provided by

financial intermediaries 24

4.3 Financial intermediary product costing 25

4.4 Policy alternatives 33

Abstract: Government controls over the price of credit have a long history, beginning in Biblical times.

In British history, borrowing and lending were first de-regulated in 1854, then were gradually re-regulated culminating a century later in a period of very tight policy control over the volume, direction and price of credit. This was followed in the late 20th Century by a second de-regulation, one aspect of which was a revival of high-cost fringe money lending. The current policy preference in Australia is to rely on competition to ensure that fair terms are available to borrowers, but in addition Victoria retains its historic interest rate caps. NSW has reacted to the recent rise of pay-day lending by moving to include fees in the calculation of the capped interest rate for short-term consumer lending. Cost analysis suggests that a structured cap would be more appropriate.

Ever since the invention of money, there has been an

intimate and often contested relationship between

government and the financial sector. Governments

issue the money on which the financial system is

based and enact the forms of contract from which it is

built. It is inevitable that the financial sector depends

heavily on government, not just for the law and order

that is necessary for all kinds of production and trade,

but for very legitimacy of the stuff in which it deals

and for the enforcement of the contracts that it writes.

Equally inevitably, there will be disagreement between

institutions in the financial sector and governments

over the expectations of governments concerning

financial sector behaviour, and over the types of

contracts that should be enforced.

A second peculiarity of the finance sector also implies

a close but potentially conflictual relationship with

government. Despite the common use of the word

‘product’ to describe the various contracts the finance

sector has on offer, the sector does not actually

produce anything directly useful. It does not produce

food, clothing or houses, or even services like haircuts.

In other words, it does not contribute directly to the

standard of living. Rather, it administers an important

part of the web of contracts on which business

relationships are founded, and having thus assisted

with the flow of production, helps to determine who is

entitled to what. Like government, the finance sector

is essential to the working of a modern economy.

Governments depend on the sector for the detailed

administration of financial obligations and

entitlements without which capitalist production

cannot take place, but do not always agree with the

patterns of production and allocations of wealth that

result.

These characteristics – money, contract law and

essentially administrative, allocative rather than

directly productive functions – bind government and

the financial sector in a close but mutually watchful

relationship. A fundamental judgement underlying

this paper is that the finance sector is in no position

to declare itself autonomous, and that governments

are entitled to exact a behavioural quid pro quo as

the price of the foundations which they provide for

the sector.

Since it is party to all economic transactions save those

conducted by barter, the finance sector inevitably

comes under scrutiny whenever the justice of

economic affairs is discussed. Where the sector is not

directly party to a transaction, but merely facilitates it

by providing the medium of exchange, the relevant

scrutiny concerns the adequacy of the sector’s

transactions services. The sector is not directly

implicated in discussions of the just price that can be

charged for food or clothing, or of the just wage that

should be paid for labour, but can come under

criticism for shortcomings in servicing these

transactions, for example excessive bank fees. These

discussions can become quite lively, because, taken as a

whole, the inter-bank payments system is a monopoly

within which there are significant opportunities for

profiteering, provided tacit agreement can be

maintained between the member banks.

The finance sector comes under rather more intense

scrutiny when it is itself party to transactions – when it

borrows and lends (Wilson 2004). On the borrowing

side, there has been a troubling history of defaults,

particularly among fringe financial institutions, but

sometimes extending into the mainstream, and

governments have stepped in to protect depositors

from fraud and over-optimism. On the lending side,

bad practices have included making loans to mates

and denying them to non-mates even when proper

risk assessment favours the latter. Market power has

also been used to exact excessive interest rates and, in

particularly bad cases, to force people into debt-slavery

(‘I owe my soul to the company store’). Malpractice in

loan allocation tends to be revealed when the loans go

bad, but the exercise of market power strengthens

rather than weakens financial intermediary balance

sheets, and is not so regularly exposed. It has not

proved hard to establish the theoretical possibility that

financial intermediaries may misuse market power, but

has proved much harder to identify actual cases of

malpractice. There has also been disagreement over

how market power should be controlled. Should

households be expected to recognise shonky deals and

steer clear of them, or should the scope for such deals

be limited by regulation? Can competition between

financial intermediaries be relied on to ensure their

good behaviour?

Malpractice is also difficult to separate from risk

management. Bias towards mates is hard to separate

from the many other factors that enter into

judgements of creditworthiness, and exercise of market

power is hard to separate from reasonable costrecovery

from high-risk loans.

Aside from malpractice, more general questions of

social responsibility arise at the level of loan allocation

policy (Manning 1995). In the daily business of

making loans, a great deal has to be left to the good

judgement of the lender, but much still depends on

general loan policy. The options can be sketched in

terms of the more general debate as to whether

corporations should maximise shareholder value,

versus ‘triple bottom line’ behaviour. According to the

former approach, a financier should invest to

maximise his profits, allocating funds so that his loans

are repaid with maximum returns. The latter approach

makes a distinction between financial rates of return

and social rates of return. Sometimes the social rate of

return is less than the financial – for example, a casino

may be highly profitable, but at uncounted cost in

broken homes and white-collar crimes. By contrast,

investment in infrastructure frequently generates high

social returns in terms of jobs created even though the

actual investment is not particularly profitable.

Judgements differ on the extent to which financial

and social rates of return diverge – at one extreme

stand those who can discern no divergence, and at the

other those for whom divergence is endemic and

justifies rigorous government control of the finance

sector. Stretton (2005) has recently argued that the

financial deregulation of the 1980s was a moral and

economic failure, and proposed a return to

government direction of lending into areas with high

social returns, such as housing. An alternative recent

approach, pioneered by Robert Shiller, argues that the

finance sector is unable to allocate funds to maximum

social benefit due to lack of appropriate financial

instruments (Shiller 2004). If the sector was rewarded

for undertaking investments with high social benefit,

and punished if its investments caused social loss, it

would develop appropriate risk pricing and

management. Shiller’s approach is salutary in that it

reminds us that the finance sector has both strengths

and weaknesses, and that reform should build on the

strengths.

And what are the strengths of the sector? Shiller would

instance its ability to manage large volumes of

transactions accurately and quickly, and its ability to

manage the risks inherent in borrowing and lending,

saving and investing. The sector’s obvious weakness is

that it contributes to booms and busts – one may

instance its collective misallocation of loans to the

paper entrepreneurs of the 1980s, and more recently

its over-allocation of funds to the urban property

markets. It also misallocates funds when social rates of

return diverge from financial rates. These two

problems are inter-related – speculation, whether in

shares or land, has a low social rate of return, and

there is an obvious case for financial innovation to

render the sector less prone to speculative

misallocation of funds. This will require careful

distinctions between the management of inevitable

commercial risks and the creation of excess risk by the

sector.

However, the concern of this paper is not at this grand

level, but rather with the millions of small consumer

credit accounts. This paper accordingly concentrates

on one small aspect of the contested relationship

between finance and government, the question of

whether government should cap interest and fees

charged for consumer credit. It thus tackles an area

where argument has been going on for centuries. With

the passage of so many centuries, the arguments both

for and against have been thoroughly rehearsed. A

historical account of the arguments is valuable, since it

puts current debates into perspective.

The notion of property, with its distinction between

mine and thine, is fundamental to borrowing and

lending. Some, but far from all, demands for

restrictions on credit practices can be traced back to

uneasiness with the institution of private property,

particularly the disruption of community caused by

the intrusion of property rights. At its narrowest, there

is a strong tradition that married couples should hold

their property in common, and therefore should not

be able to lend to one another. The moral community

whose members should hold property in common has

sometimes been enlarged to the family, or to the

village. Enlarge the village to the nation, and we

obtain the nineteenth-century left-wing slogan ‘all

property is theft’. Lest this sound like thunder from

the past, one has only to notice that native title in

Australia and the Pacific island nations is communal

rather than individual, and that there is heated current

debate as to whether it should be replaced by

individual title, the better to enable Aboriginal people

to seek debt finance for their enterprises. (The

fundamental role in capitalism of debt secured on

property is discussed in de Soto 2001.)

Despite the questions raised by native title, it may be

assumed that, in settler Australia, private property is an

accepted and foundational institution derived from

long cultural tradition – though the tradition does not

wholly justify current highly individualistic

interpretations of property rights. The advantages of

private property in terms of the care people lavish on

it, and the self-expression they achieve through it, are

such that the major religious traditions accept it as a

principle of social organisation, and also accept the

potential for borrowing and lending that comes with

it. However, the ancient texts are concerned about the

disruption of community that can result from

borrowing and lending, and in Leviticus 2535-37 and

Deuteronomy 2318-19 there are prohibitions on the

charging of interest, at least on loans made within the

Jewish community.

The Christian New Testament has extensive

commentary on property, which emphasises the

secondary importance of possessions compared to

things of the Spirit, their nature as a gift, and their

owners’ status as stewards rather than proprietors.

Despite passages highly critical of wealth

accumulation, there is no explicit prohibition against

lending at interest. Early Christian prohibitions of

usury relied on the passages in Leviticus and

Deuteronomy. The economist Alfred Marshall gave the

following typically 19th Century explanation of these

prohibitions. ‘In primitive communities there were but

few openings for the employment of fresh capital in

enterprise… Those who borrowed were generally the

poor and the weak, people whose needs were urgent

and whose powers of bargaining were very small.

Those who lent were as a rule either people who

spared freely of their superfluity to help their distressed

neighbours, or else professional moneylenders. To

these last the poor man had resort in his need; and

they frequently made a cruel use of their power,

entangling him in meshes from which he could not

escape without great suffering, and perhaps the loss of

the personal freedom of himself or his children. Not

only uneducated people, but the sages of early times,

the fathers of the mediaeval church, and the English

rulers in India in our own time, have been inclined to

say, that money-lenders “traffic in other people’s

misfortunes, seeking gain through their adversity:

under the pretence of compassion they dig a pit for

the oppressed”. (Marshall here quotes St John

Chrysostom) (Marshall 1949 p485). Based on this type

of analysis medieval Christians prohibited the taking

of usury, but somewhat inconsistently permitted rent

for the use of assets other than money, such as land.

Marshall’s view is that interest became respectable

once it was realised that capital was a factor or

production worthy of reward. An alternative historic

interpretation is that the needs of kings to borrow for

war, and of merchants to borrow to finance trade,

caused the lifting of the prohibition. In 1545 English

law caught up with practice and lending at interest

was legalised, subject to a cap of 10 per cent. This legal

maximum was gradually reduced to 5 per cent. Here

were the first caps on interest rates, introduced as part

of a redefinition of ‘usury’ from interest per se to

interest at excessive rates. (Chan J gives a history of the

relevant English law in Bumiputra Merchant Bankers

Berhad vs Meng Kuang Properties Berhad, High Court

of Malaysia, 1990.) Though they rarely state this

premise, advocates of caps still sometimes imply that

moral distinction can be made between acceptable and

excessive, or usurious, interest rates. We will encounter

this type of argument several times in the following

pages.

The Western tradition has gradually come to tolerate

interest-bearing loans, but to this day the Muslim

tradition is known for its prohibition on the taking of

interest. It is rather less well known for its approval of

profit. Combined with this approval, the Muslim

prohibition on the taking of interest can be interpreted

as an insistence, on moral grounds, that lenders

should share risks with borrowers. Joint ventures are

welcome. Buying followed by selling at a capital gain is

fine. Fees for financial services are acceptable. Not

surprisingly, despite the prohibition of interest,

modern Western merchant bankers can easily make

themselves at home in the world of Islamic banking.

However, regulations designed for standard Western

banking may be inappropriate for the Islamic

alternative. (de Jonge 1996)

Had the Venetians been Muslim rather than Christian,

Shakespeare would have been deprived of the plot of

his play The Merchant of Venice. The plot turns on an

ethically hard case, that of a merchant who can only

pay his fixed-interest debts when his trading fleet

returns to port. In 17th and 18th Century Britain

many a merchant was cast into the debtors’ prison for

no better reason than that his ships were wrecked. In

the Muslim world merchants did not bear this risk,

because their financiers were required to share the loss.

In Europe fixed interest-bearing debts continued to be

legal, but – perhaps partly as a result of Shakespeare’s

advocacy – their impact was gradually alleviated by

two financial innovations: insurance and bankruptcy.

The former spread the risk of loss from the merchant

to the insurers, and the latter required the financier to

accept losses which the merchant had no hope of

repaying. What had been true in practice – a lender

who financed a wrecked voyage would not be repaid –

now became true in law, and at least to this small

degree Western law came to incorporate the principle

of risk-sharing. From 1842 nobody in Britain could be

sent to prison for debt, and bankruptcy became an

option for non-traders as well as traders. The state

undertook to enforce contracts, if necessary by

compulsory transfer of property, but would not

enforce the deprivation of liberty. This reform, begun

in the 17th Century, was completed by the same

group of parliamentarians who legislated for the

abolition of slavery. At the time, the relationship

between debt and slavery was highlighted by the

position of the Russian serfs, whose condition was

objectively that of slaves, but who were technically

debtors.

The institution of bankruptcy put the onus on lenders

to assess and bear the risk that borrowers might go

bankrupt. The question of how far the state should go

in assisting debt recovery is still live, with South Africa

currently moving to limit debt recovery in cases of

reckless lending (Credit Law Review 2003). At the

opposite extreme, lenders in the USA are campaigning

to make it harder for debtors to declare bankruptcy.

On a more modest scale, in the Australian states

permissible debt collection practices are continuously

under review. Whether state refusal to enforce

repayments that threaten deprivation of liberty can be

substituted for caps on the cost of credit is also a live

question.

The Merchant of Venice not only concerns the

morality of exacting repayment from a merchant

whose ships have been lost at sea. The reason why the

merchant was in debt was not that he needed cash to

finance trade, but that he had made an interest-free

loan to a friend who needed cash to court a rich

widow. If the courtship succeeded, the loan would be

repaid, but if it failed it would not. A prudent

assessment would be that there is large risk in such a

loan, and the lender should quietly refuse. The

merchant, of course, was a generous man, and lent.

Even so, the play raises the question as to the moral

status of loans according to purpose. Against a

background where all borrowing and lending was

treated with suspicion, it was easier to justify a loan to

a merchant who had the prospect of profit than a loan

to finance a courtship. Given that the doyen of late

19th Century English economists justified interest as a

claim on profits, it is not surprising that earlier moralists

were uneasy about loans to finance consumption.

