1

CONTENTS 1

SUMMARY 3

1. INTRODUCTION 5

2. STRUCTURE OF AUSTRALIAN RETAIL BANKING 6

2.1 Market shares and access 7

2.2 Breadth of banking products 10

2.3 Types of bank fees and charges 10

3. MEASURING THE PERFORMANCE OF BANKS 14

3.1 Net interest rate margin 14

Net interest spread 15

International comparisons 17

Distributional issues 17

3.2 Other indicators of competitiveness 18

Market share 18

Profitability 19

Cost efficiency 20

Service quality and product range 20

3.3 Summing up 20

4. IS COMPETITION EFFECTIVE? 21

4.1 Single bank market power 21

4.2 Collusion 21

4.3 Oligopolistic pricing 23

4.4 Consumer inertia and ‘lock-in’ 24

5. USER PAYS 27

5.1 Impact of deregulation 27

5.2 One step at a time 28

5.3 Charges can be further refined 29

6. FINANCIAL INSTITUTIONS AND COMMUNITY SERVICE

OBLIGATIONS 30

6.1 Who are the disadvantaged groups? 30

Exemptions from paying fees 31

Avoidance of fees 31

Conclusion 31

6.2 Options for government 33

6.2.1 Social security payments by cheque 33

6.2.2 Further options 34

Compensation via social security payments 34

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

2

In-kind vouchers 35

General subsidies to all bank consumers 35

6.2.3 A “community service obligation” on banks? 35

6.3 Summing up 36

REFERENCES 37

3

Following the deregulation of the Australian financial system, bank fees and

charges have become part of the Australian retail banking landscape. This

development has not been without controversy. Concerns have been expressed

that fees and charges are unwarranted (resulting from exploitation of market

power), or that they impact disproportionately on certain groups within society.

This submission examines -

1. whether fees and charges are set in a competitive environment,

2. looks at their incidence on different groups within society; and

3. assesses the case for government intervention.

From a number of perspectives, it is apparent that there is now significant

competition in the retail banking industry in Australia. For one thing, no bank

has significant market power on its own. Moreover, given the number of sellers

of banking services, the diversity of banking products sold, and the demonstrated

capacity for entry to and exit from the banking industry, the pre-conditions for

collusion among banks are not present. Similarly, the high number of sellers

means that oligopoly pricing models are inappropriate. Increased competition

since deregulation is supported by evidence of a reduction in the interest rate

margin received by Australian banks, as well as other indicators such as reduced

profitability and increased cost efficiency.

An indication of the competitiveness of the retail banking industry in Australia is

that consumers in most parts of Australia can choose between many different

banks and the competitive fringe of operators providing banking services.

Although some areas (eg country towns) may have a more limited choice, the

advent of technology based services (eg telephone banking) enhances consumer

options even in poorly serviced areas.

Prior to deregulation, the Australian retail banking system was characterised by a

high number of cross-subsidies amongst different categories of consumers.

Income was derived largely from interest rate margins; there were few specific

fees. This was inefficient and penalised some consumers. Since deregulation,

fees and charges have increasingly been applied on transaction accounts, and the

proportion of income from interest rate margins has declined.

The Office of Regulation Review (ORR) considers that this development is a

desirable consequence of the competitive pressures faced by banks, which have

made it difficult for them to sustain cross-subsidies and avoid user pays pricing.

While enhancing economic efficiency, the introduction of bank fees and charges

can potentially impose a disproportionate burden on those with low incomes,

such as social security beneficiaries and students.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

4

At this stage, however, the ORR cannot find a compelling case for government

intervention on equity grounds. This is because banks currently exempt from

fees and charges most potentially disadvantaged groups within the community.

In addition, there are several avenues open to consumers to minimise or eliminate

the impact of the current structure of fees and charges. Apart from its impact on

efficiency, government intervention would also reduce the incentive for banks to

provide exemptions from fees and charges.

The structure of bank fees and charges could change in the future. In that case,

the argument for government intervention on equity grounds may become

stronger. If the government chose to intervene, the ORR considers that on

balance the most efficient approach would take the form of a “community service

obligation” on banks to provide specific groups of customers with fee-free basic

banking services, with the Commonwealth Government reimbursing the banks

for the costs of doing so.

5

The last 10 to 15 years have seen significant microeconomic reform within the

Australian economy. These reforms have had the effect of opening the economy

internationally, and of addressing regulatory impediments to competition in many

previously sheltered markets. A key element of microeconomic reform has been

the deregulation of the financial system.

Financial deregulation has brought a wider range of products and services offered

by all financial institutions, especially banks. The banks also have rationalised

their branch networks and their staffing, expanded into new fields including

stockbroking and insurance, and acquired substantial subsidiary interests overseas.

There have been, however, some downsides associated with financial

deregulation. The spate of high profile corporate failures in the late 1980s and

associated loan losses by banks have both been linked, fairly or unfairly, with

financial deregulation. Perhaps the most controversial change to be associated

with deregulation has been the introduction of a range of fees and charges on

retail transaction accounts. There are concerns within the community that many

of these fees and charges impact disproportionately on disadvantaged groups.

The concern expressed by many groups over increased fees and charges led the

Assistant Treasurer to instigate a public inquiry by the Prices Surveillance

Authority (PSA) into fees and charges imposed on retail transaction accounts by

banks and other financial institutions. (In 1992 the PSA conducted an inquiry

into a related matter, credit card interest rates.)

This submission considers whether retail banking fees and charges are likely to

be set as outcomes of a competitive marketplace, and what their impact might be.

Section 2 outlines the structure of retail banking in Australia and describes some

of the fees and charges which might apply to certain transactions. Section 3

examines various statistical measures of the degree of competition, loosely

referred to as competitiveness, and their usefulness. Section 4 assesses the

competitiveness of the retail banking services sector. Section 5 discusses how

explicit fees and charges for specific banking products and services are a

consequence of a competitive marketplace. Section 6 considers which groups in

society are most disadvantaged by bank fees and charges, and evaluates several

possible methods of government intervention to assist them.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

6

Australia’s financial system can be divided into five main categories:

Competition for the provision of retail banking services comes from banks and

non-bank financial institutions. The category of non-bank financial institutions

includes the building society and credit union movements.

