Supplementary Submission to the Financial System Inquiry 1. The Next Ten Years


1. The Inquiry is being asked to make recommendations about how financial regulation should adjust to cope with changes to the financial system over the next decade. Submissions should be aimed at helping the Inquiry to make those recommendations, so it is not unreasonable to expect them to provide some outline of how they think the financial system will evolve over the next decade. Even though no-one can forecast ahead this far with any degree of accuracy, the key forecasts (or assumptions) should be made public. In this Chapter, we set out some assumptions about developments over the next decade which underlie the RBA's view on financial regulation.

The Future Role of Banking

2. A good illustration of the need to be explicit about future developments is provided by the contrasting views that are often heard on the future of banks.[1] The two widely differing views could be summarised as follows:

  1. Banks are too powerful and too profitable, a position they are able to maintain because they can keep competitors at bay by their own efforts, and with the help of regulations which discriminate against potential competitors.
  2. Banks are a threatened species. New technology and competition from non-banks such as software firms, communications companies, retailers and specialised finance companies are taking away the most profitable parts of banking, and leaving existing banks with the unprofitable residue.

3. These contrasting views about the future of banking lead to two diametrically opposite sets of policy implications. Proponents of proposition (i) would presume that there are anti-competitive constraints on entry to banking and seek to dismantle them, i.e. compared with its present position, they would tilt the playing field in favour of the non-bank competitors. Proponents of proposition (ii), on the other hand, would say there is no need to because the non-banks will prevail anyhow. Instead, they would turn their attention to identifying the risks developing in the new growth areas and putting in place regulations to cope with them.

4. There is more than a grain of truth in both propositions (i) and (ii), but it is not necessary to opt for either extreme; they could be reconciled either by saying the truth was in the middle, or by saying that proposition (i) describes the past, while proposition (ii) describes the future. Whatever way they are reconciled, it can only be done by spelling out a view of how the next decade will develop.

5. The remainder of this Chapter attempts to sketch an outline of how the financial sector could unfold over the next decade. Not surprisingly, given the RBA's responsibilities, it concentrates on two questions. First, how important will commercial banking be in ten years time, i.e. how big will banks' balance sheets and the public's holdings of bank deposits be relative to other financial institutions and to the economy as a whole? Second, how will the developments of the next decade affect the ability of the financial system to withstand shocks, i.e. will aggregate risk rise or fall?

Some Possible Trends

(a) Simple projections

6. The best simple summary of recent trends in the relative importance of banking in Australia is given by Diagram 1. It shows that over the past 30 years, bank deposits have increased slightly faster than GDP and so the ratio of bank deposits to GDP has risen from 49 per cent in 1965 to 63 per cent in 1996. Other financial assets, however, have risen more strongly during this period, so total financial assets to GDP has shown a much bigger rise than deposits to GDP. As a result, the ratio of deposits to the assets of all financial institutions has fallen.

7. In Diagram 1, the greatest divergence between the growth of all financial assets and the growth of deposits occurred in the second half of the period, i.e. since about 1980. In that period of 16 years, the assets of all financial institutions have grown at an annual rate of 13.5 per cent compared with growth of 11 per cent for deposits.[2] If this continued over the next decade, deposits would be 23.4 per cent of total financial assets in 2006, compared with 29.2 per cent at present. This is shown as outcome A in Diagram 2.

8. However, most of this divergence between growth in deposits and growth in financial assets occurred in the 1980s when rapidly rising asset prices pushed up the market value of the assets held in superannuation funds, life offices and unit trusts. If the same calculation was repeated using the past eight years as the basis of comparison, the growth of deposits and all financial system assets is very similar, implying that there is no trend reduction in the ratio of deposits to all financial assets. This is shown as point B on Diagram 2.

9. It is not unreasonable to expect some downward trend in the ratio of deposits to all financial assets. The main explanation for the downward trend is that as people become wealthier, they choose to hold more of their financial wealth in assets that offer higher returns, even though they involve higher risk. In recent years in Australia, this has been supplemented by an element of compulsion as mandatory superannuation has become more widespread.