Against the political background of Europe in the 16th

and 17th Centuries, moralists were also occupied with

the case of loans for war finance. Wars are a negativesum

game, yet kings were often desperately anxious to

finance them. Here was an opportunity for financiers

to make high-risk loans, and for persuading kings to

guarantee the financiers’ requirements even if it meant

impoverishing their nation. The tax burdens that

resulted from wars financed by borrowing lay behind

the moral condemnation of government borrowing –

a condemnation with strong echoes in present-day

financial sector insistence on balanced government

budgets. Once again, the distant past echoes into the

present.

In the late 18th Century the movement for free trade

addressed the interest rate cap of 5 per cent, arguing

that it should be abolished in the name of liberty. The

classic defence of ‘the liberty of making one’s own

terms in money-bargains’ remains Jeremy Bentham’s

Defence of Usury, written in 1787. Bentham addresses

a series of arguments then current for interest rate caps.

1. To the argument that usury is bad by definition he

replies that it is impossible to define a usurious rate,

for who is to say that an interest rate agreed between

gentlemen is too high, or for that matter too low?

2. He counters the argument that the interest rate cap

curbed prodigality by asking whether anybody is in

a position to prevent a spendthrift from getting rid

of his assets. And who will grant loans to a person

they know will not be able to repay, whatever the

rate of interest? An interest rate cap, therefore, has

no role in the discouragement of prodigality.

3. To the argument that the interest rate cap benefits

the indigent borrower he replies that either the cap

is high enough to cover the risk of lending to an

indigent person, in which case it has no effect, or is

too low to cover the risk, in which case the indigent

person is denied the loan. By preventing the making

of loans that might be helpful, the cap is

accordingly against the interests of indigent persons.

4. He argues that regulating markets to protect lenders

from risky lending is not necessary, since lenders

should be able to assess the risks for themselves and

bargain a rate of return accordingly.

5. Finally, he argues that it is unnecessary to regulate

markets to protect simple people from deceit,

because people are not idiots.

All of these arguments are still current, particularly the

third and fifth and the underlying assumption that

competition between lenders would give borrowers

access to funds at reasonable rates. Even in the 19th

Century there was much debate. The tide of

utilitarianism rose slowly, and a lengthy campaign was

necessary before the financial deregulation of 1854,

which abolished the British interest rate cap. However,

one act of deregulation cannot quell an argument that

has been going on for millennia. Over the following

century the tide gradually turned towards re-regulation,

culminating with detailed requirements imposed on

the financial sector (particularly the banks) during and

immediately after the Second World War. We now

trace the gradual lead-up to this second phase of

regulation.

The financial deregulation of 1854 was a triumph for

free-market economics. In the following decades the

trend was towards re-regulation – gradually at first,

then with considerable force as the war economy of

1939-45 was converted to an economy managed for

full employment. We consider first the 19th Century

intellectual background, then the increase in concern

over fringe lending, and finally the rise of Keynesian

economics.

Though the interest rate deregulation of 1854 settled

the practical question of whether interest rates were to

be market-determined for the greater part of a century,

it did not settle the more academic question of the

legitimacy of interest. The late 19th Century was the

heyday of the labour theory of value. This theory

extended the work of David Ricardo, who had

concentrated on the rent of land. His argument had

had two simple steps.

• Land is scarce, and hence must be rationed.

Landowners therefore receive rents.

• However, the landowners did not produce the land

and do not have to take any positive action to

ensure that it continues to provide its services.

Therefore they do not deserve their rents.

Marx extended these propositions by assigning the

value of all production to labour. Payments to

capitalists, however much they might reflect scarcity of

capital, were accordingly undeserved. The Austrian

school of economists, Bohm Bawerk prominent

among them and Hayek their successor, defended

capitalist practice by developing the theory that

interest was a reward for waiting for the superior

productivity of roundabout methods of production; in

short, a reward for saving. Marshall was at pains to

distinguish the undeserved rent of land from the

deserved quasi-rent of the capitalist (Marshall 1949

p353). In Australia, the theory of the undeserved rent

of land was influential in the introduction of the

progressive land tax, but the legitimacy of interest was

never seriously challenged.

Marshall’s justification of the legitimacy of interest

applied only to savings that were productively

invested. Like so much else in liberal economics, the

coherence of this theory depended on demand and

supply. The supply side rested with households, which

would increase their savings if suitably rewarded with

interest payments. The demand side rested with

business, which would increase its investment in

productive capital if borrowing costs were low enough.

To quote Marshall again: ‘Thus then interest, being the

price paid for the use of capital in any market, tends

towards an equilibrium level such that the aggregate

demand for capital in that market, at that rate of

interest, is equal to the aggregate stock forthcoming at

that rate.’ (p443)

But what if social returns diverge from financial

returns? A classic case here was the railway

investments of the colony of Victoria. The colonial

railways were barely profitable financially, but opened

up the land for farming, so making possible an

increase in colonial incomes that amply compensated

for the poor financial returns of the railways

themselves. Cases like this continue to appear, and the

World Bank, AusAID and other development

financiers regularly assess them when distributing soft

loans and grants. They provide loans at less than

commercial interest rates to projects assessed as having

social rates of return in excess of their expected

financial returns.

In the 19th Century, as now, one of the major

complaints against the financial system was that of

small businesspeople short of working capital. By the

end of the Century all the Australian colonies had

founded government banks with the aim of filling two

gaps in the commercial market: providing an outlet for

the small savings of the working class, and providing

loans to small business, particularly farmers (Butlin

1961 Ch 12). In the 20th Century these savings banks

extended their loans to the support of home

ownership. In both cases the state banks provided

loans at lower interest rates than the commercial

banks. Here was a case of governments pursuing social

policies by directing flows of finance, rather than

leaving the business of lending to the market.

Effectively, people borrowing for approved purposes

had access to capped loans, though loans in general

were not capped.

It should be noted that interest rate caps for approved

purposes must be complemented by regulations which

ensure that funds are made available, and not

transferred to uncapped lending, where profits are

presumably greater. The government savings banks of

the first half of the 20th Century did this by tapping a

distinctive source of funds – small household savings.

These banks minimised borrowing costs by offering

pass-book accounts with limited transactions services

(no cheques) and low interest returns. An alternative

method of ensuring that funds are provided for

preferential, capped lending is simple command and

control. Japan, Taiwan and South Korea are wellknown

for their application of this technique to

industry finance, but it was also used in post-war

Australia.

With the encouragement of borrowing to finance

home purchase, it became normal for Australian

households to be in debt during the home-buying

phase of the life course. This provided a precedent for

mass consumer borrowing. Home purchase was not a

business investment of the kind envisaged by Alfred

Marshall when he justified borrowing to finance

investment, but it had similarities. A house is an asset

that can be used to secure a loan, and home

ownership saves on rent and has tax advantages and

therefore yields a cash flow that can be used to service

the loan – all this, in addition to advantages like

security of tenure and freedom from landlord

requirements. No wonder home purchase was a

popular justification for borrowing. Even so, the switch

to home-buying financed by mortgage breached a

psychological barrier. It was no longer necessary for

consumers to save up before they bought.

Even further removed from the world of commercial

banking than the small businesses and aspiring home

owners served by the government savings banks was

the experience of the poor. Marshall, quoted above,

implies that, by the beginning of the 20th Century,

economic growth had lifted the poor of England out

of the clutches of the moneylenders, but in this he was

unduly optimistic. Social welfare agencies reported

quite the reverse, raising the question as to the

responsibilities of the state in circumstances not

contemplated by Bentham. What should be done

when neither borrowers nor lenders were gentlemen?

Bankruptcy was an expensive procedure involving

state administration of the bankrupt estate, not suited

to the relief of poor debtors with negligible assets. A

new generation of social reformers argued that the

government should control moneylenders who were

exploiting the poor.

In Victoria, the control of moneylending began with

provisions allowing the courts to re-open loan

contracts that they considered harsh and

unconscionable. The Money Lenders Act of 1906 was

modelled on the English Money Lenders Act of 1900.

The Victorian Act excluded lending by banks and

pawnbrokers, the former because they had become a

Federal matter under the Constitution, and the latter

because their potential connection with criminal

second-hand markets was thought to require separate

legislation.

Introducing the Bill that became the 1906 Act, the

Victorian Attorney General recalled that the British

legal tradition had included Usury Acts in the past, but

that attempts to cap interest rates had been

abandoned. This had allowed moneylenders to exploit

ignorant and desperate borrowers, and there was need

to provide such borrowers with means of redress.

However, the circumstances of borrowing differed so

widely for different loans that no single rate cap was

considered appropriate. Like Britain, Victoria put its

faith in the willingness of borrowers to request review

of their loans in the courts, and the ability of the

courts to recognise and disallow harsh or

unconscionable contracts. Under the 1906 Act relief

was only available where money had been lent at 12

per cent and above, thus indicating the lower bound

of unconscionability (the prime rate at the time was

around 4 or 5 per cent). By amendment during the

debate on the Bill, the 12 per cent was to be calculated

including all fees that were part of the contract,

defined as payments apart from the repayment of

principal.

In England, the 1900 Act was found to be ineffective,

partly because the courts had difficulty in recognising

harsh or unconscionable loans. In 1927 the English

Money Lenders Act was revised to include specific

mention of 48 per cent per annum as a rate above

which the onus of proof would fall on the

moneylender to show that the rate was not harsh or

unconscionable. Below 48 per cent the onus of proof

would fall on the borrower. In the House of Commons

debate there was some discussion of how the 48 per

cent should be calculated, and it would appear that

fees were excluded, but the Act prohibited the

charging of fees for preliminary expenses. Both the

proponents and opponents of the Bill professed to

have the interests of the poor at heart. Opponents

predicted that the cap at 48 per cent would drive out

of business the ‘decent moneylender’ of small

amounts over short periods, and argued that

moneylenders’ mark-ups were not excessive compared

with those charged in other forms of small trading

business. The withdrawal of decent moneylenders

from the small-amount, short-term market would force

poor borrowers into the hands of illegal moneylenders,

who used standover tactics rather than legal processes

for the recovery of debts. In contrast, the proponents

of the Bill were on the side of the angels. Money

lending interest rates above 48 per cent was branded as

an iniquitous imposition on the downtrodden of the

slums, practised by men whose faces were those of the

devil incarnate. The last speaker in the debate made

the intriguing comment that ‘the evils of money

lending reflect the extremes of wealth and poverty in

our midst’.

Despite its propensity to follow developments in

Britain, Victoria did not update its Money Lenders Act

till 1938. The provision that the courts could re-open

contracts with harsh and unconscionable interest rates

remained, with a general indication that such rates

should be calculated including all payments in excess

of principal, including fees. However, this was not very

precisely expressed. The legislators discussed the merits

of the English specification of 48 per cent, but did not

include it in the Act. Instead, they provided that

maximum interest rates could be set by regulation,

their apparent intention being that there would be a

schedule of maximum rates appropriate for different

classes of loan. Critics of the Bill who argued that this

would not work were proved right – it apparently

proved impossible to draft appropriate regulations, so

the English cap of 48 per cent was inserted into the

Victorian Money Lenders Act by amendment in 1941.

There it has stayed ever since.

Meanwhile, market determination of interest rates

came under popular attack. In Victoria, the reputation

of the finance sector, to say nothing of the selfconfidence

of the sector itself, was undermined by the

depression of the 1890s, and a further, global

undermining occurred as a result of the depression of

the 1930s. Bad behaviour by the banks was held

responsible, and large sections of the public sought

vengeance.

An academic justification for re-regulation of the

financial system, and in particular the banks, was

provided by the development of Keynesian

macroeconomics. The exact content of the Keynesian

revolution is still debated, but one element was a sharp

break with Marshall’s theory of interest (Keynes 1936,

chapter on liquidity preference). Instead of envisaging

the rate of interest as determined by equilibrium

between the supply of savings and the demand for

new capital, Keynes incorporated into his system an

interest rate determined by the supply and demand for

money as against other financial assets. In this he was

merely formalising established Bank of England

practice – the Bank already influenced market rates by

its dealings in the money market. The difference was

that the rate of interest was no longer conceptualised

as an essential return to Thrift, but as a

macroeconomic control variable to be manipulated in

the interests of full employment.

The adoption of Keynesian macroeconomic

management after the Second World War involved the

Commonwealth capping both the borrowing and

lending rates of interest charged by the trading and

savings banks. Fees were not capped, but by 1990s

standards remained low, perhaps because of the

conservative culture of the banks but also because the

proliferation of fees had to wait for the invention of

electronic account-keeping. Lending rates were capped

so stringently that very little lending took place below

the caps. Borrowing rates were capped so as to allow

the banks a conventional profit margin. All of this was

done in the interests of macroeconomic control, and

was unrelated to the older legal tradition that

disallowed harsh and unconscionable contracts.

In 1950 the overdraft rate was capped at 4.5 per cent

with the actual average rate around 4.25 per cent. Real

interest rates were negative, due to the burst of

inflation associated with the Korean War, but they

became positive as inflation fell during the 1950s. Both

the borrowing and the lending rates gradually drifted

upwards, as did the maximum rate for housing loans

from savings banks, reaching 8.25 per cent in 1970,

with actual average rates slightly below. With the

Consumer Price Index at around 4 per cent, real

lending rates were positive (Norton, Garmston and

Brodie 1982).

Nominal interest rates lagged the breakout in inflation

in the 1970s, and for most of the decade real rates

were negative. Even in 1980 the capped lending rate of

10.5 per cent was barely positive given the inflation

rate that year. Understandably, there was a gradual

increase in bank loans exempt from the cap. Because

of these exemptions, the actual average loan rate was

above the cap for many years during the 1970s. There

was an even more rapid increase in uncapped loans

through non-bank financial intermediaries, the chief

of which were bank-owned. The home mortgage

lending rates charged by the non-bank finance

companies in the 1950s were not much above the

savings banks, but by the 1960s were diverging by 4

percentage points or so – i.e. around 9 per cent

compared with 5 per cent. In 1980 they averaged 13.4

per cent compared with regulated caps of 10.5 per

cent, and an inflation rate of 10.2 per cent.

The consistently higher returns on non-bank financial

assets resulted in the non-bank sector growing faster

than the banks. In the two decades to 1973 bank assets

contracted in real terms, and non-bank assets grew.