The banks can be classified into four main categories:

The major banks have extensive branch and agency networks and operate

throughout Australia. State banks operate principally within each State, although

some are now extending their operations to other states. The regionally operating

banks are commonly building societies which have converted to banks, and tend

to focus their activities in a niche market.

In 1994, there were 44 banks operating in Australia. This included the four major

banks, three State banks and twenty one foreign banks (RBA 1994a).

There are numerous types of non-bank financial institutions, such as building

societies, credit unions, cooperative housing societies, authorised money market

dealers, finance companies and general financiers.

While all forms of non-bank financial intermediary provide some retail banking

services, only building societies and credit unions provide the range of services

provided by banks. In June 1994, there were 28 building societies and 308 credit

unions operating in Australia.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

7

There is no single accepted measure of the size of, or the market shares within,

the Australian banking industry. This section presents two different measures:

value of deposits repayable by banks; and numbers of bank branches.

Total deposits repayable by banks in Australia amounted to $246 000 million as

at October 1994. Figure 2.1 presents the market share of the different banks in

terms of deposits repayable. As expected, the largest banks (measured in these

terms) are the four major banks.

Figure 2.1: Bank market share in terms of deposits repayable

Source: RBA 1994b.

Building societies and credit unions provide a sufficiently diverse range of retail

banking products and services for them to be regarded as substitutes for banks.

Incorporating building societies and credit unions, total deposits repayable by the

presents the market shares of the different institutions in terms of deposits. The

largest market shares also are held by the four major banks.

Deposits repayable are not necessarily an accurate reflection of market shares in

retail banking because commercial customers also hold deposits, and building

societies and credit unions concentrate on retail banking services. Branch

unions hold deposits with banks.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

8

numbers are in many ways a better measure of market share because branch

networks are basically designed to cater for retail customers.

Figure 2.2: Market share of financial institutions in terms of deposits

repayable

Source: RBA 1994b.

As at June 1994, there were 6790 bank and 950 credit union branches in

Australia. Figure 2.3 disaggregates the number of branches by institution across

Australia. The building society and credit union movements clearly have a level

of market access equivalent to each of the major banks.

The national market shares do not necessarily reflect the market shares in each

state. For example, in South Australia the State Bank has over 20 per cent of the

retail banking branches, yet State Banks nationally have only a 7 per cent market share.

For many consumers, however, the issue is the degree of access they have to their

savings. A very different story emerges when electronic access is measured.

Each of the major banks has an extensive telling machine network. In June 1994,

there were 5724 automatic teller machines (ATMs) in Australia.

Many State and regional banks, building societies and credit unions have access

to similar size or even larger networks of telling machines than the major banks.

In most cases, the State and regional banks, building societies and credit unions

not only have their own telling machines, but their customers also have access to

the networks of the National Australia and ANZ Banks for many transactions.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

9

Figure 2.3: Market share of financial institutions measured in terms of

branch numbers

Source: RBA 1994b.

For example, credit union members can use Redicard to access about 3350

approximately 2350 ATMs. Hence, from an access stand-point, most financial

institutions have telling networks that are viable substitutes for each other.

All domestic financial institutions and some foreign banks have access to the

1994. When these products are included in computations of market shares, most

financial institutions have approximately the same number of access points.

Most card holders can access almost 50 000 transaction outlets.

While branch networks have been reduced over recent years, the growth of the

autobank and EFTPOS networks has resulted in a considerable increase in the

number of retail banking outlets accessible by customers.

terminals located in post offices and in the premises of retailers acting as agents for financial institutions.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

10

Banks, building societies and credit unions all offer a vast range of banking

products and services. They all offer basic deposit accounts. These can include

card accounts, cheque accounts, passbook accounts and special purpose accounts

(eg accounts for children and pensioners).

Most financial institutions also offer variations on these basic accounts. For

example, the Elliott Committee surveyed the types of accounts offered by 15

banks in Australia, and found 56 card accounts, 41 cheque accounts, 55 passbook

accounts and 18 pension accounts (Elliott 1992). These numbers would be much

greater if all banks, building societies and credit unions were included in the survey.

Other examples of retail banking products include personal loans, fixed term

deposit accounts, credit and debit cards, home loans and investment accounts. It

is difficult to summarise the range of accounts offered by financial institutions

across Australia.

This breadth of products is due in part to technological innovation, best

illustrated by the growth of electronic teller networks in the last decade. In

addition, advances in data processing and telecommunications have enabled

banks to offer home banking (for personal customers) and cash management

systems (for corporate customers) which allow customers to undertake banking

activities from their home or office.

Banks have also diversified their operations into other areas of financial activity

beyond traditional boundaries, including superannuation, funds management,

financial planning and advice, life insurance, stockbroking, trust services and travel.

Table 2.1 summarises some basic accounts offered by a selected group of

financial institutions.

Excluding government taxes and charges, there are basically three types of fees

and charges on basic deposit accounts:

specified fee-free threshold is maintained;

impose such charges, with most offering unlimited access to the electronic network; and

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

11

withdrawals or cheques. Most institutions permit customers to make 2 or

more such transactions each week before charging customers. Deposits do

not incur charges.

All of the selected financial institutions offered exemptions from fees and

charges to specified groups within society. In most cases, aged pensioners, under

18s and full time students do not incur fees and charges.

13

Table 2.1: Fees and charges on typical small accounts

Electronic

transactions

15 per month, then 50c

each

Unlimited Unlimited Unlimited Unlimited

Exemptions Under 18; trustee accounts

for minors; home

borrowers; students; gold

card holders

Aged and war veteran

pensioners; under 21;

home borrowers;

students

Aged and war veteran

pensioners; under 21; home

borrowers; existing

customers before Jan 1

Students under 21;

trustee accounts

for minors

Most pensioners;

under 18; students

Electronic

transactions

Unlimited 7 per month, then 50c

each except through

Advance ATMs

Unlimited 10 per month, then

30c each

Exemptions Students; pre-schoolers;

most pensioners; retirees

Aged and war veteran

pensioners; students under

16; home borrowers

Aged and war

veteran pensioners;

students under 25

Trustee accounts for

minors; under 18;

Target account

a. Some institutions define withdrawals to include cheques. Deposits incur no charges.

