Submission to the Inquiry into the Australian Banking Industry C. Competition in Banking

The Results of Deregulation

21. Any evaluation of the results of deregulation should bear in mind the recentness of those changes – we have little more than half a decade of experience with the present system, after more than three decades with a tightly controlled financial environment. Furthermore, the period of the present system has involved a substantial ‘learning phase’ as decision making by participants has had to adjust to more market driven influences and less official direction. The past half decade or so has also witnessed other significant economic developments which, while not related directly to financial deregulation, have affected the behaviour of banks and their customers.
22. What was expected from financial deregulation at the time? Different groups no doubt expected different things but it was widely expected that:
    1. banks would regain market share;
    2. interest rates would be less volatile;
    3. bank credit would be more readily available and bank depositors would be better compensated for the use of their savings;
    4. banking would become more competitive and innovative, probably involving some reduction in profitability; and
    5. because banks would have more freedom and competitive pressures would be greater, they would be exposed to more risks.
23. Much of the remainder of this submission comments on the extent to which these expectations have been fulfilled; many of the issues here would appear to fall directly within the Terms of Reference of the Committee. The overall conclusion must be that there has been a significant increase in banking competition during the second half of the 1980s.

(a) Market share

24. The expectation that banks would regain market share has been fulfilled. From a low-point in 1983, when banks accounted for only about 40 per cent of the assets of all financial institutions, their share has risen to a little over 46 per cent. This has not returned them to anywhere near the degree of dominance they enjoyed in the immediate post-war period but no such return was expected. A large part of the increase in the banks' share has reflected the bringing back onto banks' own books of business that was formerly written by bank-owned finance companies and merchant banks. An additional factor has been the conversion of a number of permanent building societies into banks. Merchant banks gained market share in the early years of deregulation but lost much of these gains subsequently as imposts on the banks were reduced and some merchant banks chalked up substantial corporate losses.

(b) Interest rate volatility

25. Interest rates have fluctuated within wide limits (cash rates, for example, have ranged between 10 and 18 per cent since 1983) but in terms of day-to-day movements in interest rates, there has been a reduction in volatility.^[2] Sharp ‘credit crunches’, of the 1961 and 1974 variety, have been avoided as more of the work of monetary policy has been done by rising interest rates and less by credit rationing. For a variety of reasons, however, interest rates have probably acted more slowly in countering excess domestic demand pressures than was expected. Interest rates had to be kept at high levels for a considerable time in 1985/86 and again in 1989 before domestic demand slowed appreciably. Other factors – including expectations of sustained asset price rises – appear to have contributed to that situation. Notwithstanding the lags involved, however, monetary policy pursued through market operations has proved effective.
26. It is sometimes argued that the process of deregulation caused real interest rates to rise over the last decade. It is true that real interest rates have been significantly higher in the 1980s than in the 1970s, but this has been true for all major countries (see Table 2). The widespread use of controls in the 1970s meant that interest rates were slow to adjust to rising inflation; in fact, the catch-up did not occur until the 1980s. In addition, the demand for funds for private investment was much stronger in the 1980s for most countries while in many countries private savings rates declined.

(c) Availability of bank credit

27. Bank credit has been more freely available since direct controls over banks' interest rates and lending volumes, were removed. Table 3 shows the strong growth that occurred through the 1980s, with bank credit growing at an average rate of over 20 per cent. The fastest rate of growth was in the period from 1985 to 1989. During this time, non-bank credit did not slow by much, so that the net effect was to speed up the growth in the total provision of credit during these years. By sector, the fastest rate of growth occurred in the provision of credit to businesses.
28. In contrast to the regulated period, when the non-availability of credit was a common charge, many complaints during the deregulated phase have been to the effect that banks have provided too much credit. Certainly the growth of credit has far exceeded the rate of growth of nominal GDP, and the outstanding stock of debt as a ratio of GDP has risen, as has corporate leverage. It is fair to say that the increase in the availability of credit was greater than was foreseen – and banks would concede that they made many loans that they now regret. This is part of the learning phase for banks (and others) which is still underway.
29. Other factors, however, have been at work in generating this exceptionally high rate of growth of credit.^[3] In Australia, as in a number of other countries, business adapted to the inflationary pressures of the 1970s by pursuing strategies based increasingly on leveraged asset acquisition. Australian banks, to a large extent, accommodated this, but it is unlikely that they were the main initiating factor, nor were they the only credit providers to companies engaged in leveraged asset speculation; overseas banks and overseas holders of high-yielding (‘junk’) bonds were also prominent in many instances.