This did not seriously affect the profitability of the

banks as corporate entities – after all, they owned

many of the non-banks. However, bank loans at

capped rates were rationed with increasing severity.

The banks required top-notch collateral and

conservative valuation ratios. Even then, loans were

often restricted to people with a record of prior saving

with the bank. As a result, housing finance came to

depend on a mixture of first and second mortgages

(the first with the bank, the second with its non-bank

subsidiary) and other consumer finance became largely

the province of non-bank finance companies,

operating under state rather than Commonwealth

regulation. The states did not regard themselves as

macroeconomic managers, and made no attempt to

extend Commonwealth monetary policy to the

financial intermediaries under their control.

Given two decades of co-existence between a tightly

regulated sector and an essentially unregulated sector,

one may ask why the transfer of assets to the

unregulated sector was not faster, and why the

divergence of interest rates was fairly moderate. Part of

the answer would be that the banking sector has the

privilege of credit creation, and this advantage remains

even when its regulator insists on reining in the

growth of credit in the interests of monetary policy.

Again, the regulator allowed interest rates to creep

upwards. Allowing for risk and for other benefits to

depositors (chiefly liquidity and ready transfer), the

competitiveness gap between the banks and the nonbanks

was not as wide as their relative interest rates

would indicate.

On the borrowing side, the non-banks gradually

widened their range of consumer loans from financing

home purchase through mortgages to hire purchase of

vehicles and other consumer durables that could, in

theory, be repossessed, to straight personal loans

without collateral other than the consumer’s word and

income-earning prospects. This reflected a gradual

relaxation in social judgement from Marshall’s defence

of borrowing restricted to the finance of productive

investment to something more akin to Bentham’s

defence of the prodigal borrower. The social sanctions

against debt diminished as it became less fashionable

to judge the decisions of others.

Despite the growth of the non-bank financial

intermediaries, pawnbrokers and traditional

moneylenders fell on hard times during the 1950s, and

many ceased business. The reasons include the

following.

• Banks were discouraged from lending to

moneylenders, who accordingly had difficulty in

raising finance for their operations.

• Demand for high-risk loans fell due to the general

prosperity and high level of employment.

• In Victoria if not in other states the cap on interest

rates discouraged moneylending.

• Moneylending and borrowing from moneylenders

were both stigmatised in most sectors of society.

An indicator of the unimportance of cash

money lending was the social welfare agencies’ lack of

concern about cash loans. In evidence to the

Henderson Poverty Inquiry of 1973-5 they disregarded

moneylenders, but argued for stricter regulation of

retailer lending. One or two notorious retailers

specialised in lending small sums to high-risk

borrowers, tied to the purchase of the retailer’s goods.

Despite the lack of availability of short-term high-risk

cash loans, the social welfare agencies did not report

that there was any major resort to illegal loans, apart

from transactions associated with illegal gambling.

The deregulation of interest rates during the 1980s was

part of the government response to stagflation. Much

has been written about why the Australian and other

English-speaking governments responded in this way,

and the reasons will doubtless be re-evaluated by

historians for centuries yet. However, it is important to

note that the deregulation of the 1980s was contested

to a degree that the deregulation of 1854 was not, and

that many of the control instruments erected during

the era of re-regulation are still with us, even though

some of them are dormant like the Commonwealth

legislation allowing for quantitative controls of nonbank credit.

During the 1960s the Commonwealth viewed the

gradual contraction of the banks vis a vis the nonbank

intermediaries with concern for the effectiveness

of its monetary policies. The view that the management

of aggregate demand required further control of the

non-bank financial intermediaries gained ground, and

resulted in the Financial Corporations Act (Cth) of

1974, which obliged financial corporations registered

under state or territory legislation to supply

information on their operations to the Reserve Bank

and to the Commonwealth statistician. The Act also

provided that, in respect of financial corporations with

total assets exceeding five million dollars, the Reserve

Bank could promulgate regulations on asset ratios, the

volume and direction of lending, and maximum interest

rates. As it turned out, these powers have not been used.

The breakout of stagflation in the mid-1970s began a

remarkable reversal of policy (Hughes 1980). The

Australian stagflation had both domestic and overseas

causes. The major domestic cause was a breakout in

wage inflation, due to a breakdown in union restraint.

The imported cause was the OPEC inflation.

Theoretically the imported inflation could have been

denied entry by raising the exchange rate, but in the

post-war period exchange rates had been as far as

possible fixed, to encourage trade and investment.

Maintenance of this fixed exchange rate while export

prices increased rapidly meant that bank assets

increased faster than non-bank financial assets for the

first time since the beginning of bank interest rate

controls.

The Commonwealth’s first, monetarist, reaction to the inflation was to tighten monetary policy. However, it also began the decontrol of bank interest rates and allowed the issue of the country’s first credit cards, which bore borrowing rates well above the controlled rates for other bank loans. Full control of the money supply implied the extension of controls to non-bank intermediaries, but no attempt was made to use the power to regulate awarded to the Reserve Bank in 1974. In 1984, the waning of monetarism and the revival of neoclassical economic theory led to full decontrol of bank interest rates. Financial innovation made it difficult to define, let alone control, monetary aggregates, and the Reserve Bank withdrew its monetary targets a few years later. In effect, the banks and non-banks were put on an equal footing by deregulating the banks, rather than by extending regulation as provided in the 1974 legislation. These changes had the expected effect of allowing the banks to fold their non-bank subsidiaries back into themselves. They also permitted the banks and some building societies to engage in a disastrous adventure in the financing of speculative entrepreneurs, leading to financial breakdown and a recession in the early 1990s. The Commonwealth’s considered reaction to this came in 1998, when prudential controls under the Banking Act were extended to all deposit-taking institutions.

The neoclassical economics that came to provide the basis of Commonwealth policy is agnostic as to the purposes of national economic life, but adamant on the importance of competition as a means of reconciling the pursuits of individuals. The Commonwealth’s Keynesian appetite for direct credit controls was therefore replaced by a priority for the promotion of competition. Since 1998 the Reserve Bank has had a responsibility to promote competition in the payments system, while since 2003 the Australian Securities and Investments Commission has implemented the approach throughout the financial sector (Lanyon 2004). A principle of competition is to welcome new entrants to the market. In principle, therefore, moneylenders were once again welcome.

They were now seen as adding to the range of competitive financial services – a far cry from their pariah status in the post-war period.

To speed recovery from the recession the Commonwealth Treasury and the Reserve Bank encouraged a growth in consumer indebtedness in

tandem with a policy of reducing government

indebtedness (Brain 1999 Ch 9). Technically speaking,

they have pursued a permissive monetary policy. So

long as the inflation rate was satisfactorily low, the

financial sector was allowed to increase its outstanding

loans. The hope was that the flow of funds which

governments had released by their abstention from

borrowing would be absorbed by private business

anxious to invest in Australia’s future (in the phrases of

the time, governments could not pick winners and

should not crowd out the private sector), but the

reality was that banks found it more profitable to

increase their loans to households, particularly for

house purchase but also for general consumption. The

easy availability of loans for the purchase of dwellings

resulted in a boom in house prices, or more accurately

in suburban land prices, since increases in productivity

in construction meant that the cost of the dwellings

themselves did not increase. This in turn brought

capital gains to suburban households, who became all

the more willing to borrow. As a result of the capital

gains, the aggregate balance sheet of the Australian

household sector does not appear to be seriously

loaded with debt, but the debt-service ratio (interest

and loan repayments in relation to income) is now at

historically unprecedented levels (Manning 2004).

It has been said that Australian macroeconomic

performance during the decade from 1993 was

brilliant (Argy 2003 Ch 2). GDP has grown despite a

drought that depressed rural production and external

circumstances that were not always favourable – one

may instance the Asian financial meltdown. However,

other commentators fear the consequences of the land

boom, especially the increase in both household and

overseas indebtedness.

The debt boom reversed the factors that led to the

post-war withering of moneylending. On the supply

side, with the abandonment of quantitative controls

over the finance sector, finance is again available for

fringe moneylending, pawnbroking and the like. On

the demand side, several factors have been important.

As a direct consequence of the increase in consumer

indebtedness, there has been an increase in the

number of would-be borrowers whose bad credit

records require them to borrow from fringe lenders or

not at all. High indebtedness also increases the

chances that a household will resort to a fringe lender

for additional credit, having exhausted its creditworthiness

at the bank. Various psychological factors

are also likely to contribute. It may not be far from the

truth to argue that debt is fashionable. It is widely

advertised. The Commonwealth government seems to

approve of it. A decade of capital gains has made

households feel that it is safe to be in debt – even that

it is desirable to be in debt to harvest tax-winnings

from negative gearing. They may also reason that, if

the worst comes to the worst, the accumulating

National Superannuation lump sum provides a

backstop that can be used to repay debt, even if it

means retiring on the age pension.

More speculatively, it has been claimed that the

increase in demand for fringe credit reflects a decrease

in employment security (Wilson 2002). This claim is

supported by the increase in the proportion of the

population with casual jobs and/or jobs with variable

hours. However, it is not so strongly supported by

evidence of increased inequality in the earnings

distribution or of reduced typical job tenure. What is

certain, however, is that with the increase in average

indebtedness, household vulnerability to fluctuations

in income is now much more widespread than it was

when consumer credit regulation was last reviewed a

decade ago, and similarly households have become

more vulnerable to increases in interest rates. Australia

is in uncharted territory: consumer indebtedness has

never before been as high in relation to income.

Taking the sweep of Western history from the Middle

Ages to the present, there has been an increase in the

respectability of borrowing. In Shakespeare’s time it

was frowned on, though the upper classes might

indulge. In the 19th Century borrowing for business

purposes became thoroughly respectable, but

economics gave no justification for borrowing for

consumption. Such borrowing was the last resort of

the desperate poor, and the academic defenders of

capitalism preferred to believe that there weren’t any

such people. By contrast, consumer borrowing has

now become so respectable that it formed the

centrepiece of Commonwealth macroeconomic

strategy in the recovery from the 1990 recession. The

resulting high level of household indebtedness,

coupled with an individualistic refusal to judge the

motive for borrowing, has underpinned the revival of

fringe moneylending.

We do not know how long this revival will continue.

A business-as-usual projection in which household

debt continues to accumulate, and with it resort to

fringe moneylenders, is difficult to believe, since it

would exhaust even the current elastic tests of

creditworthiness. At the opposite extreme, there are

those who predict financial disaster. Victoria still has

the monuments if not the memory of the land boom

of the 1880s and the depression that followed it. More

recently, Japan has experienced little economic growth

since the conclusion of its land boom fifteen years ago.

Were the current high level of household indebtedness

to precipitate a financial crisis such as happened in

Victoria in 1890 or in Japan in 1990, widespread

default would result in major restructuring of the

financial system. The fate of the fringe moneylenders

in such a restructuring is hard to predict.

More optimistically, a ‘soft landing’ scenario is

currently popular, in which resources released by a

revival in household saving are switched to the

construction of infrastructure and the rectification of

environmental damage. In this scenario household

debt remains high but under control, and fringe

lenders prosper.

In addition to the gradually increasing respectability of

borrowing to finance consumption, the historic record

also makes clear the difficulty of reconciling

individualism and freedom. At its most stringent,

respect for creditor’s rights can result in individual

borrowers becoming debt-slaves. Since the 19th

Century there have been safeguards against this, the

chief being bankruptcy, but there are many other

arrangements for the management of overindebtedness

so that it does not become debt-slavery,

such as requirements that lenders should make

minimum inquiries as to the credit-worthiness of the

borrower before the state will enforce the contract.

Interest rate caps may have a place in this context. The

higher the rate, the more likely the borrower will not

be able to pay, therefore prohibiting loans at high rates

helps to prohibit loans with a high probability of

causing over-indebtedness. However, as Bentham

pointed out, this is a blunt instrument. Many loans at

high interest rates are indeed repaid and are, prima

facie at least, to the mutual benefit of both borrower

and lender; again, loans at low interest rates

sometimes become onerous, usually due to

unexpected changes in the borrower’s household

circumstances. Caps on interest and fees cannot

substitute for bankruptcy and allied provisions.

A second strand of argument that was important when

caps were reinstituted following the deregulation of

1854 concerned the relative bargaining power of

lender and borrower. Bentham was at his weakest at

this point. It is not enough just to assert that

borrowers can’t be idiots. During the second

deregulation of the 1980s governments were

concerned to equalise bargaining power by promoting

competition between lenders. They licensed additional

banks and insisted on uniform description of loan

‘products’ to facilitate comparison. However, there is

always a risk in competition between financiers, which

is that they will seek market share or short-term profit

at the expense of asset quality. Faced with failing

banks, regulators have usually taken the anticompetitive

route of amalgamating the sick bank with

another stronger one. The result is that Australia has

four major banks, and competition in banking is

oligopolistic rather than perfect (for comparison with

South Africa, see Task Group 2004). The same might

be said for fringe moneylending, where competition

tends to be limited at the neighbourhood level –

though telephone and internet-based operators are

having some effect.

The mere presence of oligopoly is hardly a sufficient

argument for price controls – after all, a high

proportion of Australian consumer purchases are from

oligopolists, starting with the two great retail chains.

However, the welfare agencies point to the existence of

two types of borrower who may require the protection

of interest rate caps. One is the eternal optimist who

seriously underestimates risks – in Bentham’s terms,

the profligate. The other is the truly desperate modernday

equivalent of the 19th Century and depression-era

indebted poor. Ramsay disputes this dichotomy,

arguing that all consumers have their moments of

qualified rationality; their impulses and moments of

over-optimism (Ramsay 2004). A cap can have the

effect of denying credit to the over-optimist, and may

reduce the need for resort to stronger remedies such as

bankruptcy. Financial optimists merge into people

who borrow out of desperation. In their case, whatever

temporary relief borrowing may offer, it is no solution

for poverty unless the proceeds are invested in income

generation. A cap that effectively denies loans to the

desperate consumer borrower may assist in driving

home this fact. (One might add that welfare agencies

administering low-interest or no-interest loans restrict

their lending to occasions where income is likely to

increase as a result, either directly through investment

in small business, or indirectly, as where a refrigerator

reduces time spent in food preparation and so releases

time for paid work.)