Sources: Bulletin, 7 February 1995; and information provided by the financial institutions.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

14

A common approach to assessing the degree of competition in the retail banking

services sector is to analyse various statistical measures. This section examines a

variety of such statistical measures. The primary statistic used in banking is the

difference between interest rates levied and interest rates paid (the ‘margin’), but

other useful measures relate to market share, profitability and cost efficiency.

Such statistical measures provide useful indicators of the competitiveness of the

market, but cannot be taken as conclusive evidence. Movements in these

measures, even over a long term period, can be a reflection of factors other than

the competitiveness of the market.

The Australian financial system has undergone substantial regulatory change

throughout the past 20 years. The major changes occurred in the 1980s. Box 3.1

outlines the most important elements of deregulation from a competition perspective.

One benefit of deregulation is that it should have been associated with decreases

in interest rate margins due to increased competition. Has this been borne out in practice?

There is a perception in the community that the net interest rate margin has

increased in recent years. For example, the ACT Consumer Affairs Bureau has

In contrast, analysis undertaken by the RBA and others indicates a decline in the

net interest rate margin since deregulation.

This gap between perception and reality can be simply explained. Statements

about behaviour of interest rate margins are often made with reference to a

comparison of one lending rate and one deposit rate. For example, the difference

between the housing indicator lending rate and a savings account deposit rate.

Such comparisons are misleading. Comparing particular deposit and loan rates

does not provide a reliable indicator of the overall interest rate margin. Banks

have many different types of loans and these are drawn from the pool of funds

built up from a variety of deposits and other liabilities. In other words, money is

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

15

‘fungible’. Interpretation must therefore be based on a bank’s overall interest rate

margin, and not on simple rate comparisons.

Financial deregulation has been a gradual process, not a one-off event. The major elements

from a competition perspective (other elements directed at prudential issues are not addressed

here) were:

controls were removed;

were accepted, and the new banks commenced their operations between September 1985

and May 1986;

$100 000;

removed; and

considered, and foreign banks were given the option of operating as branches, rather than as

(or in addition to) locally-incorporated subsidiaries.

The RBA has identified other flaws in using simple interest rate comparisons to

evaluate competition in banking. These include that:

new deposits and loans and average rates paid and received by banks on

outstanding deposits and loans;

the average cost of funds to banks even if cash rates remain unchanged; and

significant differences can persist even over the longer term (RBA 1991).

spread. This is equivalent to the difference between the average rate of interest

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

16

earned on a bank’s portfolio of interest-bearing assets and the average rate of

interest paid to its depositors.

As might be expected, the net interest spread does not fluctuate as sharply as

some indicator interest rates, and certainly by much less than some of the

common measures used in the simple comparisons.

The RBA has found that the net interest spread has followed a downward trend

since deregulation. Figure 3.1 shows the net interest spreads for the major banks

percentage points in the first half of the 1980s, and between 4.5 and 5 percentage

points in the second half (RBA 1992). In 1990 and 1991, the net interest spread

is estimated to have declined to just under 4 percentage points, and has since

stabilised at about that level (RBA 1994a).

3

3.5

4

4.5

5

5.5

1980

1982

1984

1986

1988

1990

1992

1994

Source: Data supplied by the Reserve Bank of Australia.

While spreads have been lower on average since deregulation, this result is not

deposits to banks as investors sought security for their savings.

earned by banks. The spread on banks’ other loans and investments - the average gross spread - is higher.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

17

The RBA has also attempted international comparisons of interest rate margins.

Given the structural and institutional differences across countries, however, the

RBA has expressed scepticism about the value of the data underlying such

comparisons and about the prospects of deriving any entirely satisfactory

international comparisons.

The RBA has concluded that total income and profitability of Australian banks

are in line with comparable banks in other countries. What is different in

Australia is the mix of income - while interest margins tend to be higher than in

other countries, this is offset by relatively low fees and charges (RBA 1994a).

The structure of the taxation system in Australia, which is characterised by

relatively high marginal rates of tax at relatively low levels of income, could

reduce the incentive to change the balance of these two sources of bank income.

This is so because low fees and charges constitute a non-taxable benefit whereas

higher paid interest rates (which would accompany higher fees) would be subject

to relatively high marginal rates of income tax.

There are some indications that the wholesale banking sector has benefited

considerably more from deregulation than the retail sector. Milbourne and

Cumberworth (1991) decomposed the aggregate net interest margin into its retail

and wholesale components. They found that the net interest margin on retail

activity increased steadily throughout the 1980s, whereas margins on wholesale

(mainly commercial) banking were decreasing.

The results attained by Milbourne and Cumberworth hinge on the assumptions

used to separate retail and commercial banking. They themselves acknowledge,

in reaching their conclusion, that it is difficult to calculate separate margin series

for retail and wholesale activities because the funds used are drawn from a

common pool of assets, to which different contributions incur different costs.

If the gains from deregulation have been concentrated in the wholesale

(commercial) sector, it need not follow that retail customers have crosssubsidised

commercial customers. Indeed, under the previous regulated system it

seems likely there would have been pressure on the banks to favour (crosssubsidise)

the household sector (Milbourne and Cumberworth 1991). It is also

likely that the wholesale sector has seen more intense additional competitive

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

18

pressures since deregulation than the retail sector. Most of the new foreign banks

The RBA has also looked at distributional issues. Although finding that the

average net interest margin had fallen since deregulation, the RBA acknowledged

that did not mean that every customer’s experience had been the same - some are

better off, some are worse off, and some are neither one nor the other (RBA

1992). Chapter 6 assesses which groups within the community are most likely to be disadvantaged.

The RBA found that, as a group, depositors have benefited in the past decade or

so from a rise in the average interest rate paid by banks on their balances,

especially after allowance is made for inflation. This is so for two reasons. First,

interest rates on some categories of low-interest accounts have risen over the

55 per cent of banks’ liabilities in 1981, to about 20 per cent of liabilities in 1991

(RBA 1992).