10. The long-term decline in the ratio of deposits to total assets is common to most countries, but it is most pronounced in the United States, and least pronounced in European countries. In the United States, the ratio is now down to 18 per cent, but in Germany it is 45 per cent, in France 42 per cent, and in Canada 36 per cent. The extreme position of the United States is largely a result of the heavy restrictions and prohibitions that have been placed on US banks over the years which have kept them out of many activities that banks take for granted in other countries. Some of these regulatory restrictions are spelt out in later parts of this Chapter.

11. In summary, our best guess is that in ten years time, bank deposits will have grown substantially in nominal terms, but their size relative to all financial assets could have fallen by between zero and six percentage points. Even if the outcome was at the bottom of this range, they would still comprise a substantial proportion of the Australian public's holdings of financial assets. This suggests that the Australian public would still choose to hold a substantial proportion of their financial assets as a deposit-type instrument, i.e. one where there is a binding contract with a bank for the return of the full nominal principal plus interest. There has been no significant tendency for ‘blurring’ to override this relatively stable relationship (see later).

12. A simple extrapolation of trend, as given above, is only one way of approaching this subject. Another approach is to look at the various factors which have contributed to this trend, and to ask whether they are likely to cause the trend to speed up or slow down. The following sections attempt to do this by identifying each factor, and assessing whether it is likely to have a continuing evolutionary effect, or whether it could lead to a dramatic change in trend.

(b) Technological change and financial innovation

13. Chapter 2 of this submission deals with this topic in some detail. The main effects that we can foresee will be within banks themselves, although there will also be some influence via the entry of non-bank competitors (see next section).

14. There can be little doubt that the way banks deliver existing loan, transaction and deposit products to their customers will be greatly transformed over the next decade. The role of the branch and face-to-face contact will diminish and new channels such as the telephone and the computer will increase. Where physical contact with the bank is involved, it will increasingly be through supermarkets, kiosks, ATMs and mobile banking. While these will have enormous commercial significance, with big penalties for banks that are slow to exploit the channels which consumers prefer, they are not likely to lead to a sharp fall in the relative size of the banking sector. In addition, the prudential implications are not great; the riskiness of banks is not likely to be greatly affected by these changes in delivery mechanism.

15. If financial innovation led to non-bank institutions providing deposit-type instruments on a large scale, this would have serious implications for the foregoing analysis. The evidence to date suggests that this is unlikely. Banks' share of deposit-type instruments has remained at a relatively stable 95 per cent over the 1990s (data limitations prevent longer-term comparisons). The only significant non-bank ‘deposits’ were insurance bonds issued by insurance companies, which were essentially term deposits with a small insurance component, and approved deposit funds. These owed their position to a tax concession rather than intrinsic product advantage. Cash management trusts, which are often incorrectly viewed as deposits, have failed to increase their market share over the past decade. While it is possible that non-banks will offer alternative forms of deposits, it is more likely that they will concentrate their attention on the parts of the market whose share is growing, such as payments and investment products. If they need to offer a deposit facility as a component of a broader product, commercial considerations usually argue in favour of an agency agreement with an existing bank.

16. The greater use of technologies such as the Internet and other devices which permit rapid searches and comparisons of bank products will have significant commercial implications. By increasing information and reducing switching costs, they should increase competition and enable customers to ‘play off’ banks against each other on a product-by-product basis, or access non-bank providers. This will increase competitive pressures within the industry, and hence tend to increase risks for banks, although it is unlikely to alter the risk characteristics of individual products.

(c) Globalisation

17. Globalisation has affected all Australian financial markets – shares, bonds, foreign exchange and futures – and Australian markets are now some of the most open in the world. A major beneficiary of this has been the Australian corporate sector which has easy access to foreign banks and capital markets (including the Euro $A market).

18. Further globalisation and improved communication systems will probably open up similar opportunities to access foreign banks to the Australian household sector. In this case, it is likely that consumers could hold more deposits with foreign banks and do more transactions through them than at present. This has little prudential significance as long as the banks in question are supervised entities of good repute. It has been suggested that improved communication channels, such as the Internet, may provide opportunities for consumers to access institutions of dubious quality or from unsupervised tax havens. In practice, however, these institutions would probably have great difficulty in competing with established international banks offering essentially the same services. There may be a need to protect the gullible, but this is a matter of consumer education rather than prudential supervision.