Where the concern is potential profiteering by lenders,

there is a strong argument for caps set to reflect lending

costs. To date there has been little effort to do this, but

South Africa is contemplating a cap which differentiates

between loan establishment costs (incurred at the

commencement of the loan), loan administration

costs (incurred periodically through the life of the

loan) and costs related to loan amount (the cost of

funds and the risk premium) (Credit Law Review 2003).

The question of the relationship between interest rates

and loan purpose is little discussed at present, but

shows signs of revival (Stretton 2005). In strong

contrast to policy two generations ago, is not currently

fashionable for governments to favour one loan

purpose over another. However, there are still

glimmers of purpose-related controls, for example limitations on access to credit card debt in gambling venues. Should current levels of consumer

indebtedness be judged excessive, the pendulum may

again swing, and caps be seen as part of credit-direction

policy aimed at encouraging household savings. Thus

there may be a cap deliberately set to deny credit for

risky consumer borrowing – complemented by

encouragement of high-risk venture capital for

business, and perhaps by lower caps and reserved flows

of funds for such worthy items as home purchase.

Our review of the arguments for and against the

regulation of credit may be summarised as follows.

At one extreme lies the old religious command that

lending for consumption purposes should be

prohibited; at the other the libertarian view that

whatever contracts are entered into between

gentlemen should be legal and enforced. A milder

version of complete prohibition allows lending

provided rates are not usurious – but this view of itself

cannot serve to define a particular rate which is the

limit of acceptability. A milder version of the voluntary

agreement argument is that enforcement should be

restricted to agreements that meet standards of

procedural fairness and do not threaten debt-slavery,

but that beyond this point controls infringe liberty.

Controls on the cost of credit have been advocated to

ensure that profitability is limited to acceptable levels,

and in particular that lenders do not exploit borrowers.

This argument reflects a view that competition is

inadequate to control costs and profits in consumer

lending, particularly in lending to marginal borrowers

who lack collateral and whose credit-worthiness

cannot be quickly assessed from a checklist. Counter

arguments have been mounted from two points of

view. One is to assert that competition is effective so

that there is no need for controls. Alternatively, it may

be admitted that competition is weak, but argued that

it can be fostered to the point where controls are not

necessary. An allied argument is that controls are

counter-productive, increasing the costs of borrowing

by marginal borrowers by limiting competition or

imposing extra costs on lenders.

A purpose of controls can be to prevent lending which

is likely, because of the high rates charged, to push the

borrower into a downwards debt spiral. Against this, it

is claimed that bankruptcy is available to arrest such

spirals, and that it is unfair to deny credit. A second

form of credit denial is related to the purpose of

borrowing – for example, low caps may be imposed on

consumer borrowing to discourage lenders from

making consumer loans – and is likely to be integrated

into economic development policy. The counter

argument is that governments should not deprive

citizens of the liberty of borrowing by judging purpose.

The mechanisms currently used in Australia to regulate

the cost of credit include fostering competition,

regulating the circumstances under which loans can be

recovered so that excessively costly loans are not

recoverable, and direct caps.

We have already noted that fostering competition

between lenders is not always easy, since it can result

in the relaxation of prudential standards. Secondly,

financial ‘products’ can often be varied to avoid

regulation. It is not always easy to assist consumers to

compare branded, differentiated loan contracts, nor to

ensure that consumers have the mobility between

providers that is required if high-cost lenders are to

lose custom and so be forced to lower their prices.

It has not been hard to legislate against loan recovery

when the borrower has no reasonable prospect of

being able to repay: bankruptcy looks after this.

Should it be desired, it may not be impossible to

legislate to make loans non-recoverable when the

lender has failed to carry out obligatory steps in the

assessment of credit-worthiness – the South African

‘reckless lending’ proposal is a step in this direction.

However, it is difficult to frame enforceable regulations

to make loans non-recoverable on the ground of

excessive interest charges. This is the history of the

‘unconscionability’ provision – courts can determine if

the procedures leading up to the grant of a loan were

defective, but in the absence of specific numbers (a

cap) they have not been able to determine whether or

not a given interest rate is unconscionable. This

indeterminacy lay behind the institution of specific

interest rate caps in the UK, Victoria and NSW.

The Victorian cap has been in place since 1941 at the

levels of 30 per cent for secured and 48 per cent for

unsecured lending. The meaning of ‘interest’ has

varied, at some stages including, and currently

excluding, fees. The coverage has also varied, with

large loans originally excluded, and with short-term

(less than 62-day) loans previously excluded. A

legislated cap of 48 per cent applies in the A.C.T. These

caps rank as additional state-specific provisions that

supplement the Uniform Consumer Credit Code,

which does not contain caps.

In recent years, public debate over interest rate caps

has been most lively in NSW – not unexpectedly, since

the land boom has resulted in greater household

indebtedness in Sydney than in the other major cities.

We have already described the history of legislation

regulating moneylenders in Victoria, where the cap of

48 per cent has been imposed since 1941. In the same

year NSW enacted its Moneylenders and Infant Loans

Act, which provided that the courts could re-open

credit contracts that were harsh or unconscionable,

particularly if they considered the interest rate or the

amounts charged for expenses, etc., to be excessive.

Unlike Victoria, NSW did not impose a cap.

The Moneylenders Act and other Acts passed during

the post-war period dealing with hire-purchase were

superseded in 1981 by a new Credit Act, which was

further updated in 1984. Consultation with Victoria

raised the question of an interest rate cap, one

suggestion being that moneylenders who lent at

more than 50 per cent should be deregistered.

The proponents of a cap were only partially successful,

for the 1984 Act provided that the Commercial

Tribunal of NSW could set a maximum interest rate,

but did not have to. It preferred not to, and discussion

of the merits of interest rate caps continued against the

background of very high commercial interest rates

resulting from the Commonwealth’s then very tight

monetary policy.

In 1991 the government asked the Commercial

Tribunal to investigate the costs of consumer lending.

The Tribunal reported (1992) that high-interest lending

was mostly carried out by small businesses, many of

which had low overheads – some did not have offices,

but operated from the owner’s home. The clientele

comprised people with low incomes, requiring small

loans of less than $2000. Such loans were not available

from the banks, due to the high administration costs

of each little loan. The Tribunal judged that the small

moneylenders were exploiting their clients, because

the clients were gullible and there was little

competition between moneylenders – each

moneylender tended to serve a particular suburb. The

Tribunal recommended a cap of 48 per cent, in line

with that in Victoria. Meanwhile the government had

changed, and the Bill to introduce the 48 per cent cap

was introduced by cap-proponents who were now in

Opposition. This, and popular agitation about high

interest rates, prompted the new government to act,

and in 1993 it amended the Credit Act.

To differentiate itself from the Opposition, the

government set a cap that would be relative to the

prevailing level of interest rates, rather than an

unchanging percentage. Its 1993 amending Act set a

cap equal to four times the interest rate prescribed

under the Supreme Court Act, a rate which in turn was

subject to six-monthly review. At the time this rate was

10.5 per cent, yielding a cap of 42 per cent. The

amending Act provided that, for loans of less than

$2000 where there had been no previous credit

relationship between the moneylender and the client,

the cap could be exceeded by seven percentage points

(i.e. 49 per cent) or by $35, whichever was the larger. It

was argued that that these exemptions reflected Credit

Union practice, and would be sufficient to cover

establishment costs for small loans. It was intended

that the cap would be calculated to include all fees and

charges, but the legislation did not specify a precise

formula for calculating the cap rate, and it appears that

lenders could still charge fees provided they were

careful in their wording. There was some principled

opposition to the introduction of an interest rate cap

from the mainstream financial sector, but they had to

acknowledge that the caps would not affect them in

practice, and the amendment was passed with

bipartisan support.

The 1993 amendment did not last very long. The 1995

Consumer Credit (NSW) Act provided that the

Uniform Consumer Credit Code, developed for all

Australia, applied in NSW. However, the Act included

additional provisions outside the Code, among which

was an interest rate cap which, reverting to the 1984

approach, was to be set by regulation. A regulation was

duly promulgated precluding lenders from recovering

interest in excess of 48 per cent. As a result, the interest

rate cap for loans of more than 62 days duration is

now similar in both NSW and Victoria, with the courts

having power to re-open unjust contracts even where

the rate of interest is less than 48 per cent.

The 1990s credit boom saw a revival of moneylending

in NSW, as in other states, including the new concept,

imported from the USA, of ‘pay-day loans’ or very

short-term consumer credit. A consequence at the

national level was a review of the Uniform Consumer

Credit Code, which, as originally enacted, did not

apply to short-term credit. Presumably this was

because consumer-protection provisions were not

thought applicable to short-term instruments of

largely transactional nature. The new moneylenders

took advantage of this. In 2000 the State of

Queensland, as keeper of the Uniform Consumer

Credit Code, appointed a Working Party to investigate

the regulation of payday loans. The Working Party

(2000) reported that payday lending was of recent

origin, and was carried out by a small number of

businesses that ran branch or franchise networks and

advertised their services. This would appear to be a

different business model from the small low-overhead

operators whose activities persuaded NSW to impose

an interest rate cap in 1993. The main

recommendation of the Working Party was that the

Uniform Consumer Credit Code should be amended

to bring payday lending under the Code, without

‘unintentionally catching other short-term products

offered by mainstream lenders, such as bridging

finance’. Inclusion under the Code would bring an

obligation to disclose the interest rate charged for each

loan, and the Working Party indicated that regulators

should issue guidelines to ensure that this quoted rate

included all fees and charges.

The Working Party considered the question of whether

there should be an interest rate cap, which would

completely ‘remove any concerns about usury or

exploitation of vulnerable consumers by payday

lenders’. The Working Party limited itself to

considering a single cap rate that would include fees.

Given the minimum costs of loan establishment,

estimated at 45 minutes of clerical time, a cap of 48

per cent would be very tight for short-period loans

while permitting possibly unconscionable interest rates

to be charged for long-period loans. A cap set as low as

48 per cent was effectively a prohibition on short-term

small-amount lending other than by credit card. The

Working Party believed that it would be preferable ‘to

allow the industry to operate in a regulated way rather

than to kill the industry altogether and force

consumers into the jaws of loan sharks’. Accordingly, it

was against interest rate caps, and indeed

recommended that ‘in order to maintain uniformity,

NSW and Victoria should review the 48 per cent

ceilings on interest rates in those States and the effect

this will have on the payday lending market’.

This report found its way to the Ministerial Council on

Consumer Affairs, who adopted its main

recommendation. From December 2001 the Uniform

Consumer Credit Code was amended to include short term

(less than 62-day) loans where the fees charged

are greater than 5 per cent or the interest rate is greater

than 24 per cent. This brought payday lenders under

the requirements of the Code throughout Australia.

The extended coverage of the Uniform Consumer

Credit Code applied in all states and territories

including those with interest rate caps. Despite the

Queensland Working Party’s recommendation that

they reconsider, both Victoria and NSW retained their

caps.. NSW eventually altered its regulations to define

the 48 per cent cap for loans of less than 62 days as an

annual percentage rate including fees, with a detailed

formula which makes it very difficult to keep fees out

of the cap. For longer-duration loans, the detailed

formula did not apply.

These provisions engendered a lively debate in the

NSW Legislative Assembly, with a strong sense of déjà

vu to anybody who had read the equivalent debates

from a century ago. The case for the cap was put, with

much indignation, by people who considered payday

loans clearly usurious due to their high annual interest

rates. The main opposition to the cap came from the

Revd Fred Nile, and independent member who

combined conservative Christianity with knowledge of

the finances of poor people derived from his

experience running a mission in Sydney’s red-light

district. In 1993 Mr Nile had supported the imposition

of the interest rate cap, but in 2003 he argued that its

extension to payday loans would simply drive poor

borrowers into the clutches of criminal loan sharks.

The cap has indeed had an effect on the availability of

payday loans in NSW. Whether or not the criminal

loan sharks are fattening is debateable. However, there

have been complaints that short-term moneylenders

in NSW are trying to survive legally by concentrating

on loans of 63 days duration. This has caused the NSW

government to extend the fee-inclusive cap to all

consumer loans whatever their duration. This new law

is scheduled to commence on 1 March 2006.

The de-regulatory, free market fervour which swirled

out of the USA and through the English-speaking

countries during the 1980s was not uniform in its

effects elsewhere. There are many countries where

the financial system much more closely resembles the

regulated Australian post-war system than the current

deregulated system. Again, not all countries have

experienced a consumer boom of Australian or US

proportions, and Australia is almost alone in the

vulnerability of its indebted consumers. The following

account is based on papers by Schierenbeck (2004) and

the Department of Trade and Industry (S.Africa) (2004).

In Western Europe, with variation between countries,

the level of consumer indebtedness is not generally as

high as in Australia and the distribution of income is

generally more equal. One would therefore expect a

lower demand for fringe credit, but this is not a

variable that is easily measured. Though there are

moves towards the negotiation of a European Union

Directive on consumer credit, the area has remained

one of divergent national regulations. In the UK, the

48 per cent statutory cap that survives in Victoria and

NSW was abandoned in 1974, despite a

recommendation by the Crowther committee that it

should be retained. Among Western European

countries the UK stands with Luxembourg, Portugal,

Sweden and possibly Spain as lacking controls on

consumer credit interest rates. In the remaining

countries the controls take various forms.

• In Ireland there are no ceilings, but controls are

imposed in the course of licensing financial

intermediaries. A moneylender who claims high

operating costs will be refused a licence.

• There may be a flat statutory ceiling, as was once

the case in the UK and is still the case in Greece and

Switzerland.

• Ceilings may be defined by the courts and varied

with market interest rates, as in Austria.

• There may be a range of ceilings, with different

interest rates for particular types of loan.

• Ceilings may be determined periodically by

government, as in Belgium.

• Ceilings may be defined in relation to a reference

rate or rates. In France usury rates are defined by

the Banque de France every quarter at 133 per cent

of the average market rate for each type of loan.

A similar but more complex system applies in Italy.

Germany has a rule of double the market rate – save

that the ceiling is less than double when the market

rate gets into double figures. Finland uses the

European Central Bank reference rate.

• There may be a specific cap on default interest, as in

Austria and Finland.

In the USA the situation is not unlike Australia

in that consumers are heavily indebted. The income

distribution is unequal, and many incomes are

insecure. There is no national cap, but twelve states

have caps. These caps are effective for payday loans,

but not for credit cards, since they can be avoided by

basing the credit card operation in an uncapped state.