The trend increase in average interest rates paid by banks has translated into a

corresponding trend increase in the average interest rates sought by banks on

their loans and other investments. On average therefore, borrowers are paying

more for their loans than they did a decade or so ago. But the benefits are

flowing to those people who lend money to or deposit money with the banks

(RBA 1992).

Indeed, it is pertinent that the “household sector”, as defined in the national

accounts, is the major source of domestic saving in Australia (Fitzgerald 1993).

It can therefore be argued that the household sector, overall, is better off as a

result of higher average interest rates paid by banks.

Deregulation of the financial sector has had impacts in areas other than the net

interest rate margin. It is worth looking at some of these impacts in order to

determine whether the results are further indications of competitiveness.

Before deregulation, the restrictions placed on banks and the high cost of their

operations induced substitution away from banks towards non-bank financial

institutions. The result was a loss of market share by the banks to the non-banks.

Bank, Hongkong Bank of Australia, and Bank of New Zealand.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

19

Since deregulation, the combined market share of banks has consolidated, and the

loss of market share to non-bank financial institutions has been stopped. Finance

the latter have become banks.

The additional competition stimulated by financial deregulation was expected to

reduce the profitability of banks. The recent large profits earned by the major

banks are sometimes cited as evidence that deregulation has failed. The more

systematic evidence, however, appears to confirm a general downward trend in

bank profitability since deregulation.

Profitability of the finance industry as measured by the return on equity fell

steadily from approximately double the all-industry average in 1981-82 to less

than two percentage points higher than the all-industry average in 1985-86.

Since then, profitability decreased due to corporate bad debts, but has since

returned to approximately 1985-86 levels (ABA 1994). In a more complex study,

Harper and Scheit found that the shareholders of banks generally, and of the three

major private banks in particular, have not earned supra-competitive risk-adjusted

rates of return on their shareholdings since deregulation (Harper and Scheit

1992).

However, data on bank profits cannot be relied on in drawing conclusions about

the adequacy of competition. Profits studies have a notorious history as an

indicator of the effectiveness of competition and are plagued with measurement

For example, several hundred studies published between 1950 and the early

1970s found that, in most cases, industry profits rose with seller concentration.

This positive correlation was regarded as a consequence of monopolistic and

oligopolistic market power. However, by the end of the 1970s, the interpretation

of this correlation between profits and market concentration was viewed as

ambiguous: the above-normal profits could be the result of superior efficiency,

market power, or both. By the end of the 1980s, further more sophisticated

studies demonstrated that most, if not all, of the correlation between profitability

and market concentration was almost surely spurious (IC 1994).

associated with profits studies.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

20

A major goal of financial deregulation was enhancement of the efficiency of the

financial system. This was to occur through competition forcing institutions,

particularly banks, to operate at lower cost levels.

Bank operating costs as a percentage of average assets for the major banks have

indeed fallen since deregulation (Ackland & Harper 1992). Real deposits per

employee have increased, and unit labour costs have fallen (Milbourne &

Cumberworth 1990). One manifestation of greater cost efficiency is the

increased automation of routine procedures previously conducted manually by bank staff.

One of the most obvious effects of deregulation is the wider range of products

and services offered by all financial institutions, especially banks. The

introduction of electronic funds transfer, 24 hour/7 days per week automated

teller machines, over-the-counter sales of shares and insurance products, the

burgeoning array of financial products including swaps, futures and options, lowstart

and reverse mortgages, securitised loans - all are examples of new products

and services introduced into retail and/or wholesale financial markets since deregulation.

There are numerous statistical measures which have been used to show the

competitiveness or otherwise of the retail banking services sector in Australia.

These measures include the net interest rate margin and the rate of profitability.

None of these measures is perfect, and all can be misused. Nevertheless, it is

significant that all the available measures are consistent with the retail banking

services sector having become more competitive since financial deregulation.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

21

The PSA (1995) has identified competition as one of the key issues of its inquiry.

In principle, competition should stimulate the banks, building societies and credit

unions to provide, at least cost, the retail services desired by their customers.

This section assesses the degree of competition in retail banking. It draws on

At one extreme, there could be ineffective competition if a single bank could

no bank has a market share that is even close to the levels needed for unilateral

market power. For example, the combined market share (in terms of Australian

deposits) of the four major banks amounts to only around two-thirds of the

Antitrust case law and recent econometric evidence suggest that a single firm

would need to occupy at least two-thirds of a market to secure significant market power (IC 1994).

Another reason for ineffective competition might be that there is collusion. The

main avenue for collusion by the banks is tacit: the development of implicit

The prices of retail banking services are publicly known which aids tacit

existence of numerous competitors, the entry and expansion of new banks, the

lack of a standard product and a rapidly changing business environment. For

example, implicit understandings may break down owing to conflicts over the

most suitable price, the complexities of co-ordinating pricing across a diverse

range of products or the simple presence of a maverick institution. As time

passes, destabilising pressures will build due to long-run substitution and the

threat or actuality of entry by new competitors.

Any collusive arrangement in retail banking would call for the co-ordination of a

plethora of existing products and services, as well as the regular introduction of

new products and technologies across a range of established and new financial

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

22

institutions. For example, as noted in section 2, a survey of 15 banks found 56

card accounts, 41 cheque accounts, 55 passbook accounts and 18 pensioner

accounts. Moreover, these numbers would be much greater if building societies

and credit unions were included.

Nonetheless, the banks no doubt note carefully the actions of their rivals,

including the costs of basic products, and a degree of copy-cat behaviour is

likely. However, it is not easy to distinguish competitive from collusive pricing

on this basis. A simultaneous price rise may reflect industry-wide cost changes

or rising demand. Moreover, identical prices would be expected in any market

where buyers have good information and the products of each seller are similar.

As a practical matter, any understandings to reduce competition in retail banking

would have to be virtually market-wide in coverage. Otherwise collusion would

be defeated by competition from non-participating institutions. Moreover, if

there were some collusion in the fixing of account fees, that would only induce

institutions to compete with greater intensity in other areas. Non-price

competition is generally too subtle and widespread to be constrained by an

informal agreement. For example, prior to financial deregulation, there is

evidence that banks responded to interest rate controls by competing in the size

and scope of their branch networks and other forms of non-price competition.

demand or the pace of technological and product innovation are strong. The

Martin Committee observed:

In the past few years the way the average Australian has carried out daily banking has

changed greatly. No longer is it usual to operate a stand-alone cheque account to pay

bills and a passbook savings account for daily transactions and deposits for a particular purpose.

