RBA Annual Conference – 1996 Regulatory Policy Issues in Australia Graeme Thompson

1. Introduction

The pace of development in the Australian financial system over the past decade and a half has been dramatic. The size of the sector has more than doubled in real terms, while the range and sophistication of products and instruments has expanded rapidly. The use of complex products such as derivatives has grown apace. Technological advances are revolutionising the delivery of financial services and are making available much more sophisticated alternatives in the payments system. Greater sophistication in financial arrangements is also allowing services to be offered in new ways and by new players. In Australia, as elsewhere, a trend for funds-management vehicles to grow faster than the traditional institutions is evident, as is the related growth of capital markets at the expense of intermediaries. Consequently, competitive pressures on established institutions, including the banks, are intensifying. While professional financial markets have been globalised for some years, modern communications technology offers the prospect of a more globalised market for retail finance.

Some of these trends, which are explored in greater detail in the paper by Edey and Gray in this volume,[1] have raised important questions about Australia's financial regulatory structure and for the framework of bank supervision. This paper seeks to address some of these. Section 2 considers the broad objectives of financial regulation in Australia. The following sections discuss some of the key regulatory issues.

2. The Objectives and Structure of Financial Regulation in Australia

Although the features of financial systems vary from country to country, depending on their stage of economic development and the structure and philosophies of government, it is possible to identify common themes or objectives underlying financial regulation. Broadly speaking, regulation is justified on three grounds:

  • a concern with the stability of the financial system;
  • a desire to protect the interests of users of financial services in situations where information concerning the characteristics of products, or the riskiness of institutions offering them, is hard to assess; and
  • the need to encourage appropriate levels of competition and efficiency in markets.[2]

Ultimately, the success of a financial system is measured by its capacity to facilitate the nation's long-term economic growth and prosperity. And the success of any system of financial regulation depends on its achieving those stability and other desirable characteristics in the financial system at reasonable cost in terms of moral hazard risk[3] and restraint on institutional flexibility and innovation.

It is useful to classify financial regulation in Australia according to its various types and objectives. Competition policy is left aside, since there need be no unique competition policy for the financial sector.

Prudential supervision, which is the main subject of this paper, may be defined as supervision directed at institutional solvency, and is exercised over banks, building societies, credit unions, life insurance, general insurance offices and some superannuation funds – entities which have obligations with a promised minimum value. The promised value of such liabilities will, of course, increase over time, as interest is credited to deposit accounts or bonuses added to insurance amounts. The agencies which conduct such supervision are the Reserve Bank (RBA) in respect of banks; the Insurance and Superannuation Commission (ISC) for insurance companies and defined-benefit superannuation; and the Australian Financial Institutions Commission (AFIC) for building societies and credit unions. These agencies, together with the Australian Securities Commission (ASC), are represented on the Council of Financial Supervisors, established in 1992 with the main objective of promoting effective liaison and co-ordination among the regulatory agencies.[4]

Other regulation and supervision focuses not on the viability of legal entities, but on the products (or services) offered and the competence of those offering (or advising about) them and the integrity of markets in which products are traded. Several related objectives can be identified under this heading of product regulation. One is to ensure that information disclosed by providers is sufficient for investors to make well-based decisions (which may of course include a decision to invest in a highly risky venture), with the ultimate objective of promoting efficiency in financial markets. Much of the ASC's regulation in relation to managed funds and securities traders (as well as non-financial corporations) has this goal. There are also regulations dealing with the competence and integrity of investment trustees and managers. Another area of product regulation is concerned with fair treatment of consumers – ensuring that they are properly informed about the conditions of contracts, including charges; providing for avenues for complaint and redress when disputes arise; and so on. Sometimes it is required that information be provided in a way which facilitates comparison of competing products or services. Examples of such regulation include the ISC's disclosure standards for life insurance policies, the uniform consumer credit laws and the code of practice agreed between the banks and the Government. Distinctions are often made between retail and professional customers.

A grey area is the oversight of accumulation superannuation funds, where the ISC exercises quasi-prudential supervision in seeking to ensure that trustees' investment strategies are not excessively risky; for instance, it requires asset portfolios to be diversified and has recently promulgated standards for risk-management systems governing the use of derivatives. This role has arisen because Government policies which promote saving in the form of superannuation give rise to community expectation that funds will be managed responsibly. Such supervision does not, of course, have the objective of institutional viability or solvency.

This classification system is useful when considering how financial regulation might be organised. Prudential supervision is about risk management, while product regulation is largely about standards of service and quality of information.[5] Consequently, the skills and knowledge required by the relevant regulators are different. Moreover, as long as financial products, such as home loans, are not uniquely identified with particular institutional groups, the relevant product regulator will necessarily deal with several groups. As financial services are provided by a widening range of institutions, product regulators, in the interests of competitive equity, need to expand their purview.