(d) Competition and profitability

30. Deregulation was expected to lead to an increase in competition in the banking industry, and probably involve some reduction in profitability in the process. There are many aspects to be examined here. This section of the submission examines competition in banking by considering, in turn, the concentration of the industry, trends in profitability, changes in interest rate margins and range of services.

(1) Concentration

31. A common starting point for studies of competition within an industry is to look at its degree of concentration – for example, the proportion of industry turnover accounted for by, say, the four or five largest firms. Industry turnover can be defined to include all banks, or it can be widened to include all financial intermediaries. The wider definition recognises that banks compete with building societies, finance companies, credit unions, and other institutions. In Australia, there has been a number of studies of industry concentration, but none specifically directed at the banking industry. Table 4 shows concentration ratios for a number of major Australian industries derived from a recent study by the Australian Bureau of Statistics; we have added figures for banks, which show the proportion of assets of all banks accounted for by the four largest banks.
32. Many industries in Australia have concentration ratios that are high by international standards; indeed, some major industries are near-monopolies. On the data shown in Table 4, banking comes roughly in the middle of the field. The concentration ratio in Australian banking, measured on this basis, rose from 66.9 per cent in 1978 to 79.1 per cent in 1983, following the mergers between the Bank of New South Wales and the Commercial Bank of Australia to form Westpac, and between the National Bank of Australia and the Commercial Banking Company of Sydney, and the absorption into ANZ of the Bank of Adelaide. The ratio has since fallen – to 68.5 per cent in 1988 and 66.9 per cent in 1990 – but will rise again when the State Bank of Victoria/Commonwealth Bank merger starts to reflect in the figures.
33. By international standards, the concentration of banking in Australia is not unusual. Apart from the United States, which has an extremely fragmented banking system of around 14,000 separate banks, virtually all other countries show a fair degree of concentration. For example, in the United Kingdom, Canada, Australia, New Zealand, the Netherlands and Sweden, the bulk of domestic banking business is accounted for by four or five large banks. Table 5 shows concentration ratios for 9 countries, where concentration is measured by the percentage of assets of all financial intermediaries held by the largest 3, 5 and 10 firms. Again Australia is in the middle of the field. (This ratio is lower than the one shown in Table 4 because its denominator is all financial intermediaries, rather than all banks.)

(2) Bank profitability Recent trends

34. One guide to whether an industry is competitive is the profitability of firms in that industry. Abnormally-high profits usually indicate a lack of competition, while normal or below-normal profits may indicate (assuming firms are efficient) that the industry is competitive.
35. Determining what is a ‘normal’, or appropriate, level of profits in an industry is a matter of judgment. A comparison often drawn, however, is with rates of return available on alternative investments. A widely-used benchmark is the interest rate on government bonds, which provides a measure of the risk-free rate of return on capital. Investors in shares look for a return above that because of the greater risk; the higher the risk, the greater the expected return needs to be to attract capital. Another benchmark is rates of return in other industries, although such comparisons need to take account of differences in risk across industries.
36. Bank profitability can be measured in a variety of ways. The most widely-accepted measure, and the one that can be compared most readily with other industries, is return on shareholders' funds. This is usually measured as net profit after tax as a percentage of shareholders' funds. Another measure is return on assets – i.e. net profits after tax as a percentage of total assets – but this measure can be affected by changes in the composition of banks' balance sheets and is also more difficult to compare with other industries.
37. Returns on shareholders' funds for the four major banks and yields on 10-year Commonwealth Government bonds are shown in Graph 2 for the period covering the 1970s and 1980s.^[4] The year-to-year variability in profits means that not too much emphasis should be placed on profits in any particular year, but conclusions can be drawn by looking at a run of years. The graph shows that:
    • average returns rose gradually over the 1970s, from a little over 10 per cent to about 16 per cent – this rise was more or less in line with movements in government bond yields but, on average, returns exceeded bond yields by 4 percentage points in the 1970s;
    • through the first half of the 1980s, returns on shareholders' funds were fairly steady, averaging 16 per cent – over this period bond yields rose and the margins of bank returns over bond yields fell to 2.5 percentage points on average;
    • returns fell sharply over 1985, 1986 and 1987 – both in absolute terms and relative to bond yields – following the progressive moves towards deregulation, including the licensing of new banks. Profitability rose in 1988 and 1989 due largely to the reduction in banks' costs of funds resulting from the ‘flight to quality’ by investors after the sharemarket crash of 1987. However, it again fell in 1990, as these effects passed and banks were burdened with large volumes of bad and non-performing loans. This followed the sharp expansion in their loan portfolios in earlier years.
38. On average in the second half of the 1980s, banks' profitability fell to a rate which was not very different from the government bond yield. The fact that banks were not able to earn a premium on the risk-free rate of return suggests strong competitive pressures. In the Bank's view, deregulation and foreign bank entry were major sources of the increased competitive pressure.