Instead there are combinations of accounts to satisfy almost every banking need.

Accounts can now be combined so that money from a cheque, savings or credit card

account can be accessed by a single card through an Automatic Teller Machine 24

hours a day (pp. 366-7).

Electronic banking is marginalising the advantages of large branch networks and,

more importantly, is making small or regionally based institutions almost

indistinguishable from major banks because all offer similar card access.

The number of retail outlets has increased from about 7,000 branches to over

50,000 access points due to the emergence of telling and EFTPOS networks.

Electronic transactions account for almost half of all transactions while EFTPOS

and Poscash usage is growing an annual rate of 30 per cent. In such a market,

collusion could not endure unless all or most institutions can agree on the

rationalisation of their existing branch networks and their rates of expansion into

new markets and technologies.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

23

Even if there is no collusion, there has been long-standing suspicion that

competition is less than vigorous in concentrated markets. While each bank may

act independently, its decisions must reflect the expected reactions of its rivals.

A pattern of non-aggressive behaviour may emerge as a result.

Many factors beyond concentration bear on the ability to co-ordinate prices

without collusion in the normal sense. They include the heterogeneity of

products, buyer strength, the psychology of the players, and the relative

uncertainty about how rivals are behaving or will behave (Areeda 1984). Hence,

although there may be no need for formal agreements or tacit understandings, the

weight of the evidence is that there is a need for clearly understood signals and

patterns of leadership by one or two of the market’s largest firms (Scherer 1987).

There is some anecdotal evidence that high concentration tends to dampen

competition. But there is also anecdotal evidence to the contrary. For example,

there is strong rivalry within the airline and long distance telephone duopolies.

In addition, the PSA has examined seven different three-member oligopolies

since 1991. It found enough signs of competition to justify the removal or

replacement of prices surveillance in the majority of these inquiries.

The Industry Commission (1994) noted in its submission to the PSA’s review of

prices surveillance declarations that, on best empirical evidence:

is often based on cost or price advantages;

competition;

this process may be slow;

two other suppliers usually results in effective competition; and

little impact on pricing.

The ORR considers that Australia has a moderately concentrated retail banking

sector. The declining significance of branch banking and ready access to

electronic technology allows most banks, building societies and credit unions to

win market share from under-performing rivals.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

24

Effective competition in retail banking may be impeded if consumers are

reluctant to shop around. Some consumers may be locked-in to their existing

institution if retail banking is an ancillary feature of a long-term relationship

premised on a mortgage. It is also possible that consumers may over-look retail

banking fees and the costs of switching out of a long term arrangement when

entering into a mortgage.

Gaps in the availability and quality of consumer information are common-place

in real markets; this creates room for otherwise competitive suppliers to overprice

their products without losing appreciable market share (Stigler 1961).

Moreover, consumers are constrained by income, time and their capacity to

process information when making decisions (Becker 1993).

For example, consumers select financial institutions not only on the basis of

price, but rather their preferred mix of price, service, quality and other features.

Their demands vary over time with economic conditions and requirements such

as the need for a home mortgage. However, the Martin Committee (1991)

expressed dissatisfaction with the amount of consumer information that is

available about banks and noted the complex financial calculations required for

detailed product comparisons.

The Governor of the Reserve Bank has observed that:

If [consumers] shop around, they are likely to discover that they can get better deals

on particular products from different banks. But in practice, many borrowers are

reluctant to shop around for a number of reasons, including inertia and the

convenience of the current ‘packaged’ service (comprising housing loan, cheque

account, credit cards and so on), reluctance to try non-traditional sources of funds,

and the actual and perceived costs of switching some or all transactions from one

bank to another (Fraser 1994, p. 16).

Consumer inertia and switching costs appear to be less of a problem than in the

past. For example, electronic funds transfer, telling machine and EFTPOS

networks have made it easier to conduct business with more than one financial

institution. Hence, the modest market power that flows from gaps in consumers’

knowledge about their banking options should be in decline due to technological

and product innovation and the intensification of marketing efforts following

deregulation. Of course, interest in new products and technology, and the

capacity to use them may be less in certain disadvantaged groups. Thus this

feature of the benefits of deregulation and competition may not be evenly spread

in the community. This issue is discussed in Section 6.

Consumers are often aware of the possibility of lock-in and act to protect

themselves before an initial purchase is made (Klein 1993). For example,

Australian adults already use an average of 1.92 financial institutions and 10 per

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

25

cent of customers change their main financial institution each year. Moreover, a

1989 survey revealed that only 30 per cent of borrowers chose their lender

because they were banking with them at the time of deciding to seek a loan. The

large majority of borrowers compared price, service and availability of different

institutions (ABA 1991).

Most firms have an interest in attracting new business and retaining the goodwill

and regular patronage of their existing customers (Arthur 1994). In addition,

sellers are usually aware of the extra profits from locking-in customers and offer

rebates to attract this type of buyer (Shapiro 1995). For example, the offer of

discounts to regular customers is a common commercial practice.

Thus, in a market with several providers, each seller faces a choice: build a

reputation for serving all customers well, or run down its goodwill by taking

advantage of locked-in buyers. If the actions of a seller undermine customers’

trust, or alter market expectations about future behaviour, the loss of goodwill

may spread to the other products sold by the firm (Shapiro 1995). News of

opportunistic behaviour by a financial institution can spread both by word of

mouth and through the mass media.

For example, studies of the impact of automobile and drug recalls, and the

detection of deceptive advertising, show that the share prices of the companies

concerned fall by amounts far in excess of all plausible out-of-pocket expenses

that may be incurred at the time and in any later litigation. The implication is

that the share market anticipates a significant long term decline in sales and

profits due to a general loss of goodwill (Peltzman 1981, Jarrell and Peltzman

an industry or it is operating in a sharply declining market (Shapiro 1995).

Some financial institutions already appear to be aware of the potential adverse

reaction of consumers to retail banking fees in the context of long term

relationships. For example, some institutions exempt home mortgagees from fees

on their savings, passbook and card accounts (see Table 2.1).