Japan has a national cap, currently set at 29 per cent.

However, as dramatised in Miyabe’s novel ‘All She Was

Worth’, there is a substantial illegal lending sector

whose debt recovery practices leave much to be

desired.

South Africa is in the process of replacing a statutory

usury cap set in relation to the prime lending rate.

Within this cap, it was not possible to lend profitably

to small, short-term and high risk borrowers. Small

loans were accordingly exempted from the cap. New

legislation is now before the South African parliament

to introduce caps that will apply to all consumer

credit, with the actual levels being specified by

regulation. It is proposed that the caps should be

structured so as to reflect the most common approach

to loan product costing.

Leaving aside countries where credit is strictly

controlled as part of macroeconomic or economic

development policy, there are two major approaches

to controls on the cost of consumer credit. The first is

based on the concept of a usurious rate of interest that

should not be exceeded for any loan, with the

implication that lenders should either get their costs

down below the cap rate, or if this is not possible

should simply refuse to lend. The second approach is

directed towards preventing lenders from exercising

market power, and is much more sensitive to the

structure of lender costs.

Concerning the first of the two approaches, we have

noted that it is difficult to define a single usurious

interest rate. To get round this difficulty, some

countries using the approach have simply resorted to a

high rate defined as some multiple of the prevailing

rate, often differentiated by loan type. This approach is

most defensible where the main purpose of the

interest rate cap is to discourage over-indebtedness by

preventing lenders from making loans at interest rates

which are likely to lead borrowers into a downwards

debt spiral. However, such interest rate caps imposed

to prevent debt spirals are both blunt and indirect.

They are blunt because the correlation between credit

prices and unrepayable debt is poor: many highinterest

loans can be and are repaid; some low-interest

loans contribute to the impoverishment of the

borrowing household. They are indirect, since the

obvious method of tackling spiralling debt is improved

loan assessment, backed up by limitations to

permissible debt-recovery techniques (so that lenders

bear some of the cost of controlling debt burdens).

Even so, they may have a role as a second-tier

mechanism.

The second approach, directed more towards

preventing lenders from exercising market power, has

much in common with the price caps imposed on

monopoly providers in other industries, such as the

caps on bulk transmission and distribution tariffs in

the National Electricity Market. The intention of such

caps is to allow service providers to earn market

profits, and they are usually accompanied by service

standard requirements to ensure that the providers do

not profit by cutting costs at consumers’ expense. The

applicability of this approach to consumer credit is

contested by those who argue that that credit

provision is not a natural monopoly, and that

competition can be relied on to minimise prices. As we

have seen, the contrary argument is that the consumer

credit market is at best oligopolistic, and that

individual borrowers may find themselves in a

position where the lender has a considerable degree of

monopoly power. It can also be pointed out that,

because lenders are dealing in contracts which are

ultimately enforced by government, and the scope of

which is limited by government, standards of service

are already regulated, much as in a natural monopoly.

Why not, therefore, add a price cap? A fundamental

judgement underlying this paper is that the market

power approach to price caps for consumer credit is

worth exploring, and that it might even yield insight

into the prevention of excess indebtedness.

Following the natural monopoly precedent, the next

step is to carry out a detailed analysis of industry costs

and cost drivers. Though this is the approach proposed

for South Africa, so far as is known it has not been

applied in any other jurisdiction. The following

account of industry cost patterns owes much to work

done for the South African reformers, and ultimately

to the Microfinance Product Costing Tool prepared by

the Product Costing Resource Centre of CGAP (2004).

The discussion relates particularly to the costing of

small consumer loans, and does not cover the costing

of other financial services except in passing.

To begin with the basics, financial intermediaries are

businesses providing two broad classes of service,

transactions services and financing services.

Transactions services facilitate the settlement of debts.

Though cash transfers are the hallmark of the transfer

services provided by financial intermediaries, transfer

services may also involve facilitating the sale and

purchase of non-cash assets. Various kinds of brokers

exist to facilitate transfer of assets such as shares and

businesses, while real estate agents specialise in the

transfer of land and buildings. Businesses that facilitate

the transfer of illiquid assets make considerable use of

cash and near-cash transfers, and may have strong

links with, or even be part of, businesses that provide

cash transfer facilities. In this area there is no absolute

boundary separating financial intermediaries from

other businesses.

Financing services arise as a result of transactions

separated by significant periods of time. These can

involve the intermediary both in acceptance of

liabilities (it receives deposits) or acquisition of assets

(it makes loans).

Transactions and financing service are closely related.

Simple cash transactions are commonly accomplished

by transferring the ownership of deposits. At the other

end of the spectrum, complex business sales involve

financing as well as asset transfer, and are commonly

packaged by merchant banks. Despite these

relationships, it is possible to distinguish businesses

that are almost wholly providers of transaction services

(for example, postal or telegraphic money-order

businesses) and businesses that are almost wholly

providers of financial services (for example, traditional

moneylenders). Our present concern is with financing

services, and more specifically loans, but the

relationship and frequent complementarity with

transactions services should be remembered.

Financial intermediaries with significant financing

business may be identified as institutions whose

balance sheets are heavily weighted with financial

assets and generally also with financial liabilities. They

incur costs in administering these assets and liabilities,

and recoup their costs by charging the parties to

whom they lend more than they reward those parties

who lend to them. The simplest contract for this

purpose is the bill of exchange: the borrower promises

to pay a specific sum on a specific future date, which

the lender discounts into a smaller present sum. Given

the period in question, the discount can easily be

expressed as a rate of interest, but for analytical

purposes it comprises at least three portions.

• Compensation to cover income foregone from

possible alternative uses of funds – if compensation

for risk is separately accounted, this will be the socalled

‘risk free’ interest rate.

• Compensation for risks that the sum will not be

repaid.

• Compensation for administrative and transaction

costs incurred.

In current practice, financial intermediaries recoup

these three types of cost in two ways: by charges

calculated proportionally to principal and time, and by

imposing fees which are precipitated by particular

events such as the signing of a loan contract. All three

types of cost may be recouped either way or by a

combination of the two ways.

An important issue in the design of financial

instruments, and hence in the determination of

charging patterns, is the allocation of risk. Legally

speaking, there are two classes of loan:

• Those where the liability is determined in the

contract as a cash amount or series of such

amounts, relief being legally available to the debtor

only in very restricted circumstances such as

bankruptcy or the contract being found

unconscionable. Lenders bear the risk that contract

will not be fulfilled due to bankruptcy as well as due

to lawbreaking (as when a debtor absconds), but

otherwise all risks (other than inflation risk) are

borne by the borrower.

• Those where the liability is determined in the

contract as a contingent amount, typically

dependent on the profitability of the borrower’s

business ventures. There is an obvious moral hazard

in such loans – the borrower will be tempted to

fritter the loan away, show a loss and not have to

repay – so the contracts generally include rights for

lenders to supervise the use of the loan, rights

which are not usually included in fixed-interest

contracts.

This distinction is not hard and fast. A contingent

element can be introduced into the first type of loan

by insurance contracts, or other provisions limiting the

circumstances in which the debtor has to pay.

Similarly fixed-liability elements can be introduced

into contingent loans, such as limits on the profitshare

to which the creditor is entitled. However, the

basic distinction remains. It is convenient to refer to

loans with a schedule of definite cash payments from

the debtor to the creditor as fixed-interest loans.

Even the simplest of financial intermediaries making

fixed-interest loans – say a moneylender who makes

loans out of his own funds – faces the decision as to

how much cost to recoup from interest and how

much from fees, and within the fee category, what

events should precipitate charges, and how much for

each. Venture capitalists making contingent loans face

the much more daunting task of defining the

circumstances in which they will be entitled to

dividends and repayments.

With balance sheet complexity comes complexity of

decision as to how much to charge for each loan

product and how much to reward each class of

creditor. There is considerable discretion in these

decisions. Intermediaries can raise funds in various

ways, and often have discretion as to what

combination of interest rates, services and fee

discounts to offer for these funds. Similarly,

intermediaries can loan funds in various ways, again

with discretion as to the combinations of interest and

fees. Finally, intermediaries that also provide

transactions services have discretion as to how to

charge for these services. Since financial services

necessarily involve transactions, intermediaries that

impose transaction fees are likely to include these as

part of their cost-recovery for financing services.

On the other hand, it is common for intermediaries

that serve high-income personal customers to provide

free transactions as a quid pro quo for low returns on

deposits. Such free transactions are worth more to the

customers than interest income on which they will be

taxed.

The conventional cost analysis for loans follows the

steps in the loan-making process. Once again we

follow the Product Costing Resource Centre (2004).

The first step in making a loan is to set up a business

capable of doing so. This gives rise to business

establishment costs, which, once sunk, do not vary

with the number or value of loans made – at least until

a major change is required in the scale of the business.

In addition, maintenance of the business will require a

minimum of ongoing annual costs. These sunk and

ongoing costs have to be spread over the loans made

(and any other products), and mean that there is likely

to be an element of economies of scale in financial

intermediation. (This is not the only reason for

economies of scale, for which see below.)

Cost reductions in business overheads have been

sought in two contrary directions. One is to seek to

expand the business without increasing the overheads,

generally by mechanisation; the other is to seek

economies of relationship.

The increased scale approach seems to offer more

scope for cost reductions in transactions processing

than it does in lending, since the loan assessment

process is more difficult to automate while

maintaining any particular standard of risk assessment.

The basis of credit assessment is knowledge of the

probability of a loan being repaid. For large business

loans there are obvious economies of scale in

knowledge of industries and of the business

performance of potential borrowers, not to speak of

economies of scale related to sophistication in risk

management. In consumer lending the amounts are

generally smaller, and large lenders have tended to

adopt a credit scoring approach. This has been much

facilitated by access to credit bureaux.

Whether or not there are economies of scale, total

business overheads are likely to increase with broad

indicators of the size of the business. Both size of

portfolio and events can be used as such indicators,

and overheads do not seem tied to one or the other in

particular. Accordingly the cost analysis provides no

particular rule about whether they should be recovered

from fees or interest.

A second step involves the recruitment of borrowers,

who may also be customers for other products offered

by the intermediary, such as depositors. It reduces

costs later in the process if borrowers can be recruited

in such a way that the intermediary already knows

something about their reliability as debtors.

Recruitment can be fairly simple if a customer base

already exists, as for store-based credit, but except

where customers are personally known to the

storekeeper this method of recruitment says little

about credit-worthiness. Recruitment from established

affinity groups, as by credit unions, can reduce

advertising costs, improve credit assessment and,

through group moral suasion, improve loan recovery

rates.

However it is done, recruitment adds to overheads,

and again there is no particular reason why these costs

should be recovered from fees or interest.

Making loans involves a number of activities. Potential

borrowers may need advice, which may involve the

lender in significant work in client education. An

application form must be made available, completed

and assessed, and if the loan is approved it has to be

disbursed. Most of these activities involve staff time,

including indirect staffing costs: training to ensure that

staff know what to do, and supervision to ensure that

they do it and to guard against corrupt lending.

An important function in loan assessment is refusal of

credit where it is unlikely to be repaid. Examples

where extension of consumer credit is not in the best

interests of either borrower or lender include the

following.

• Where debt is added to a load that is already

difficult or impossible to service.

• Where prospects of repayment are otherwise dim

through lack of uncommitted income (including

people with high spending commitments on

necessities, and those who spend on addictions).

• Where repayment is reasonably certain but is

expected to require sale of collateral essential to the

operation of the borrowing household.

It will be noted that the last of these is a trap where

lending assessment emphasises collateral, while the

other two are traps where lending assessment

emphasises income prospects.

Lending assessment based on collateral is low-cost and

simple when borrowers can offer marketable assets to

which they have secure, non-contested and readilytransferable

title. The obvious instance is mortgage

lending, though even here there is an irreducible

minimum of work required to check the documents.

In the case of mortgages on property in suburbs

affected by the recent land boom, the lender may also

want to discount values in case they fall.

Pawnbroking provides another example. The work

required to estimate the metal content of jewellery is

fairly small provided expert staff are available, but

there are often uncertainties as to whether the

borrower has secure title to the item – pawning is just

too easy as a method of getting rid of stolen goods.

Collateral of any kind thus has assessment costs. As

against these, it has the advantage that seizure is a

possibility in the event of non-repayment of the loan

(or, in the case of pawnbroking, simple non-return of

the pawned item – though here storage costs are

incurred). Mortgages, pawnbroking and other loans on

collateral therefore have considerable establishment

expenses but low risk of default provided the loan is

safely below the sale value after subtraction of recovery

costs. Apart from the difference between pawnbroking

and mortgages, the amount of checking required is

likely to be independent of the size of the loan.

Mortgage establishment costs are such that a mortgage

will not be considered worth arranging for short-term,

low-value loans.

An alternative approach is lending based on income

prospects. This is inherently more risky, in that there is

nothing to seize in the case of default, though

garnishee orders provide a substitute if incomes are

steady and their source is known. In rural

communities, local lenders often have the necessary

knowledge, or can easily check it. The greatest

uncertainties lie in the areas of prior commitments

and excess household costs (such as those caused by

addictions), where local community knowledge is

advantageous but may be used to discriminate

unfairly. Within any given income/occupational

group, the amount of checking required is likely to

increase with the size of the loan, with fairly little

checking being required for small loans to people who

can claim unencumbered status and a regular income.

Financial institutions may attempt to minimise the

costs of loan assessment by standardising conditions

for the grant of loans. Inevitably this omits significant

difficult-to-observe variables, such as whether the

applicant has relations who are likely to siphon off the

loan proceeds. Such variables tend to be common

knowledge in small affinity-based credit unions, and

are likely to form grounds for loan refusal.

The high long-run costs of bad debts give lenders a

strong incentive to engage in rigorous loan assessment.

However, bad lending decisions (not counting the

inevitable occasions when a risk was detected and

turned out badly) occur for several reasons.

• Failure to allocate sufficient, or sufficiently skilled,

resources to the task.

• Failure to appreciate the risks properly – this is

particularly likely to take place during cyclical

booms, when lenders are seeking market share.

• Failure to carry out an arm’s length assessment.

It is an important role of prudential regulators to

recognise these problems when they are incipient and

insist that they be remedied. There is a strong case for

recouping the cost of prudential regulation from the

affected institutions. Such costs add to the costs of

loan assessment.