Factors affecting banks' profits

39. Profits reflect the difference between revenues and costs. The two main sources of revenue for banks are net interest income and non-interest income (e.g. fees for service). Costs can be divided into operating costs and costs of credit risk. Movements over the 1980s in these various components for the major banks are discussed below.

Net interest income

40. Net interest income of the major banks – the difference between interest charged on loans and interest paid on deposits – averaged 3.7 per cent of assets in the first half of the 1980s, but fell to 3.3 per cent in the second half. Several factors contributed to this fall (discussed in more detail below) but, importantly, over this period the margin between interest rates on loans and those on deposits narrowed.

Non-interest income

41. Non-interest income of banks (again measured in relation to assets) was slightly lower in the second half of the 1980s than in the first half (1.7 per cent and 1.8 per cent respectively). Although banks widened the range of services they provided to customers over the period, and greatly expanded the volume of some (such as bill finance), competition brought about significant reductions in the fees for many of these services. This was particularly noticeable, for example, in the fees banks charge for bill finance. Typically, acceptance fees for larger companies were 1.5 per cent in the early 1980s, but fell to 0.5 per cent by 1987.

Operating costs

42. This is another area where competition appears to have had a major impact, raising the level of banks' operational efficiency. Operating costs of the major banks averaged 3.9 per cent of assets in the first half of the decade, but declined to 3.2 per cent in the second half. This reduction was achieved by more efficient use of personnel (assets per employee have risen strongly) and by the introduction of new technology. It is reflected also in a fall in the ratio of operating costs to total income – this fell from 0.7 in the first half of the 1980s to 0.6 in the second half. The reduction in these ratios suggests that banks are now operating more efficiently than in the early 1980s.

Credit risk

43. In the first half of the 1980s, costs of bad debts averaged only about 0.2 per cent of assets. (The cost of non-performing loans – i.e. interest forgone – is taken into account in the measure of net interest income discussed above.) In recent years, however, and particularly over the past year, these costs have risen sharply; charges against profit for bad debts accounted for 0.5 per cent of assets per year over the period from 1986 to 1990, peaking at 0.9 per cent in 1990 – see Graph 3.
44. Some of the increase in bad debts over the past year or so results from the contraction in economic activity, and should be partly reversed as the economy picks up. However, a further large part of the increase reflects the recent fall in asset prices, after their rapid growth during most of the 1980s. Had these bad debts been foreseen, they should have been charged against profits in earlier years, in which case the apparent pick-up in profitability in 1988 and 1989 (see Graph 2) would not have occurred. In other words, there would have been a steady decline in the return on shareholders' funds in the second half of the 1980s, rather than the variations shown in the actual figures. Part of the rise in bad debt expenses above that prevailing in the first half of the 1980s might also reflect a structural shift by banks into higher-risk forms of lending.
45. Table 6 summarises the net impact on banks' profit margins of the various factors discussed above. Profits, measured as a percentage of assets, fell between the first and second half of the 1980s, from 0.8 per cent to 0.7 per cent. This fall occurred despite a substantial increase in the efficiency of banks, as indicated by the reduction in their operating costs. Part of the reduction in operating costs was absorbed by higher bad debt expenses, but most of it was passed on to customers through lower interest margins and fees – suggesting the operation of substantial competitive forces.

Comparison of bank profits with other rates of return

46. The decline in bank profits following deregulation occurred against the background of a slight increase in the general level of profitability of companies in Australia. As a result, while returns on shareholders' funds for all banks exceeded the average of other companies in the first half of the 1980s by an average of 6 percentage points, in the second half of the 1980s the margin was only 1 percentage point (and was negative on average in 1989 and 1990). For the major banks, the margin recently has averaged 3 percentage points, well down on that in the first half of the 1980s – see Table 7.
47. Graph 4 shows rates of return of companies listed on the Stock Exchange, classified by industry. In the first half of the decade, banks were among the most profitable companies listed on the Stock Exchange but, in the second half, they fell in the middle of the field.