Attempts by an institution to over-charge for retail banking services are likely to

impact on new business and encourage enough existing customers to shop around

that such a strategy should be unprofitable. Hence, the discipline exerted by

mobile consumers, to the extent that they are informed about alternatives, offers

some protection to customers who are poorly informed or locked-in.

in the value of companies; and that new information about a company is capitalised into

its share price within the day that it is released (Easterbrook and Fischel 1991; ORR

1995).

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

26

There is evidence from the PSA’s last inquiry into retail financial services that

suggests that the introduction of fees may actually reduce consumer inertia.

Consumers appear to be relatively insensitive to variations in credit card interest

rates. However, evidence from overseas suggests that consumers are quite

responsive to variations in the fees levied on credit cards (PSA 1992).

The PSA (1992) found that the removal of the prohibition on fees is likely to

result in more diverse pricing of credit cards and lower interest margins and lead

to a closer match of consumers’ preferences and credit card products. A similar

rationale applies to the introduction of fees on retail transaction accounts.

patterns of different groups of consumers.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

27

The pricing regime used by financial institutions prior to the 1980s made

relatively little use of fees based on costs of particular services (user pays). A

large part of this was due to regulations which restricted price competition. It

also reflected more conservative management in banks and less demanding

customers. This meant there were a number of cross-subsidies, such as

depositors subsidising borrowers, infrequent transactors subsidising heavy

transactors, and urban customers subsidising geographically isolated customers.

These cross-subsidies were funded by the net interest margin, and were largely

hidden and possibly unintentional.

Deregulation meant that additional competition was introduced into the domestic

banking sector. Competitive forces placed increased pressure on banks to justify

the use of resources in given activities. The more substantial cross-subsidies

ceased to be sustainable because competitors could always avoid providing the

underpriced services, and undercut where services were overpriced.

The effect of strengthening competitive pressures has been summarised by the

Commonwealth Bank:

If you try to avoid user pays and others are using it or coming in to compete with you

in areas where they do not have the same burden of transactions to service, then they

can pay interest rates which will woo your customers away and leave you stuck like

the proverbial shag on a rock with all the business that is costly (Martin, 1991, p.

104).

The principle of ‘user pays’ pricing is the simplest way to avoid cross-subsidies.

It also increases overall economic efficiency as customers face prices based on

the costs of their use of different services, rather than being encouraged to overconsume

some apparently ‘free’ services and under-consume others. Indeed, to

encourage the provision of innovative and cost-effective banking services, it may

not be in the long term interests of many customers to offer them some crosssubsidised services.

User pays is manifested by the imposition of explicit charges for identifiably

separate services - sometimes referred to as ‘product unbundling’. It was

anticipated that:

Institutions which cut their net interest margins and introduced fees and charges

would expand their borrowing and lending and eliminate profit haemorrhage through

overuse of underpriced ancillary services (Ackland & Harper, 1992, p. 58).

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

28

The PSA itself has noted the process by which increased competition is changing

the way that banks do business, stating that competitive forces have:

led banks to pay closer attention to individual product costing and pricing. The shift

in focus away from intricate structures of product cross-subsidisation to “user pay”

pricing principles ushered in the era of bank product “unbundling”. This involved

identifying as far as possible distinct products and services to achieve a more

satisfactory connection between the costs of producing and delivering those particular

products and services and the revenue they generated (PSA 1992, p. 11).

Deregulation has thus seen a move by banks towards user pays pricing. Banks

have introduced transaction charges for deposit accounts, establishment fees for

loans, together with an increased range of other charges. The proportion of total

income received by the major banks from sources other than the net interest

margin has increased since deregulation of the financial sector. An additional

factor spurring the adoption of user pays by the major banks is the reduction in

the proportion of low-cost deposits held by them.

While both banks and non-banks have made some moves in the direction of user

pays pricing, consumer resistance has been strong. There has also been concern

that fees and charges have had a disproportionate impact upon disadvantaged

groups within society.

It would appear that consumers have become accustomed to using cards as a

transaction device without fees and consequently are resisting fees (PSA 1992).

Credit unions have made a concerted effort to accustom their clientele to fees and

charges as a means of controlling costs in the face of stiffer competition from

banks (Ackland & Harper 1992).

Some commentators believe that the move to user pays has been too slow,

thereby imposing avoidable costs upon the community. Harper is one such critic:

the incomplete use of user pays is sub-optimal because the fear of any one bank being

pilloried, especially by the political mechanism, has made them naturally reluctant.

The awkward thing is that they pass that higher cost back to the community in forms

which are not easily perceived and which, frankly, are inequitable (Martin, 1991, p.106).

The political dimension is illustrated by the inquiries, including the current PSA

inquiry, which have resulted from the introduction of fees and charges.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

29

The structures of fees and charges is yet to accurately reflect the cost structure of

retail banking, so that some cross-subsidies remain. For example, electronic

transactions involve significantly lower costs than manual paper-based

transactions (the banks have estimated that the cost of an ATM withdrawal is

between 30 and 60 per cent of that of a withdrawal over the counter (Martin

1991)). While banks sometimes charge differently for manual and electronic

transactions, the difference is rarely as large as the difference in costs and so does

not yet provide adequate incentive for consumers to make efficient use of

electronic facilities.

As consumers become more used to ‘user pays’, and as banking technology

advances further, fees and charges are likely to be applied more broadly and in a

manner which more accurately reflects the cost structure of retail banking. As

this will take some pressure off interest margins as a source of profitability,

margins could be expected to narrow further.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

30

The evidence presented in this submission indicates that banking fees and charges

are set in a relatively competitive market. Consequently, they are likely to

promote efficiency in the sense that resources are allocated in response to the

costs of and demand for services.

Any move away from explicit fees and charges for identifiably separate services

would result in the re-emergence of cross-subsidies and lead to wasted resources,

in the form of the types and levels of banking services available. Increased

interest rate margins would likely emerge.

But it should be recognised that explicit fees and charges can impose a relatively

heavy burden on some groups in our society. Who they are and what might be

done are addressed in this section.