19. Like all industries, there will probably also be increased foreign ownership of Australian financial institutions, and further offshore diversification by Australian institutions. Foreign ownership, including by acquisition, is already pronounced in insurance, funds management and stockbroking. It is a good deal lower in banking, and until Bank of Scotland's recent purchase of BankWest and Rabobank's purchase of Primary Industry Bank of Australia, foreign takeovers of Australian banks were unknown. In the future, however, there is reason to expect that such events could become more common. Again, it is hard to see new prudential implications.

(d) Competition from non-banks

20. In lending to the corporate customer, commercial banks have always faced competition from investment banks, finance companies and securities firms. These competitors have played a major role in the development of capital markets and the process of securitisation. In recent years, there has been increased competition from non-banks for the household customer and this trend should continue. Some of the main areas are set out below.

The growth of finance companies

21. In the United States, the growth of finance companies, such as General Motors Acceptance Corporation, GE Capital, etc. has played a large and well-documented role in the declining share of US banks. Finance companies' loans to business are now equivalent to 63 per cent of banks' loans to this sector, compared with 31 per cent in 1980. While this is a striking development, it appears to be largely confined to the United States. In Australia, finance companies' loans to business represent only 18 per cent of business lending by banks, down from the 38 per cent they represented in 1980.

22. In this area, and in others such as securitisation (see later), it is unwise to generalise from US examples because US banks have faced higher regulatory restrictions than those in other countries. As well as restrictions that kept them out of insurance and the securities business (the Glass-Steagall Act), there are other regulations which have impeded their ability to compete in their traditional field of lending. For example, the McFadden Act restricted interstate branching, and the Community Reinvestment Act limits regional diversification even within States.


23. The capital markets have always been a strong competitor for banks. Securitisation has taken this a step further by allowing loans, which formerly would have been on banks' balance sheets, to be packaged together and sold to investors such as superannuation funds and life offices. The type of loan which is most suitable for securitisation is the residential mortgage, and in Australia mortgage originators have recently gained about 10 per cent of the new mortgage market (equivalent to 1.5 per cent of all financial assets). This has prompted people to ask how much further it could go. At one extreme, is the US example where two-thirds of mortgages are financed by securitisation (although many of these are originated by banks). That is possible here, but unlikely because the secondary mortgage market in Australia has not received the same degree of official encouragement as in the United States.[3] At the other extreme, mortgage origination could go the same way as cash management trusts. After achieving a market share of about 1 per cent in 1983, the share of cash management trusts remained essentially stable thereafter as banks retaliated with similar products. The contribution of cash management trusts to the Australian financial system was mainly to help bank customers get a better deal from their bank, rather than to take those customers away from banks. The recent reduction in the interest rate margin on bank mortgages suggests that there may be some similarities between these two financial innovations.

24. The factor that could speed up securitisation in coming years would be the securitisation of new types of loans. The US experience suggests that credit card receivables might go down this path, but it is hard to see much beyond that. The ‘bread and butter’ of commercial banking is loans to small and medium sized businesses. This is where the biggest risks are, and where the biggest information problems lie. Despite some success in standardising credit scoring for this type of loan, no significant progress has been made in securitising them. It is reasonable to assume that these loans will remain on the books of commercial banks, and could become a larger share of their assets as easier-to-securitise assets are on-sold.

Payments system

25. Another area where non-bank competitors are entering is in the payments system. As explained in Chapter 2, there are various levels of the payments system. Some of these, such as the provision of credit-based payments instruments, have long been open to competition from non-banks such as charge card providers (Amex, Diners Club, etc.), but such competition has been limited until recently.[4] With new communications systems, and increasing use of electronic payments instruments, they could be joined by software companies, communications companies and others. This applies also to such possibilities as electronic cash or cyber cash. These should provide benefits to consumers without any significant change in risk. There is also nothing to stop these non-bank competitors entering the first tier of the payments system by providing credit, but if they offered conventional deposits, they would then be liable to regulation as banks. Such new entrants would, of course, be competing with banks of established reputation, who were capable of offering similar technical products either of their own design or in a joint venture. So their advent is unlikely to cut deeply into the market share of the existing banking sector.