(3) Banks' interest rates

48. Following deregulation, there have been two major developments in banks' interest rates:
    1. with the lifting of controls the average interest rate paid to depositors has risen substantially. In 1980, about 45 per cent of banks' deposits attracted an interest rate of less than 6 per cent. Today, despite a lower rate of inflation, about 13 per cent receive less than 6 per cent. In other words, depositors – other than those who, because of inertia or for other reasons, have elected to retain their savings in low interest accounts – now receive higher, more market-related, interest rates on their savings; and
    2. banks' interest margins have declined – i.e. the full extent of the increase in deposit rates has not been passed on to borrowers.
49. There are various ways of measuring changes in bank interest margins. One is to take the difference between a selected deposit rate and a selected loan rate. This approach, however, takes no account of changes in the relative shares of deposits raised at different rates or of changes in the mix of loans and other assets held by the banks. It does not allow, for instance, for the shift to higher cost deposits noted above, or for the fact that interest is now paid on a much higher proportion of bank deposits, including cheque accounts.
50. A better approach is to measure the net interest income of banks as a proportion of their assets. The figures shown in Table 6 are on this basis. As noted earlier, this ratio has declined in the post-deregulation period, reflecting the net result of several factors:
    • the removal of interest rate controls and competition among banks for deposits have tended to raise average interest rates paid by banks, while competition for lending business has limited the scope for banks to pass on these higher costs of funds to borrowers. Taken together, these factors have tended to produce a lower interest margin;
    • the growth of offshore business, where net interest earnings have been narrower than on domestic assets has worked in the same direction. Banks in most countries earn higher rates of return on their domestic business than on their overseas business, reflecting their greater competitive advantage at home;
    • also tending to depress the ratio has been the growth of non-interest bearing assets, such as bill acceptances, on which the banks earn a once-off return as acceptance fees rather than as interest; and
    • working in the opposite direction has been the reduction in the severity of regulations, particularly the Prime Assets Ratio and the Statutory Reserve Deposit arrangements, which required banks to hold low-interest assets. The replacement of these assets with assets earning higher interest rates – mainly loans – has tended to push up the ratio.
51. If we put aside offshore business and non-interest bearing assets, and look only at the difference between average interest rates paid on domestic deposits and average interest rates charged on domestic loans, a similar picture emerges. Information available to the Bank indicates that the average interest spread measured on this basis has declined by 0.4 percentage point in the second half of the 1980s, from 5.0 per cent to 4.6 per cent.
52. This does not mean that interest margins have been uniformly lower in the second half of the 1980s. At times, especially after the stockmarket crash in 1987, when the banks gained large inflows of low-interest deposits in a ‘flight to quality’, and again for a time in 1990 when banks were slow to reduce loan interest rates at a time of large bad debt losses, margins widened temporarily to around the average levels of the early 1980s. Those wider margins, however, were not sustained. Suggesting that community pressures and competitive forces were strong enough to prevent a permanent return to earlier levels.
53. Nor do the lower average margins in the second half of the 1980s mean that all depositors and borrowers have benefited equally. Some depositors – for example, those who, for whatever reasons, choose to hold deposits in low interest bearing accounts may not have benefited at all. It might be argued that competition for corporate lending was stronger in the period 1987–1989, leading to a presumption that corporate borrowers fared better than retail borrowers. This presumption is difficult to test because of the controlled interest rate loans remaining in banks' housing loan portfolios and the lack of data on which to make accurate comparisons. It seems clear, however, that margins narrowed for most, if not all, borrowers during the second half of the 1980s – by a greater degree for some than for others.

(4) Range of services

54. Under deregulation there has been a proliferation of products and services, with ‘new’ banks and non-banks prominent in this development. In addition, the number of alternative types of deposit account offered by most banks has expanded, allowing customers a wide choice of combinations of interest return, fee structure, and access to payments services.
55. Table 8 lists the main product innovations since 1985, and tentatively identifies categories of potential beneficiaries. In some cases, the innovations reflect the ‘unbundling’ of products and services which had formerly been combined; in other cases, they reflect services not available because of interest rate and exchange controls. More generally, they represent responses to perceived customer demand in a highly-competitive environment.