Prudential supervisors, on the other hand, must be concerned with a financial institution as a whole – its deposit-taking, lending, investment in liquids, capital structure and so on – since risks to its solvency may arise in any area.

Inevitably there will be overlaps in regulatory authority where a prudentially supervised institution offers services which are subject also to product regulation, such as information disclosure. In principle, there need be no inconsistency between the two, but in practice issues can arise which require some co-ordination among the respective agencies.[6]

3. Some Key Regulatory Issues

Structural changes in the Australian financial system raise many inter-related policy issues, four of which are discussed in the following sections:

  • how prudential supervision should be organised;
  • the particular question of the central bank's role;
  • the competitive impacts of bank supervision, which will come under closer scrutiny as technological and other changes increase the range of potential players in the system; and
  • evolution in bank supervision.

This discussion touches only tangentially on supervisory issues associated with the increased globalisation of finance.[7]

4. Organisation of Supervision

The present organisation of financial supervision and regulation has been under scrutiny recently, with the thrust of much comment being toward combining the existing agencies into a smaller number. The main justifications advanced for this rely on ‘blurring of the distinctions’ among the main traditional groups of financial institutions, as well as generalised concerns about the ‘burden of regulation’ which might be reduced if there were fewer agencies.

One could, of course, argue that institutional arrangements are considerably less important than ensuring that the objectives of supervision are appropriate, and that it is conducted competently with full regard for costs and benefits. But, how supervision is organised can affect its efficiency and cost, and have competitive effects. Capable supervisory resources will be scarce, and there might be considerable diseconomies if identical supervisory tasks are spread over more agencies than necessary. On the other hand, of course, a monolithic supervisory structure might bring about inefficiencies through stifling innovation and forcing financial activities into the one mould. (US debate on regulatory organisation often goes as far as extolling the virtues of competition among regulators of similar financial activities.)

The most extreme position is that all regulation and supervision for the financial system should be placed under control of a single regulatory authority. This model envisages an agency covering not only prudential supervision but consumer protection, market conduct, competition and so on. A major drawback would be that the range of tasks falling under the all-encompassing agency would be exceedingly broad – from issues associated with conduct in financial markets (dealer practices, market conventions), consumer codes and protection, legal questions associated with disclosure and the like, through to the complex and often technical issues associated with risk management within financial institutions. Clearly, the training and skills required to carry out these functions are very diverse, and bringing them into the one agency is unlikely to result in any efficiencies. The differences in objectives and cultures would produce an institution which was difficult to manage and unlikely to be clearly focussed on the various tasks for which it had responsibility. Further, some of these responsibilities, for example, competition, should be conducted on an economy-wide basis and it would be inefficient to have an agency specifically concerned with the financial system also dealing with them.

The remainder of this section considers the case for combining prudential supervisory functions. As indicated earlier, there are only three important prudential supervisory agencies in Australia – the RBA, the AFIC and the ISC.[8]

The first justification, in principle, for combining agencies could be that present arrangements have individual financial institutions (as opposed to conglomerates comprising two or more institutions) being supervised by more than one agency, leading to overlaps, conflicts and inefficiencies which could be resolved by collapsing these agencies into one jurisdiction. This argument has no force in Australia, where the statutory responsibilities of the prudential supervisory agencies are clearly defined, and no individual institution is subject to oversight by more than one. Institutions will, as noted earlier, be subject to both prudential supervision and product regulation. The relevant agencies should aim to co-ordinate their requirements to resolve any inconsistencies but, given their different objectives, overlaps should not be a major problem.

A second argument for combining agencies is that the current institutional groupings on which prudential supervision is based are no longer meaningful, and that all financial institutions are converging to one standard model or, at least, to very similar ones. The evidence for this contention is weak. Notwithstanding the substantial innovations of recent years, it remains possible to identify two broad classes of financial institutions, distinguished primarily by the nature of their contracts with those supplying funds to them.

The first category is firms which offer a ‘capital-guaranteed’ or ‘capital-backed’ product to savers – that is, which promise to repay the initial investment or some other fixed sum, and whose promise is backed by the holding of capital. These include deposit-takers (or intermediaries, in Edey and Gray's discussion) and insurance companies (for some products). The purposes to which institutions put the money raised from the public are conventionally restricted, and capital is required to be held as a buffer against losses which could threaten the repayment of funds. Supervision is justified on the grounds that it is inefficient for households and others to have to make judgments in the conduct of their daily affairs about the health of various complex financial institutions.

The second category comprises firms which offer to manage investors' money on a best endeavours basis – sometimes with undertakings about ‘capital stability’. These are the various funds-management vehicles, including accumulation-based superannuation.