Whether for large or small enterprises, there are likely

to be a modicum of paper, printing and computer

costs, and maybe also the costs of reference to a credit

bureau.

The more thorough the assessment of an application,

the better the lender’s appreciation of the risks

involved. This appreciation can result in variations in

charges for risk, and should also result in refusal of

loans where it is likely that they can only be repaid

under stress. Thorough assessment is partly a matter of

the resources devoted to it by the lender, but is helped

by backup resources: credit bureaux and other means

by which the lender can gain information about the

potential client’s balance sheet and prospective

income. If collateral is taken, or a claim is entered

against income, it is important that these claims

should not be subjected to disturbance by subsequent

lenders (and should not disturb the claims of prior

lenders). Thorough assessment reduces the risks borne

by the lender, but at a cost: both the cost of the

assessment itself, and the cost of the income potential

of marginal loans rejected.

Lenders sometimes seek to attract custom by offering

rapid assessment. This raises costs by being less

thorough than time-consuming assessment, and also

because extra personnel have to be employed to deal

with the ebb and flow of applications. However, rapid

assessment is of undoubted benefit to people who

need money quickly to deal with unforseen financial

demands. This raises a question for caps: should they

be based on rapid assessment costs, or on regular

costs? The problem here is that the people most likely

to be in need of rapid assessment are also likely to be

those who require careful assessment, at least at the

yes/no level.

It obviously reduces costs if the intermediary can

discourage hopeless applications, or recognise and

refuse them quickly – but again there will be costs, in

that some good applications will be discouraged or

rejected too quickly. It also reduces costs if applicants

know the process, and do not have to be guided

through it. This reduces costs for people who are

applying for the second or subsequent time – an offset

being that too many applications may be an indicator

of poor financial status.

The conventional cost driver suggested for loan

application costs is the number of applications. Other

drivers may include

• Amount applied for – intermediaries generally

devote more time to the assessment of the larger

applications.

• Source of application – assessment may be simpler

and easier for some classes of customer than others.

In co-operatives, members have borrowing rights,

and the costs of loan assessment are transferred to

the approval of membership.

• Repeat applications – repeat customers have a track

record and know the process, and the lender has

already collected information on their financial

affairs.

• Presence of collateral, which needs to be verified

and may occasion legal and inspection costs.

• Guaranteed speed of assessment.

The conventional cost driver approach suggests that

the costs of making loans should be recovered from a

once-off application fee, which may be related to loan

size. However, there is also an argument for recovery

from interest rate differentials, in so far as thorough

assessment reduces the risk of default and is more

costly, because more thorough, for the larger loans.

Intermediaries may also wish to vary application fees

for different classes of customer. Lower fees for repeat

customers are consonant with the cost analysis, but

may conflict with marketing strategies. They also

conflict with the aim of emergency credit, which is to

meet a once-off credit need which is often of short

duration. Higher fees for loans with collateral may be

offset against lower interest rate differentials for risk.

Once a loan is in place, it requires ongoing

administration. Where a borrower meets all payments

on time, this is a simple transaction function, with

attendant accounting costs. The conventional driver

would be the number of transactions, proxied by loans

outstanding – hence a justification for a periodical

service charge. Many intermediaries are likely to

consider this cost so trivial that it can be absorbed into

the interest charge.

The cost of funds lent depends on the liability side of

the lending institution’s balance sheet. It is also

affected by items on the asset side, including lowreturn

assets essential to the conduct of the business

(premises, cash and other liquid asset reserves) and,

where funds are limited, may be affected by rates of

return available in alternative lines of lending.

The cost of liabilities varies by class of liability.

• Deposits may bear interest rates well below the

lending rate, but often occasion additional costs,

such as transaction costs not recouped from fees,

and variability that requires balancing holdings of

cash and liquid assets. The less liquid the deposit,

the higher the rate of interest which must be offered

to compensate the financier’s creditors for reduced

liquidity, but the less the requirement to hold liquid

assets to guarantee its prompt repayment.

• The cost of the medium to long-term borrowings of

a financial intermediary depends on its credit rating.

Intermediaries perceived as making risky loans are

likely to have to pay risk premiums. Lending

policies thus feed back to borrowing costs.

• Financial intermediary liabilities also include equity.

There has been a tendency for prudential regulators

to reduce their insistence on cash holdings as a

guarantee of prompt repayment of deposits (other

liquid assets being recognised as close substitutes)

but to increase their insistence on capital adequacy,

as a guarantee that all creditors will receive their due

in the event of the business being wound up. As in

non-financial businesses, equity bears the residual

risks and is accordingly more expensive than loan

capital.

The cost of funds lent depends not only on the cost of

funds borrowed, but on the costs of low-return assets

maintained for prudential reasons. These costs vary

across financial intermediaries, depending on prudent

assessment of the need for funds of various degrees of

liquidity and in turn on the availability of backup such

as borrowing of last resort and deposit guarantees. In

addition, funds may have to be held in assets bearing

less than desired returns during periods when lending

opportunities are limited. The greater the proportion

of funds which is maintained in low-return assets, the

smaller the asset base available for lending out at

profitable rates, and the higher the lending rate

required to generate profits.

The cost of funds for any particular loan thus

depends on

• costs incurred on the liability side of the balance

sheet

• costs incurred on the asset side of the balance sheet

(low-return assets) and

• cost allocation conventions.

Whatever the cost of funds allocated to any class of

loan, it is conventional practice for the cost to be

expressed as an interest rate; that is, to vary with the

funds committed and the duration of the loan. Cost

differentials for different kinds of loan become mixed

up with risk assessments, and costs of funds may be

quoted which include a risk assessment for the loan

class. This may be fair practice, especially if the

intermediary’s borrowings are at enhanced interest

rates due to the risks perceived in its loan portfolio.

However, there is also a case for making a distinction

between the cost of funds in a cash flow sense (cost of

liabilities plus cost of prudential assets) and risk

premiums that are the result of the intermediary’s

assessments. These in turn may be distinguished from

the cost-plus mark-up.

Delinquent loans absorb staff time chasing the loans

and seizing collateral, not to speak of the costs of

write-offs. These costs provide a cost-based rationale

for delinquency fees, which, however, cannot recover

bad debts, and indeed increase the likelihood that a

debt will go bad.

Delinquency fees are not so open to limitation by

competition as other fees, though consumers may take

them into account when selecting a credit card.

Another possibility for a consumer threatened by a

delinquency fee is to take short-term credit from a

marginal moneylender. If the term of loan required to

avoid delinquency is short, the bridging loan may

come at very high cost in annual percentage rate terms

but still be cheaper than incurring the delinquency fee.

Outside credit cards, the mainstream approach to

delinquent loan costs is to recover them from interest

rate differentials – from the risk premium charged over

and above the cost of funds. In other words, the lender

makes a risk assessment, and adds a risk premium that,

on average, is expected to cover delinquency costs.

This implies that the cost varies with the quantum and

duration of funds on loan, coupled with the type of

loan: loans secured with collateral are less likely to go

completely bad than unsecured loans, but there is a

risk that the revenue from sale of the collateral will be

less than the loan, after allowing for costs. It is arguable

that some delinquent loan costs are independent of

the amount outstanding, in which case they may be

recouped from a periodical service charge.

The opposite of delinquency occurs when loans are

repaid early. Lenders argue that this imposes costs –

the administrative cost of cancelling the loan, and the

opportunity cost of interest foregone while they find a

new borrower. The former cost is brought forward

from the due repayment date, and may be somewhat

increased by being out-of-course. The latter cost can be

minimised and perhaps eliminated if the borrower has

to give due notice of early repayment. There are likely

to be economies of scale, in that lenders with large

portfolios are constantly adjusting their asset patterns,

and the early repayment of a particular loan scarcely

perturbs the balance sheet. In other words, the direct

costs of early repayment are small and perhaps

negligible.

Despite this, a number of lenders impose fees for early

repayment. This strategy is particularly noticeable

where monetary policy is loose, and lenders are

accordingly seeking to maximise loans outstanding,

particularly to credit-worthy borrowers – and what

borrower could be more credit-worthy than one who

seeks to repay early? Accordingly the fee can be

interpreted as a discouragement to repayment rather

than a bona fide recovery of costs.

Early repayment can also occur in the course of debt

re-finance. The comparison here is between fixed and

variable interest rates. A borrower who agrees to a

variable market-related interest rate will make

payments that vary according to the lender’s cost of

funds. A borrower who agrees to a fixed interest rate

benefits if the interest rate rises, but pays more if it

falls. If borrowers can freely re-finance if the interest

rate falls but the rate is capped if it rises, there is an

apparent asymmetry. It can be argued that this justifies

a renegotiation fee greater than that required to recoup

administrative costs. On the other hand, if lenders

offer fixed interest contracts that are in effect inflexible

upwards but flexible downwards, a market price can be

calculated to allow for the asymmetry of risk. There is

nothing intrinsically unfair about this asymmetry.

Policy has been divided on early repayment fees.

They are prohibited in some countries on the grounds

that early repayment without penalty encourages

household saving. These countries accept the

consequence that fixed interest contracts will be

flexible downwards, and will accordingly (if competition

prevails) bear a higher differential vis a vis flexible

interest contracts than they will in countries which

make it difficult for borrowers to renegotiate.

The above cost analysis assumes that the costs of joint

products can and should be wholly allocated to

individual products. However, this may not be

appropriate where a financial intermediary bundles

products. A typical case is the requirement that a

borrower must also have a deposit account, and may

be required to maintain a minimum balance in that

account. The reasons may include establishment of

savings habits, provision of capital for the

intermediary, establishment of a credit record, and

convenience in administering loans. Charges such as

membership fees may relate to more than one

product, and low-earning deposits may be a (serial)

quid pro quo for low-rate loans. Cost analysis may

reveal apparent cross-subsidy which disappears once

allowance is made for packaged products.

From a regulatory point of view, it will be important to

understand the more common of these relationships.

Judgement as to whether they are desirable can then

be explicit.

In most areas of production – manufacturing, retail

and the like – the simplest and safest business pricing

rule is cost-plus, achieved by adding a profit margin to

costs. Provided competitors’ cost structures are similar,

a business that follows this rule will generate normal

profits, more or less. However, in financial

intermediation prices are complex and joint costs are

prevalent, typically covering a range of products with

impacts on both sides of the balance sheet.

Intermediaries have considerable discretion as to the

pattern of fees and interest rates that they adopt, and a

wide variety of patterns has the potential to cover costs

and earn profits. The range is further increased by the

existence of numerous options to manage risk, and

hence a wide variety of cost patterns, not to speak of

residual risk exposures.

As in most markets, in the absence of price controls

the ultimate limit to pricing patterns is imposed by

competition. In any jurisdiction comprising an array

of financial markets, competition will establish

boundaries to pricing in each market. However, the

processes by which it is recognised that products are

mispriced are slow and uncertain. Under-priced

products generally do not come to light until there is a

financial crisis, usually in the course of the trade cycle

– which is currently running at about ten year

intervals on an international basis. Over-priced

products may persist indefinitely if competition is

limited in the markets concerned. The competitive

limits to pricing thus take years to be established, and

even then may be quite broad: hence the recourse to

cost analysis to establish whether products are underor

over-priced, in the sense of under- or over-recovery

of costs.

As in our discussion above, cost analysis allocates costs

to products, and hence indicates the costs to be

recovered if each product is to contribute equally to

profitability. Analysis of loan cost drivers helps in

identifying costs which are related to amounts lent

and to the duration of lending (and hence suited to

recovery from interest) and costs which are related to

events (and hence suited to recovery from fees),

though, as we have seen, there are many costs which

exhibit both relationships. Cost analysis is particularly

helpful in setting interest rates and fees if the

intermediary can be reasonably certain that the cost

analysis that it uses is similar to that used by its

competitors. If this is the case, and after assuming that

its costs are not out of line with its competitors, it can

apply cost-plus pricing rules and expect to be

competitive.

An alternative to the simple cost-plus approach with

joint costs spread more or less evenly across products

(according to the rules developed in the cost analysis)

is that the intermediary ‘cross-subsidises’ – it loads

some products with more than their rule-based share

of joint costs. This is particularly likely in cases where a

firm is operating in several markets with different

degrees of competition and therefore has the

opportunity to price keenly in its competitive markets

and monopolistically in markets where there is less

intense competition. Loss leaders are also common

practice where a firm is endeavouring to enter new

markets.

In an industry with such high joint costs as financial

intermediation, unconventional pricing may also

result when particular firms adopt an unconventional

costing model. The resulting lower prices for at least

some products may attract an unconventional

customer base. Leaving the herd is risky but sometimes

the decision pays off. Just as decisions to introduce

new products, shave profit margins and the like are

fundamental to competition under cost-plus pricing

(which otherwise deteriorates into oligopoly), so

decisions to introduce new products and revise cost

allocations are fundamental to the maintenance of

competition in financial intermediation.

The definition of over-pricing as over-recovery of costs,

resulting in excessive profits, contrasts with the

approach commonly found in the welfare-oriented

literature, which instead regards credit as over-priced

when it is unaffordable to an important social group,

particularly poor people. It is perfectly possible for

credit that barely recovers costs (and hence is neither

under- nor over-priced in relation to costs) to be

considered excessively priced in its impact on the

budgets of low-income households. This impact may

be diagnosed in two ways.

• Credit is an outstanding case where the poor pay

more than the rich. The sense of the injustice of this

can remain even if it is shown that the higher prices

reflect higher costs.

• In jurisdictions where high-cost loans are permitted,

and are taken up by poor people because no other

credit is available to them, it is not usually hard to

find people who are living in penury due to debt

burdens.

Reaction to these impacts can be as follows.

• Where the impact is due to lenders over-recovering

costs, reliance has been placed chiefly on

competition to offer the borrowers a better deal. In

many countries this has emphasised the fostering of

financial institutions aimed at providing low-cost

loans for poor people. This works all the better if the

new institutions have ways to reduce costs.

• Competition has also been promoted by measures

to make it easier for potential borrowers to compare

the prices of loans.

• Loan targeting has been tried, with financial

intermediaries directed to make a certain proportion

of loans to poor people. The intermediaries

naturally resist unless it can be shown that the loans

are profitable.

• There is also a long tradition of capping costs to

borrowers, to be considered below.