Table 2.1 shows the fees and charges relevant to basic banking accounts with a

range of financial institutions. It shows that fees and charges have tended to be

imposed on low balance accounts, and on high frequency transactions.

Consequently, it can be argued that fees and charges impact disproportionately on

people with low levels of income. Also, because some fees and charges are

imposed on over-the-counter withdrawals but not on automatic teller machine

transactions, aged persons and those with physical or intellectual disabilities tend

to be disadvantaged.

A large group likely to be disadvantaged by fees and charges are those dependent

on Department of Social Security payments. The Commonwealth Government

requires that all social security and family payments be credited to banking

accounts, which makes banking services essential to most of those who are in

need of financial assistance. In addition, social security beneficiaries are more

likely to have relatively low balances in their accounts. Bank fees and charges

therefore have a disproportionate impact on this group because their bank

balances are often low enough to attract charges, and because the fees and

charges may be perceived as large relative to their income.

A second group likely to be disadvantaged by explicit fees and charges is low

income earners. Like social security beneficiaries, many low-income earners

maintain bank accounts with low balances. Unlike social security beneficiaries,

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

31

however, some low-income earners are at least not compelled to maintain a

banking account.

A third group who might be disadvantaged by explicit fees and charges is

children and full-time students, who often have low account balances.

Banks have voluntarily provided exemptions from fees and charges to many

customers who might be disadvantaged.

Table 2.1 shows that many financial institutions currently exempt some social

security beneficiaries from paying bank fees and charges. For example, most

institutions do not require aged and war veteran pensioners to pay any fees and

charges. Some financial institutions do not levy bank fees and charges on people

under 21 years of age. This means that the 20 per cent of unemployment

beneficiaries who are under 21 could avoid paying bank fees and charges (DSS

1993, p. 272).

Table 2.1 also shows that many financial institutions provide exemptions from

fees and charges for children and for full-time students. These exemptions are

not offered by as many institutions as for aged pensioners, but are offered by

enough institutions to give those consumers a choice.

There are several ways in which consumers who are not exempt can avoid or

minimise paying bank fees and charges. These methods have all been given

some publicity by the banks themselves. For consumers with a sufficient level of

funds, the easiest way is to maintain account balances above the fee-free

threshold, which may involve the consumer consolidating several different low

balance accounts. Of course, for many social security beneficiaries and lowincome

earners, maintaining even a single account balance above the threshold

may not be possible. But even those persons, by ensuring that across-the-counter

withdrawals are limited to some 6 to 8 per month, will be able to minimise bank

fees.

After taking into account the exemptions currently offered by most banks, the

groups most likely to be disadvantaged by bank fees and charges are those social

security beneficiaries not exempted as under 21 years or in receipt of aged (and

war veterans) pensions, and employed low-income earners. To incur fees and

charges, the individuals must maintain low account balances (below $100-$500)

and rely on over-the-counter withdrawals.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

32

Transaction Amount Balance Fee

Week 1 Opening balance $100

Social security benefit (adults) $536 $636

Withdrawal (counter) $300 $336

Withdrawal (counter) $100 $236

Week 2 Deposit $50 $286

Withdrawal (ATM) $100 $186

Family payment (children) $178 $364

Withdrawal (counter) $100 $264

Week 3 Withdrawal (ATM) $200 $64

Social security benefit (adults) $536 $600

Withdrawal (counter) $300 $300

Withdrawal (counter) $100 $200

Week 4 Withdrawal (counter) $100 $100

Family payment (children) $178 $278

Withdrawal (counter) $200 $78 $1.50

Withdrawal (ATM) $50 $28

Monthly fee as balance went below $500 $2.00

Annual fees and charges on equivalent transactions $42.00

Annual flow through account $18 564

Fees as a percentage of flow 0.25%

a. The example assumes a family of 2 adults with 2 dependent children under the age of 13.

The family uses only one account which can be accessed both by card (with an ATM) and overthe-

counter.

b. There would also be some government fees. Financial institutions duty levied by the States

at 6 cents per $100 of credits would amount to 90 cents each month. In this example, there

would be no bank account debits tax because the account is assumed not to have a cheque facility.

Box 6.1 illustrates how fees and charges could affect a family solely dependent

on social security payments. In this illustrative example, the family receives a

social security payment each fortnight of $536 credited to their bank account, and

a family payment (for two children under 13 years of age) each fortnight. These

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

33

this suggests a minimum requirement might be four withdrawals per month. The

example makes provision for 10 withdrawals during the month, of which 3 are by

automatic teller machine and incur no fee. Of the other 7 withdrawals which are

seventh (and any subsequent withdrawals) in the month incurs a fee, in this case

$1.50. In addition, there is a $2.00 account keeping fee because the balance in

the account fell below the $500 threshold set by this bank. If the account were

operated in this way throughout the year, fees would amount to $42 for the whole

If the family could manage with just one less over-the-counter transaction, or

make more use of ATM withdrawals, total fees could be reduced to $2 each

Given the number of exemptions currently available and the methods available

for avoidance of fees, the number of people adversely affected to a significant

extent would be low.

Nevertheless, bank fees and charges are a relatively new phenomenon in

Australia. As noted, it is likely that they might be increased progressively and

their incidence widened over time as banks gather better information on what it

costs them to provide particular services. In this regard, it is noted that fees and

charges are significantly higher in some other countries than they are in Australia.

A significant increase in the extent and impact of fees on disadvantaged groups

may prompt government to intervene on equity grounds. Some options are

briefly discussed below.

Governments generally recognise the need to provide for persons who are

genuinely in need of assistance, due to low levels of income, or physical or

mental limitations. This section considers the merits of different possible forms

of government intervention.

If social security beneficiaries were given the option of receiving payments by

cheque, then they would have greater flexibility in managing their funds to

minimise fees and charges. Such people would no longer be compelled to

maintain a banking account, but realistically most people would see it as

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

34

desirable to have a banking account for reasons of security and good management.

Payment by cheque would not be costless to the Commonwealth Government.

The introduction some years ago of electronic payment of social security benefits

provided considerable savings to the Government, which would be partly lost as

people opted to receive cheques. Problems with fraud control could be

encountered. In addition, the introduction by banks of charges for the cashing of

such cheques could not be ruled out.