(5) Availability of Information

56. For bank customers to gain the benefits that flow from greater competition, they need to be properly informed about the services available, the interest rates to be paid or received, and all other fees and costs involved.
57. Banks were probably slower in responding in this regard than in most of their other responses to competition. In part this reflected the rapid expansion of services, the problems faced by their own officers in comprehending the various features of new products before being able to explain them to customers, and the costs involved in communicating with customers. For their part, customers were sometimes slow in seeking adequate detail in advance of signing up, and perhaps unwilling at times to admit that they did not fully understand the fine print.
58. After a slow start, a good deal of progress has been made in the past couple of years in setting standards of conduct, in the disclosure of information, and in the handling of customer complaints and disputes. Two specific developments have been:
    • implementation of the Code of Conduct for electronic funds transfer (EFT) transactions which details the rights and obligations of users and providers of EFT services – institutional compliance with the Code is now being monitored by the Australian Payments System Council; and
    • establishment of the Banking Industry Ombudsman in mid 1990.
59. The Bank believes there is scope for further improvement in standards of disclosure which it would like to see made in ways consistent with the flexible, adaptive operation of financial markets. Both directly and through its involvement with the Australian Payments System Council, the Bank is supporting initiatives to improve standards of services and protection for consumers. It is mindful that the costs of such initiatives be balanced against the benefits to be achieved given that, ultimately, the costs of customer protection are borne not by the banks but by the customers seeking to be protected.

(e) Entry to Banking

60. One test of competition is the extent to which new entrants are able to enter an industry. At present, entry to banking is restricted in a number of ways:
    • The Banks (Shareholdings) Act limits the degree of ownership by a single person, or company or associated group. A dominant shareholder poses the risk that a bank's deposits might be used for the benefit of such a shareholder (not itself subject to central bank supervision) or that public confidence in the bank would be compromised by business problems experienced by the dominant shareholder.
    • An applicant for a banking authority must satisfy the Bank and the Treasurer of the viability of the proposed bank in terms of capital availability, management competence, and other requirements.
    • Applicants must be joint stock companies. The main short-comings seen in co-operative or mutual organisations relate to the problems in establishing and maintaining a strong sense of ownership among members; the potential lack of effective discipline on management; and limited access to new capital.
61. Additional foreign banks are not envisaged under current policy. The most recent foreign bank entrants were the fifteen authorised over 1985 and 1986. Since then, foreign banks have been able to establish non-bank financial subsidiaries in Australia and a substantial number have done so. It is arguable whether a more open approach to foreign bank entry would add significantly to competition in the banking sector, or merely add to surplus capacity. The entry of additional foreign banks would hardly reduce competition in the banking sector but would probably not enhance it significantly either, unless foreign banks were permitted to take over or merge with a significant domestic bank. A non-competition argument in favour of more open entry is that such a policy change could make it easier for domestic banks to establish operations overseas, particularly in countries where reciprocal treatment is part of official policy.
62. Foreign banks, with a small number of ‘grandfathered’ exceptions, have been required to establish in Australia as locally incorporated subsidiaries, rather than as branches of the parent bank. Some foreign banks argue that this adds to their costs and limits their capacity to compete effectively. They argue that branches would be able to operate on the basis of the parent's total capital base, giving more effective access to wholesale banking opportunities. The contrary arguments, which helped to determine the present policy, relate to the capacity of the Australian authorities to supervise a bank that is not established under, and controlled by a board of directors subject to, local legislation; and to the capacity of authorities in other countries to determine the behaviour of a bank operating as a branch in Australia. The task of protecting local depositors might also be more complex if a branch is involved. This issue is under discussion within the Bank, and between the Bank and the Government. Some foreign banks have argued that their non-bank financial subsidiaries in Australia should also be able to operate as a branch of the parent bank. The Reserve Bank does not favour this course, basically because any such institution, bearing the name of the parent bank, would itself be seen as a bank, although legally and in other ways this would not be the case.

(f) Increase in risk

63. An increase in risk was an expected feature of a deregulated banking market, for a number of reasons, including:
    • a reduction in the previous incentive to lend only to the lowest-risk borrowers after interest rate ceilings were removed;
    • increased competition encouraged banks to expand their activities into newer areas in an effort to maintain or increase their market share;
    • greater pressure on banks' managements to make decisions previously made or heavily influenced by the government, e.g. how to price deposits and loans, how to assess and price risk;
    • a reduction in the proportion of banks' funds held compulsorily in government securities or deposits at the Reserve Bank, with more held as loans to the public; and
    • the spread of operations to other countries in a variety of currencies.
64. Coming to terms with this increase in risk is at the centre of the on-going learning phase of deregulation for the banks. The Reserve Bank's response can be seen in the introduction of formal prudential controls; these are detailed in Part D of the Submission.

Footnotes

R.G. Trevor and S.G. Donald, ‘Exchange Rate Regimes and the Volatility of Financial Prices: The Australian Case’, Economic Record Supplement, 1986, pp 58–68. [2]

See I.J. Macfarlane, ‘Money, Credit and the Demand for Debt’, Reserve Bank Bulletin, May 1989 and ‘Credit and Debt: Part II’, ibid., May 1990. [3]

Figures for all banks show similar movements, although the average level is lower. [4]