(e) Banks versus banking

26. It is often claimed that banks will continue to thrive, but that banking will become smaller and less significant over time. That is, banking – the activity of raising deposits to fund loans which are kept on the balance sheet – will decline because of the various factors described in this Chapter, but banks will respond by moving into other profitable areas of financial activity. There is a lot of truth in this characterisation, but it is often overstated. Our forecast of the future of banking was presented earlier, when we put forward the view that the share of deposits in financial claims could decline moderately (by between zero and 6 per cent of total financial assets). This would still leave banking as a very important part of the economy.

27. Where there is most truth in the claim is that banks have diversified their sources of income. First, an increasing proportion of their assets is now funded from non-deposit sources such as onshore and offshore commercial borrowing. Diagram 3 shows that banks' total assets (as opposed to deposits) have remained relatively constant as a proportion of total financial system assets. Second, they earn a significant proportion of their income from fees, guarantees and trading profits. So it is certainly true that banks have been able to grow a lot faster than they would have if they still relied almost exclusively on their traditional staple of financial intermediation.

(f) The growth of derivatives

28. The rapid growth in the use of derivatives has been widely chronicled in Australia and elsewhere. Banks have played a major part in this in Australia and account for 80 per cent of the major exchange traded and OTC derivatives markets. This development is another example of how banks have moved into rapidly growing ‘off-balance sheet’ areas, and suggests that comparisons of the relative importance of types of financial institutions based on the size of their assets understates the importance of banks.

29. The net effect of the growth of derivatives on financial system risk is a much debated question. In principle, the use of derivatives should reduce aggregate risk as it allows institutions involuntarily incurring risks over which they have no control to redistribute them to others who are in a better position to handle them or disperse them further. In practice, it does not always work out this way because the ability of futures and options-based contracts to permit a rapid increase (or decrease) in risk can put excessive pressure on internal control systems. Recent collapses or near collapses of financial institutions (Barings and Daiwa) and trading institutions (Metallgesellschaft and Sumitomo) have illustrated this tendency. The best that can be said is that the net effect on system stability is still an open question.

Implications For the Riskiness of the Financial System

30. The foregoing analysis has argued that most technological change and other innovation is aimed at improving the method of delivery of existing financial products, and so has little implications for the risk involved in those products. However, the riskiness of the whole financial system is more than the average of the risks involved in the different products. The following paragraphs attempt to set out the main factors that will affect the stability of the financial system over the next decade.

31. The biggest single influence will be the underlying macro-economic environment, in particular whether the next decade will contain an asset price boom and bust. History shows that systemic financial crises usually coincide with asset price falls and associated recessions (or, in earlier times, depressions). There is reason to hope that if the present low-inflation environment can be maintained, the size of asset price booms and busts could be reduced, along with the variability of financial prices such as interest rates and exchange rates. On the other hand, the experience of Japan over the last decade, which had the lowest inflation in the OECD area, but probably the largest rise and fall in asset prices, is not reassuring. The current level of US equity prices also has many observers characterising it as a bubble, or at least evidence of irrational exuberance.

32. Another important influence will be the degree of competition, which all observers agree will become more intense. As competition increases in any industry, there are considerable benefits to consumers, both from lower prices and better products, but risks for businesses rise as profits become harder to earn. In the financial services sector generally, and in banking in particular, competition will bring down profits in areas where it is easy to standardise products. There will be an incentive for firms to move away from these areas into newer areas at the higher risk/higher return end of the spectrum. There is nothing inherently wrong with this as it will bring benefits to users of financial services, but it could place strains on financial institutions' ability to handle the changing environment and to price appropriately for the additional risk.

33. It has often been claimed that banks face an increasingly risky future.[5] The argument is that banks' assets will become more risky on average as they lose their housing loans through securitisation, and their top quality business loans because more and more firms can access the capital markets directly. As a result, they will depend more heavily on lesser quality business credits, where risks are greatest, and where opportunities for securitisation are virtually non-existent. There is no doubt that this has happened in the United States and it is the major explanation for the widespread fall in the credit ratings of US banks (often to levels well below their better customers). The extent to which it happens here is still uncertain, but the direction of movement seems to be clear in that banks' assets should become riskier on average.