The fiscal cost of making social security payments by cheque could be mitigated

by limiting this option to selected beneficiaries. For example, of Australia’s

889 000 unemployment beneficiaries in 1993, about 280 000 received an

additional allowance for rental assistance (DSS 1993, p. 272) and they appear to

be in particularly pressing financial circumstances. Such targeting, however,

runs the risk of discriminating against people in even more difficult circumstances.

The PSA would have to give close consideration to the cost and relative merit of

paying social security by cheque to particular groups before making any such

recommendations. The principal difficulty would be to target those not already

exempted from fees and charges by the banks.

There are other options available to the Commonwealth Government for assisting

social security beneficiaries and low-income earners to cope with bank fees and

charges. These include income compensation, banking vouchers and general

subsidies to bank consumers.

Access to banking services among the financially disadvantaged in the

community could adequately be provided for by increases in social security

payments designed to compensate for increases in fees and charges. (Current

indexation arrangements are unlikely to provide more than partial compensation).

Such a system would also not restrict consumer choice.

The difficulty would be to avoid compensating those already exempted by the

banks. Nor does this approach assist low-income earners not in receipt of a

social security benefit. Further, the banks would face an incentive to cease or

narrow exemptions with the expectation that the government would pick up the tab.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

35

Voucher systems could be used to implement in-kind transfers while

incorporating some of the desirable features of cash transfers such as retaining an

element of consumption choice, transparency and non-interference in the

commercial activities of enterprises. As consumers retain an element of choice

between providers of banking services, competition between banks would not be impaired.

Vouchers may not be a practicable option for retail banking, because the

frequency with which retail banking services are used is likely to make the

administration of a voucher system far too costly.

A general subsidy would ensure that on some basic banking products no fees or

charges would be levied. The program could be paid for by the banks through

cross-subsidies or by the Commonwealth Government. But such general

subsidies are an ineffective means of income support for those genuinely in need

because they do not distinguish between different users of banking services.

Those who could afford to pay for banking services would also receive the subsidy.

Another option would be for government to introduce regulatory requirements on

the banks to provide services at reduced fees or free of charge to disadvantaged groups.

If the banks were required to bear the costs of operating such accounts, that

would be a form of taxation on them. There would also be efficiency losses

resulting from cross-subsidies, and increased net interest rate margins are likely

to emerge (see Section 5.1).

An alternative would be for the Commonwealth Government to fund the

program, via contracts with banks to provide fee-free or concessional accounts to

specified groups of low-income earners (equivalent to a “community service

obligation”). Banks could tender for the right to operate such accounts.

There are clear benefits associated with Commonwealth funding, as opposed to

bank funding, of such a program:

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

36

having to worry about cross-subsidising underpriced accounts. Distortions

due to inefficient resource allocation would be minimised;

encourage more thorough assessment of the policies, because taxpayers and

policy makers could clearly identify and assess the cost of such a social policy; and

government department and that the bank is accountable only for operations.

There are, however, some problems associated with the provision of fee-free

basic banking services:

encourage overuse of basic banking facilities by some consumers; and

imposition of a “community service obligation” would provide them with an

incentive to phase out their voluntary exemptions and pressure government to

extend its subsidy to those persons affected. This might result ultimately in

security recipients and low-income earners.

At this stage, the ORR cannot find a compelling case for government intervention

on equity grounds. This is because banks currently exempt from fees and charges

most potentially disadvantaged groups within the community. In addition, there

are several avenues open to consumers to minimise or eliminate the impact of the

current structure of fees and charges. Apart from its impact on efficiency,

government intervention would also reduce the incentive for banks to provide

exemptions from fees and charges.

The structure of bank fees and charges could change in the future. In that case,

the argument for government intervention may become stronger. However, any

such intervention would not be without cost. On balance, the ORR considers

that, if the government chose to intervene, the most efficient approach would take

the form of a “community service obligation” on banks to provide specific groups

of customers with fee-free basic banking services, with the Commonwealth

Government reimbursing the banks for the costs of doing so.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

37

Administration.

Committee on Banking, Finance and Public Administration.

Ackland, R. and Harper, I. 1992, “Financial deregulation in Australia: Boon or

Sydney, pp. 45-71.

Areeda, P. 1984, “Introduction to Antitrust Economics”, in Fox, E. and

Economics, Section of Antitrust Law of the American Bar Association, Chicago,

pp. 45-59.

pp. 1-76

Becker, G. 1993, “Nobel Lecture: The Economic Way of Looking at Behaviour”,

Journal of Political Economy, Vol. 101 (3), pp. 385-409.

Harvard University Press, Cambridge.

Representatives Standing Committee on Banking, Finance and Public

Administration, AGPS, Canberra.

Canberra.

Sydney, June, pp. 9-17.

Harper, I. and Scheit, T. 1992, “The Effects of Financial Market Deregulation on

271.

SUBMISSION BY THE OFFICE OF REGULATION REVIEW

38

Jarrell, G. and Peltzman, S. 1985, “The Impact of Product Recalls on the Wealth

Klein, B. 1993, “Market Power in Antitrust: Economic Analysis After Kodak”, 3

Supreme Court Economic Review, p. 43ff.

Representatives Standing Committee on Finance and Public Administration,

AGPS, Canberra.

Milbourne, R. and Cumberworth, M. 1990, “Australian Banking Performance in

an Era of Deregulation: An Untold Story?”, School of Economics, University of

New South Wales, mimeo.

Milbourne, R. and Cumberworth, M. 1991, “Australian Banking Performance in

Practices Act 1974, Submission to the Treasury, February.

and Economics, Vol. 24, December, pp. 403-459.

Rates, Report No. 45, PSA, Melbourne.

Transaction Accounts by Banks and Other Financial Institutions, February.

pp. 1-6.

Scherer, F. M. 1987, “Market Structure”, in Eatwell, J. Milgate, M. and

Performance, 3rd edition, Houghton Miffin Company, Boston.

Shapiro, C. 1995, “Aftermarkets and consumer welfare: making sense of

Kodak”, Antitrust Law Journal, Vol. 63, pp. 483-511.

Vol. LXIX (3), June, pp. 213-225.