• In the last analysis, loan refusal has its place. Unless

they are subsidised as part of a program of

redistribution, financial intermediaries are limited in

the extent to which they can contribute to the

uplift of the poor. They are not set up to make gifts.

Having outlined the theory of financial product

costing, the next step should be practical analysis.

This hits the snag that accurate product costing

information is only available within lending

businesses. Not only is it commercially confidential,

it accords to accounting conventions and policies that

vary from lender to lender. Even with the best will in

the world, data from different firms are not always

comparable. Worse, though regulators can require the

provision of information, the regulated parties who

provide the information have so much at stake in the

regulator’s decision that games of cat and mouse are

inevitable. A quantitative investigation of the costs of

credit, undertaken as part of the South African reform

process, struck all these problems, but even so the data

conformed to expectations, and were considered

sufficiently reliable to form the basis of for credit price

controls. An alternative approach to the data problem

is for the regulator to run his own financial services

business just to gain accurate information – this was

one of the purposes of the state banks of the first half

of the 20th Century.

Despite the uncertainties of detail, the broad outlines

of cost structures are common knowledge, and can be

checked from the specification of products on the

market. If competition is semi-effective, it can be

sufficient to set caps that catch outliers but leave the

bulk of the market unaffected. It is also possible to

design controls that bear down heavily on some

elements of price (for example, early termination fees)

while allowing scope for largely market determination

of other elements (for example, the interest rate

differential over prime rate).

On our analysis, a cost-based system of loan charges

would comprise

• an application fee (charged on all applications)

and/or a loan establishment fee (charged only loans

granted)

• possibly, a periodical service fee,

• an interest rate differential, and

• possibly, delinquency fees.

• Early repayment fees would probably be prohibited.

It would be expected that application or loan

establishment fees would cover part, but not necessarily

all, of the costs associated with application. They might

vary by class of loan, with differences (eg) for secured

and unsecured loans. They might include an insurance

component (in which case it would be reasonable to

allow for additional annual premiums for long-running

loans). It would be expected that the fee would be

higher for classes of loan where collateral is required,

and that, generally, higher application fees would be

associated with lower interest rates. If the fee is regulated,

this should include any charges loaded into the price

of goods and services bought, though this would be

difficult to enforce and might require further thought.

A periodical service fee would cover current

transaction costs and part of loan delinquency costs.

If insurance is offered over a long period, this fee

would also be included.

The interest rate would cover the reasonable cost of

funds (i.e. not cross-subsidising other products offered

by the same intermediary), plus mark-up, plus risk

premium if not covered by the periodical service fee,

and delinquency costs not covered by delinquency

fees if allowed.

Delinquency fees are a moot point. It can be argued

that delinquency fees are appropriate for credit cards

provided to well-off people, who might otherwise

neglect to make regular payments. However,

delinquency fees have in the past been used to tie

people up in debt. Perhaps the general approach

should be prohibition of fees with a capped rate on

delinquent outstandings (as in Austria), with a

discretion to allow fees on products aimed at the wellheeled

market.

We now consider a variety of proposals.

The classic interest-only cap is expressed as a

percentage of the outstanding loan, calculated per

specified time period. This diverges considerable from

an apparently-similar cap expressed in cents per dollar

lent without reference to time. The latter cap is

effectively much higher for short-term loans, and

much lower for long-term.

As our product costing has shown, and Islamic bankers

have demonstrated, conventional interest rate caps are

likely to be no more than cosmetic as controls over the

overall price of credit. The flexibility of product costing

makes it possible to run a profitable lending business

without any resort to interest charges, substituting fees

and other techniques such as asset sale and repurchase.

From the point of view of capping monopoly pricing,

interest-only controls can easily be circumvented.

However, the public readily understands an interestrate

cap. The psychological advantages of an interestonly

cap include assuaging vague public feelings that

interest rates should not be too high, and limiting the

ongoing exposure of debtors, at least in the case where

the lender makes up for the cap with a stiff

establishment fee. (The limit to ongoing exposure is

less certain if the lender also imposes periodic,

termination and delinquency fees.)

Where an interest rate cap is specified, with or without

controls on fees, the costing model requires that the

cap should vary with the cost of funds. In the costing

model, the costs to be recovered from interest

comprise the cost of funds, measured by a risk-free

interest rate, and the various allowances for risk. There

is ample overseas precedent for specifying an interest

rate cap related to a selected low-risk or prime rate, or

perhaps an average of rates. The cap will be well above

this risk-free rate, to allow for the fact that it applies to

risky lending. To guard against changes in market

structure that alter the practical relevance of marker

rates, there is much to be said for inserting a break at

this point, in the form of a suitably august institution

which will revise the cap having regard to changes in

general interest rates, rather than tying the cap to a

particular multiplier of a particular rate.

Interest-only caps can be supplemented by controls on

particular types of fees, for example early termination

fees. If controls are applied to some fees but not to

others, the fee controls should be justified in their own

right. The regulator should be under no illusions that a

patchwork of controls can contain overall credit

pricing. Cost-recovery and profit-making will merely

shift to the uncontrolled areas. Circumstances in

which particular fee controls may be justified include

the following.

• The regulator wishes to control a particular aspect of

pricing, for policy reasons. Examples are the

prohibition of early termination fees to encourage

household saving, and limits to delinquency and

late payment fees to reduce the chances that the

debt burdens of defaulting households will

accumulate out of control.

• The regulator wishes to ensure that certain elements

of cost are recovered indirectly. In Victoria there is a

longstanding prohibition of direct charging for the

legal fees incidental to loan establishment.

• The regulator believes that the particular aspects of

pricing are anti-competitive but that other aspects

are competitive. Regulators may thus prohibit or

limit types of fee that are easily hidden in the small

print in order to force cost-recovery into areas where

prices are more exposed.

Interest-only caps have also been specified with

interest payments broadly defined to include all

payments from the borrower to the creditor other than

recovery of the loan principal. This approach has been

taken in NSW in applying the traditional 48 per cent

cap to short-term credit. Having defined interest

broadly and published a formula to calculate the

annual percentage rate, the next step is to pick a high

number for the cap and stick by it. This approach can

be defended by starting from our first rationale for

caps on the cost of credit – the general feeling (perhaps

folk memory from the days before 1854) that action

should be taken against usury. A supplementary

rationale is the demand from welfare agencies that

action be taken against high-cost loans that threaten

to precipitate borrowers into downwards debt spirals.

However, as pointed out above, this is a blunt

instrument. The correlation between high interest rates

and debt spirals is fairly loose. Again, the approach

does not accord with the cost analysis. Effectively, it

prohibits short-term small-amount loans, where

establishment costs are inevitably high in relation to

the amount lent and cannot be fitted within a 48 (or

whatever) per cent all-encompassing interest rate cap.

Equally, it exercises virtually no pressure on large or

long-term loans, even in cases where the lender is

profiteering. If the policy goal is specifically to prohibit

short-term small-amount lending, the instrument is

ideal, but policy-makers should be aware that this

indeed is its effect: legal short-term small-amount

loans will disappear from the market except for those

who have unused credit card debt limits or those with

items to pawn. If the policy aim is to prevent

profiteering in lending, the all-encompassing singlerate

cap is fairly useless.

A cap specified in cents per dollar lent makes sense

from a product costing point of view only if limited to

short-period lending (say under three months) and if it

is inclusive, rather than interest-only. For short-term

loans, administrative costs dominate, and time-variant

costs are not enormously different for a two-week as

compared to a two-month loan. Such a cap is specified

in terms that accord with the customary pricing of

payday loans, and accordingly should be understood

by borrowers. However, difficulty may arise in aligning

this cap with whatever cap is applied to longer-term

loans.

If the cents per dollar cap applies merely to ‘interest’

and fees may be charged in addition, it is useless

except for cosmetic purposes.

In so far as the aim is the control of monopoly pricing,

cost analysis leads inexorably to a structured cap, the

basic elements of which were detailed above. A

structured cap requires definitions of fees and interest,

constructed in such a way that every possible element

in price is included in one or the other. The suggestion

is that interest should be the residual category, with all

payments in excess of loan principal and not defined

as fees incorporated into the calculation of interest

payments.

The cost analysis suggests the following classification

of the elements in the price of credit.

• Fees precipitated by the act of application or by the

establishment of a loan.

• Periodic service fees, precipitated by the passage of

time but not related to the size of the loan.

• Periodic interest payments, precipitated by the

passage of time and related to the size of the loan.

• Delinquency payments, precipitated by breach of

any condition of contract, including both lumpsum

charges and interest on arrears.

• Early repayment fees, precipitated by repayment of

the loan before term.

Delinquency and early repayment fees are not part of

the price of a loan contract. In both cases they relate to

costs that the lender incurs over and above the costs of

a contract where no condition is broken, and in both

cases incentive effects may influence the pricing and

result in lenders over-recovering the relevant costs.

Failure to regulate can therefore result in excessive

cost-recovery from borrowers who repay early, and

from borrowers who fail to keep up regular payments.

Not only is this unjust, there can be indirect effects.

Over-recovery of early repayments can discourage

household saving, while over-recovery for default can

increase borrowing costs for borrowers with fluctuating

incomes. This can occur even in the absence of caps –

the banks have been accused of publicising their

‘reasonable’ rates of interest on credit cards, while

making their profits from the unpublicised charges

levied on delinquent borrowers.

Possible approaches to these imposts include the

following.

• As argued in the cost analysis above, prohibition is a

strong contender for early repayment fees, and a

possible contender for delinquency fees.

• A possible approach to delinquency fees and early

repayment is to bring them within some aspect of

the general overall cap.

• Alternatively, they can be separately defined and

separately capped.

In a country that wishes to encourage household

saving, the main argument against prohibition of early

repayment fees is that they discourage re-financing of

fixed-interest loans in the event of a fall in interest

rates. Such refinancing does not contribute to

household savings – indeed, sometimes the reverse, as

households take advantage of the lower interest rates

to increase their indebtedness. However, in the context

of a structured cap a disincentive is still provided by

the need to pay establishment fees on the new loan.

Prohibition of delinquency charges would reduce the

probability of borrowers being precipitated into

downward debt spirals, but condones the decisions of

delinquent borrowers who miss due payments without

pressing cause, and so has unfortunate incentive

effects. Delinquency charge related to costs and

significant enough to have an incentive effect in these

cases can accordingly be justified.

Delinquency charges can be brought under a general

cap in two ways.

• Delinquency charges may be allowed until the cap

is reached for each individual loan, after which they

are prohibited. This would be quite difficult to

administer, and could offer perverse incentives to

deliberately delinquent borrowers (as distinct from

those whose delinquency reflects financial

necessity).

• A ‘standard’ level of delinquency could be assumed

in determining whether a given contract respects

the cap, the actual charges under that contract

being disregarded.

Neither of these approaches accords with the cost

analysis, which recognises delinquency as a distinct

cost driver. We return, therefore, to a separatelyspecified

cap for delinquency charges, separately

defined. This cap should be plausibly cost-related,

sufficient to have a disincentive effect, but otherwise

defined on the low side so as to limit its contribution

to snowballing debt. An alternative approach would be

to consider re-finance as part of a continuing credit

contract, and hence require it to fit within the cap (if

any) on periodic loan service fees. However, this could

only be enforced if the loan was refinanced with the

same lender. To maintain equity between same-lender

and new-lender refinance, it would be preferable to

treat refinance as a case of establishment of a new

loan.

Periodic services costs tend to be small, and there is

something to be said for incorporating them into the

interest rate cap. The contrary argument is that such

costs do not vary with loan size. Incorporating them

into the interest rate therefore penalises lenders more

heavily for small loans than for large.

Periodic service fees are particularly important for

continuing credit contracts, such as credit cards.

Because of this importance, it is likely that a structured

cap would include a periodic fee cap.

Upfront fees raise a number of questions.

• Should there be separate caps for application fees

and establishment fees?

• What about membership fees charged to join a

group prior to applying for a loan?

• How far should a cap on upfront fees be structured

to take into account differential costs of loan

assessment? Two relevant drivers have been

suggested: increases in costs with loan size, and

increases in costs with the presence of collateral.

• Should the cap on upfront fees also include any fee

payable to cover administrative costs on

termination at the end of the loan, or should this be

regarded as a form of periodic payment?

The simplest approach to the application/establishment

distinction is to disregard it, with a joint cap on

upfront fees for both purposes. Whether lenders

charge application or establishment fees will then be

left to their own marketing decision. If they want to

encourage applications, they will keep application fees

low or zero; if they want to minimise costs by

discouraging frivolous application they will charge

application fees. There is no particular social interest in

interfering with this marketing choice. However, the

law should be so worded that application fees paid to a

third party are included in the definition.

Membership fees are not currently a major concern,

and can probably be left to the market – i.e.

unregulated – since loan clubs will not be attractive

unless they can match open market lending. However,

the law should be so worded that bogus membership

fees, which are in effect application fees, are caught.

The cost assessment strongly favours variation in the

upfront fee cap with loan size and between loans

where collateral must be valued and secured and those

where less extensive legal work is required. The

obvious problem is that these two variations

complicate the cap, at least from the point of view of

customer understanding. (The paperwork of

calculating a cap that varies with loan size and security

is a cinch compared with the GST, and should cause

no problems to lenders, to consumer advocates or to

the regulator.)

According to conventional cost analysis, loan

termination costs are low, and any regular-course

termination fees (as distinct from early termination

fees) are probably best included in the calculation of

the interest rate.

The cost analysis suggests two things.

• As discussed above, interest should be capped as a

mark-up on a marker rate.

• The mark-up for risk should be less for loans with

collateral than for unsecured loans. This differential

is already present in the Victorian legislation.

The mark-up for risk effectively determines the default

rate that the lender can bear, and accordingly the cutoff

assessment of credit-worthiness at which a rational

lender will be willing to lend and the importance of

collateral. The more vigorous the debt-collection

mechanisms which the state sanctions, the lower the

mark-up required for a borrower of given creditworthiness;

however, this hardly constitutes an

argument for standover tactics in debt collection.

Better to combine reasonable debt collection

techniques with a moderate mark-up, and accept the

consequence that not everybody will qualify for

additional loans.

It should be remembered that an interest-rate cap is no

answer to the problem of over-optimistic lenders, who

get both borrowers and themselves into trouble by

approving ill-judged loans. The financial sector is

notorious for its bouts of over-optimism, which are an

important contributor to the trade cycle. However, the

curbing of over-optimism is not the primary task of

consumer regulators, but is rather a responsibility of

the Reserve Bank.