34. There are also some developments which are making banks less vulnerable than in the past. The main one is their increasing capacity to diversify their businesses away from dependence on pure intermediation. We have already seen how banks have become the major players in Australia's bond, money, foreign exchange and derivatives markets. In addition, they play a large role and derive considerable income from providing guarantees and liquidity to markets such as the commercial paper market, which are often seen as competitors to banks. This trend should continue as banks are better placed than competitors because of their large capital bases (see Folkerts-Landau (1995) and Rajan (1996)). While these developments reduce risk because they allow diversification, it is not an unmixed blessing. As explained earlier, it also means banks are moving into newer activities, many of which involve considerable risk, particularly to the uninitiated.

35. It is also important to recognise the contribution that central banks and other supervisors have made to improving disclosure and encouraging the development of risk management techniques. Gray (1996) sets out the development of disclosure requirements in Australia over the past decade and shows how much more information is now provided to markets. With the help of this information, ratings agencies and the investment community more generally can keep a closer watch on financial institutions than in the 1980s. The better financial institutions, in turn, have improved their risk management techniques and these practices are now spreading more widely. Finally, the introduction of Real-Time Gross Settlement (RTGS) in virtually all developed economies, including Australia, will virtually eliminate settlement risk on high value transactions. This will be a great help in reducing collateral damage in the case of a bank failure, but it is only part of the picture. Systemic financial crises invariably result from credit risk, and RTGS does not solve that problem, or the problem of contagion.

36. The net effect of the various influences increasing and decreasing risks in the financial system is difficult to judge. Several well-known analysts of financial markets have recently claimed that they see risk increasing on average in coming years (see Henry Kaufman (1996) and Richard Dale (1996)). We are sceptical of some of these claims, and would fall back on the more defensible course of assuming that the level of risk will not be very different to what it was over the past decade. To the best of our knowledge, no experienced observer has gone further and claimed that risks are declining. To do so would constitute a leap of faith which would be an unwise foundation for building a system of prudential regulation.


37. The foregoing projections and discussion of likely influences over the next decade lead to two conclusions. First, although banking will probably decline slightly in relation to the total assets of the financial system, it will still represent a substantial part of that system. We also see no evidence to suggest that non-banks will make significant inroads into banks' domination of the provision of deposit-type instruments. Second, while there are a number of influences that will increase risks in the system, there are also a number that will reduce them. Our assumption is that the level of risk of the financial system, and of banks, will not be significantly different to what it was over the past decade.


Throughout this chapter, in the interest of simplicity, the term ‘bank’ refers to an institution that takes deposits from the public and holds loans on its balance sheet. In the Australian context, it encompasses banks, building societies and credit unions. [1]

Over this period, nominal GDP grew at 9 per cent per annum, so deposits growth exceeded GDP growth by 2 per cent per annum, and total financial assets growth exceeded GDP growth by 4½ per cent per annum. [2]

In the United States, the securitisation of mo rtgages is dominated by three government-sponsored entities. The Government National Mortgage Association (known as Ginnie Mae) provides government guarantees to securities issued to finance housing loans to disadvantaged groups. The Federal Home Loan Mortgage Corporation (Freddie Mac), and the Federal National Mortgage Association (Fannie Mae), were created by Act of Congress to promote the secondary mortgage market, but are privately owned. They package and securitise mortgage loans, and enjoy a range of advantages. For example, they are exempt from income tax and prospectus requirements; their paper is eligible for purchase by the Federal Reserve in open market operations and as collateral when banks borrow from the Fed discount window; and they have lines of credit from the US Treasury. According to a US Treasury report, investors believe that Federal sponsorship provides a de facto guarantee for these entities. [3]

There has, however, been an increase in competition within the group broadly defined as ‘banks’ in this Chapter, as building societies and credit unions have become active in non-cheque payments such as direct entry and credit/debit cards. [4]

Edwards (1993), Edwards and Mishkin (1995), etc. [5]