Similarly, an interest rate cap is no answer to the

problem of lenders who make high-risk loans in the

confidence that they will be able to extract repayment

by unconscionable means. This problem has to be

addressed directly.

An obvious problem with the structured cap is its

complexity – though the structure is not much more

complicated than the elements in the Schumer Box

used to provide a standardised outline of loan

conditions in the USA.

Complexity is unavoidable when caps must respect a

variety of cost-drivers that differ loan by loan.

However, in the case of short-term low-amount loans

it may be possible to simplify the cap by multiplying

out the components and specifying this as a cents-perdollar

cap. This approach could be applied, for

example, in the case of pay-day loans, where all loans

for less than 63 days and for less than (say) $5000

could be given the same cents-per-dollar cap as a 63-

day loan for $5000. Such a specification would fit

neatly into a more general structured cap.

A related problem is that of stifling innovation. Most

obviously, structured caps with an emphasis on costrecovery

through interest rate differentials are not

suited to the development of Islamic banking, but

more generally innovative lenders may wish to exceed

the cap in some directions while remaining well

within on others. This problem could be circumvented

by multiplying out the arguments of the cap for each

loan, specifying the result as an equivalent annual

percentage rate, then allowing lenders to vary the

components provided they meet the overall cap.

However, this introduces an additional complexity

into the determination of excessive credit price. The

alternative may be to allow exemption for specific loan

products, such as Islamic contracts.

A claimed advantage of the structured cap is that it

encourages a distinction between costs recoverable

from interest and those recoverable from fees. As

compared with an overall cap expressed in annual

percentage interest rate terms, this allows the quoting

of relatively ‘reasonable’ interest rates. This in turn

may encourage the entry of mainstream lenders into

short-term small-amount lending, since they can do so

without quoting interest rates that the public may

regard as outrageous.

The implication of the above discussion is that a

structured cap would include sliding scales related to

cost drivers. Such differentiation is no problem in two

circumstances.

• The cap element is determined by formula from a

readily-defined base. An example is the interest rate

cap, which is determined from the loan principal.

• The cap element reflects readily-defined

circumstances, for example secured and unsecured

loans, or contingent and non-contingent fees.

The sliding scales incorporated into our discussion of a

possible structured cap do not go beyond these limits.

However, discrimination between loan types can

become contentious if there are not based on tightlydefined

differentials. Two examples follow.

• There are strong economic arguments for

discrimination between loans by purpose. Lenders

used to be instructed to discriminate in favour of

house purchase, but so many mortgages are now

used effectively for other purposes that this is no

longer possible. More generally, purposes are so

difficult to define, and there are so many ways of

ignoring the definitions, that general instructions to

favour loans for certain purposes are unlikely to be

obeyed.

• There is a potentially vague boundary between

secured and unsecured loans. The boundary is

precise enough when secured loans are restricted to

those where the collateral is mortgages over real

property or marketable financial assets, but becomes

hazy when the collateral is not easily repossessed or

is of uncertain market value. In the above discussion,

it is assumed that a fairly high standard of collateral

is set as the cut-off point for a secured loan.

• If desired, it may be possible to maintain a

distinction between continuing and terminating

credit contracts. However, financial innovation has

been blurring this distinction. The cost analysis does

not give it any great prominence as a cost driver, so

there is no particular call to include it in the

arguments of a structured cap.

As observed in recounting the history of credit

regulation, the re-regulation of lending after the deregulation

of 1854 began with allowing the courts to

re-open harsh or unconscionable contracts. The courts

proved able to recognise procedural unfairness, but did

not consider themselves competent to declare prices

unconscionable. The 48 per cent cap was introduced

to counter this understandable reluctance.

When the problem is identified as unconscionability, it

may be either a general sense that the price of credit is

‘too high’ or that unmanageable credit is being

granted. As we have seen, the first of these problems is

not quantifiable – hence the reluctance of courts to

deem unconscionability. The second is more

interesting.

The primary line of defence against the granting of

unmanageable credit is and must remain the lender’s

interest in being repaid. It should be remembered that

granting credit always carries risks, and there will

always be loans that become unmanageable due to

bad luck, such as natural disaster, unemployment or

sickness. Such cases of unmanageability cannot be

controlled by improved lender practices, and require

policies to prevent unethical debt recovery, to assist

over-indebted consumers with debt management, and

at the limit to relieve them through bankruptcy.

However, there are cases where unmanageable loans

are granted through lender malpractice or negligence.

These include the following.

• The lender has failed to carry out checks of creditworthiness.

A hard case is where the applicant has

lied and this has not been picked up by the lender’s

independent checks. Exposure of lenders to this

type of error can be reduced by positive credit

reporting, but this is opposed on civil liberties

grounds. Attempts can be made to introduce

disincentives to lying, such as criminal

punishments, but these can get out of proportion to

the offence, and may not be particularly effective

when the borrower is more confused than

mendacious. The South African Bill includes a

provision that loans will not be recoverable in the

case of reckless lending, defined in practice as failure

to carry out elementary checks of credit-worthiness.

• The lender is relying on unacceptable methods of

debt recovery. In addition to standover tactics, these

may include jumping the queue of creditors.

Enforcement of fair debt recovery techniques and of

the law of bankruptcy is the obvious answer here.

In neither of these cases are high credit prices any

more than a warning sign that other malpractice may

be taking place. A regulator who is primarily motivated

to curb unmanageable lending will keep a weather eye

on credit prices, but will devote major effort to

improving checks of credit-worthiness and policing

the limits of acceptable debt recovery.

It has been suggested that, rather than disallow

unconscionable lending, the authorities should

disallow lending at exorbitant rates, with exorbitant

defined as excess cost recovery. This would resemble a

structured cap, save that the cap would not be

published but would be administered on a case by case

basis. To administer such a system by assessing loan

contracts in arrears and declaring those with high costs

unenforceable, the regulator would have to gather

much the same cost information as is required to

specify a structured cap. Charges which appear

exorbitant (or profiteering) by the standards of this

body of information would be disallowed, and if there

is appeal a body of case law would develop which

would yield the equivalent of a structured cap –

probably a very complex one. If there is no scope for

appeal, the regulator is likely to disallow only the more

outrageous cases, in which case there is likely to be

little pressure on excess profits. This may, of course, be

the policy intention.

The Irish have developed the non-prescriptive

standard further by shifting the emphasis from the

regulation of contracts to the licensing of lenders.

They require lenders to submit cost analyses as part of

their licence application. Those whose costs appear

high are simply not licensed. Managed carefully, this

can solve the information problems that arise when

structured caps have to be specified – the lenders

provide the information in the course of application to

be licensed, and have an incentive not to over-state

their costs lest they be not licensed. What ensues is

not so much a price cap as a cost cap.

Even under this system, many of the difficulties of

administering a structured cap are likely to remain.

Keeping costs low is ultimately a surrogate for not

making high-risk loans, so the regulator cannot avoid

the problem of the risk cut-off. If licensing is to bear

down on costs across the whole range of risk,

including relatively low-risk loans, the regulator still

has to second-guess aspects of the lenders’ cost

structures. Similarly it is likely that the regulator will

require that accounts be prepared in a standard format,

and this may discourage innovation in cost analysis

and control. De-registration may also turn out to be a

blunt instrument, as in the case of a lender whose

overall costs are acceptable but who, according to the

regulator’s cost analysis, is providing a mixture of

‘high’ and ‘low’ cost loans.

An interesting problem could arise if lenders were

involved in particular classes of loans where there is

considerable competition, and in other types of loan

where there is restricted competition – say housing

loans and micro loans. In this case competition will

result in generally low cost recovery in the first case,

and high cost recovery in the second. If the regulator

does not have an independent cost model but relies

on reported averages, the result will be the

institutionalisation of excess cost recovery in the highrecovery

area – the exact opposite of the regulatory

intention. This result can also arise under a structured

cap.

Experience in the regulated utilities shows that, where

productivity gains are available, it is possible to

regulate prices downwards without affecting output

quality or quantity, but this experience will be difficult

to transfer to a multi-product industry like consumer

finance, where outputs are not easily measured and it

is therefore difficult to estimate the scope for

productivity gains.

From a business point of view, the Irish approach has

the advantage that it does not interfere with any of the

details of credit pricing. Providers can innovate and

cross-subsidise, and will probably not be prevented

from writing loans which are high or unconscionably

priced when these are offset by others which are low

priced – assuming that regulators give up on trying to

second-guess cross-subsidies. A disadvantage from a

lender point of view would be the uncertainty of never

quite knowing the boundary of acceptable business

practice.

From a consumer point of view, the effect of the Irish

system should be to bear down on business costs and

hence on prices. The effect on consumer awareness

would depend on product disclosure regulations,

which would presumably be included in the licensing

requirements. The absence of a published cap would

deprive consumers of a possible measure of credit price

– the differential between the quoted price and the

cap. We do not know the effect of this, since there are

no current instances of structured caps, and we

therefore do not know whether the cap would act as a

comparison rate. However, consumers would certainly

be deprived of an objective measure for whether or not

they should report a credit provider to the regulator. It

is a moot point whether this would result in more or

less reports.

In Australia, cost data is not required as a condition of

state licensing of lenders, but the Commonwealth’s

requirements through ASIC include the provision of

financial information (Lanyon 2004). It is possible that

ASIC might implement a version of the Irish approach,

leaving the states with a somewhat more tractable task

of regulation at the level of individual contracts.

We have isolated three main arguments for controls

on credit pricing. The first is that, despite the

deregulations of 1854 and 1984, custom demands that

interest rates should be capped. In Victoria the 48 per

cent cap meets this requirement, but has little effect

since it is above market rates for nearly all capped

loans. By contrast, the NSW cap of 48 per cent allinclusive

(including short-period loans) is of rather

more than symbolic effect. It is below cost-recovery for

short-period loans, which it therefore prohibits.

The second argument is that prohibition of high-cost

credit assists in preventing consumers from

contracting unmanageable debt burdens by preventing

lenders from making high-risk loans where their costs

are more than the cap. However, a cap is a blunt

instrument for this purpose. It does not address all

causes of over-indebtedness, and also prevents loans

being made that would not result in overindebtedness.

It is necessarily secondary to other

approaches, such as controls on debt recovery, controls

on reckless lending, debt counselling and bankruptcy.

The third argument is that lenders exercise monopoly

power to over-price credit, at least in some sectors of

the market. The sector of most concern is that serving

poorly-informed, marginally credit-worthy consumers.

The conventional answer to monopoly pricing is the

promotion of competition. Thus the Uniform

Consumer Credit Code attempts to facilitate

competition by the imposition of uniform disclosure

provisions, the purpose of which is to prevent lenders

from creating monopolistic niche markets by

bamboozling borrowers. It is a matter of judgement

whether pro-competition strategies of this kind are

practically effective. However, the consequences for

consumers of failure of competition, in terms of

wrecked household accounts, are such that a case can

be made for controls that supplement competition

policies by disallowing high credit prices.

This argument leads inexorably to a cap set at levels

that allow lenders reasonable cost recovery and profit.

Such a cap has to cover all credit-related charges –

credit costing is so flexible that lenders who are aiming

for excess profit will have no compunction in

increasing their uncontrolled charges. The logic of this

leads to a structured cap that reflects major cost

drivers. It also requires that the regulator make a

judgement (or judgements) as to the cut-off level of

risk for acceptable lending.

A structured cap can be published, or alternatively can

be kept in the regulator’s private conscience and used

with discretion to identify outrageous cases. This latter

approach leaves a gap as to how the outrageous cases

are to be prosecuted. The non-prescriptive approach is

likely to work best when it is used as a condition of

discretionary licensing of providers, rather than to

disallow particular contracts.

Approached this way, the various options of legal

specification tend to merge. The fundamental

distinction is between a system with gaps, so that

profits can be made by raising the uncapped portions

of total credit price, and a system with complete

coverage. If coverage is complete, the choice is either a

flat-rate cap or a structured cap. The former, such as

the NSW all-inclusive 48-per cent, is effective in

prohibiting short-term high-risk loans, but is safely

above market for most other loans, including

profiteering loans. In principle, structured caps can be

specified to bear down on profiteering in all areas of

the market, but this may require an impractical level

of complexity coupled with detailed and accurate cost

analysis. It must also be admitted that structured caps

are a new idea, and there is little experience with their

administration. Indeed, the costing principles on

which a structured cap is based are also a fairly new

area of analysis, and experience may prove that the

costing is not as robust as required for regulatory

purposes. Certainly there are difficulties in obtaining

accurate data on which to base the cap, though these

should not be insurmountable. Given the deficiencies

of other specifications, it is difficult to get away from

the structured cap as a possibility for further

investigation.

Allen and Unwin.

Economics and Liberty

Melbourne, Longmans

Commercial Tribunal (NSW), 1992, Report on interest

rate ceilings, NSW Parliamentary Papers

Credit Law Review, Department of Trade and Industry,

South Africa, 2003, Detailed Analysis of Policy

Options.

Finance, Monash University, working paper.

London, Black Swan

Department of Trade and Industry, South Africa, 2004,

departmental research paper.

and Robertson

Interest and Money, London, Macmillan.

Lanyon, E., 2004, ‘Changing direction? A perspective

on Australian consumer credit regulation’, paper for

the Department of Justice, Victoria, second national

consumer credit conference.

London, Macmillan

Manning, I., 1995, ‘The Social Responsibilities of

Manning, I., 2004, ‘Are we heading for a fall?’, paper

for the Department of Justice, Victoria, second

national consumer credit conference.

A.Birnbaum, Boston, Houghton Mifflin.

Norton, W.E., P.M.Garmston and M.W.Brodie, 1982,

Ramsay, I., 2004, ‘Consumer credit regulation as “the

third way”?’ paper for the Department of Justice,

Victoria, second national consumer credit conference.

Schierenbeck, H., 2004, ‘Consumer credit regulation’,

paper for the Department of Trade and Industry, South

Africa.

Edition, Melbourne, Scribe.

Treasury and the South African Reserve Bank.

Wilson, D., 2002, ‘Payday Lending in Victoria’,

Research Report for the Consumer Law Centre

Victoria.

Report to the Minister for Fair Trading.