This basic two-way classification is likely to be durable in the face of financial system change because there are severe practical difficulties in the one legal entity offering both types of product. There has, of course, been some practical blurring in the sense that some investment managers, through their choice of assets, can offer products with similar characteristics to bank deposits. The main example would be cash management trusts (introduced in 1981); another is some Approved Deposit Funds which have certain characteristics similar to term deposits. The legal form and characteristics of such products, however, remain distinct from capital-backed deposits.

The main significance of the classification is that only institutions in the first category are subject to prudential supervision, since only they may become insolvent. Funds-management institutions are (with the exception of accumulation superannuation funds) subject only to forms of product regulation – such as disclosure regimes and competency standards for trustees and managers. Moreover, there would appear to be only limited synergies between the two forms of official oversight, notwithstanding the combination of life insurance and superannuation regulation under the ISC which arises, in part, because life offices have historically been substantial offerers of superannuation investments.

Indeed, there are strong reasons for their not being carried out by the one agency. Prudential supervision, as defined, necessarily carries with it some official ‘comfort’ – though usually, as in Australia, short of a guarantee – for depositors about solvency and the maintenance of promised nominal values.[9] This may involve the use of public funds, but can also be achieved through the organisation of a takeover by a stronger institution or an officially managed workout. There should be no expectation of such support in the case of managed funds. Yet investors might assume that some obligation for adequate investment performance is implicit if the same agency has responsibility for both types of institution, the more so when a capital-backed institution has an ownership linkage with the funds manager. Any arrangements which encourage the perception that the generality of managed funds are no more risky than deposits would constrict the risk spectrum, increase moral hazard and the implicit ‘safety net’, and reduce the efficiency of the mechanism by which savings are allocated to investments of varying characteristics, including risk. If investors, being well-informed of the risks, are prepared to accept uncertainty about investment returns, the government should not intervene in the absence of external effects.

Beyond the broad two-way classification, institutions in the capital-backed class fall into two[10] distinct sub-categories:

  • intermediaries or deposit-takers, whose liabilities are relatively liquid and which hold out that money invested will be repayable in full either on demand or at the completion of an agreed period, usually with interest added. Whether all deposit-takers should be subject to the same prudential rules depends largely on community preference. Among deposit-takers, banks have generally been the more closely supervised for reasons which are discussed below; and
  • insurance companies, which offer products combining a simple contract promise (payout on death or other future event) and annuities. With such products, the exact amount of payouts will depend on investment performance. Life insurance offices also offer investment-linked products managed on a best-endeavours basis. However, their capital-backed and investment-linked products are segregated into different statutory funds.

Clearly, there are significant differences in the modes of operation of these capital-backed institutions and in the nature of the risks involved in their meeting their obligations. For instance, insurance companies have long-term liabilities with ill-defined value, while their assets are generally marketable with readily ascertainable values. Banks, in contrast, tend to have relatively short-term liabilities, with assets which are difficult to liquidate and to value. Consequently, the applicable prudential supervisory regimes are different and there would be few (if any) efficiencies in bringing their supervision together.[11] Furthermore, as discussed later, the failure of an intermediary is more likely to be a concern for financial system stability.

Another argument for combining supervisory agencies draws on the fact that certain institutional groups are now offering products which traditionally have been the preserve of others. An example would be home lending by life insurance companies. The argument is that these products should be subject to the same prudential capital rules in each case. The capital adequacy rules for banks' housing loans have, however, been determined as one part of the supervisory regime for the overall activity of banks – with the ultimate objective of bank solvency – and it might or might not make sense to align a particular component of those prudential requirements across different groups. Of course, even if that were desirable, harmonisation would not necessarily depend on any amalgamation of regulatory agencies.

The final argument for combining supervisory agencies is that the challenges of different institutions being joined under common ownership in financial conglomerates can be met most efficiently by having one regulator for the entire entity.

Many conglomerates will, of course, be answerable to more than one agency because of the different jurisdictions to which constituent companies (often including a capital-backed intermediary and an investment manager) belong. There might be some administrative economies for such conglomerates if they were answerable to only one agency, but these will be limited while ever different supervisory and regulatory requirements apply to their components.

More substantial arguments for having one agency overseeing conglomerates stem from the related tasks of limiting contagion risk from one entity in a conglomerate to another, and of appraising the overall health of a conglomerate with several capital-backed components. Where the major entities in a conglomerate are supervised,[12] these tasks call for effective communication and co-ordination among the relevant regulatory agencies, and put a premium on transparent corporate structures. As long as prudential supervision of different members – such as banks and insurance companies – remains specialised, it is questionable whether a single agency would be more effective than two or three working together. An international consensus is emerging that, for most financial conglomerates, a convenor or lead regulator should be nominated to organise group-wide financial assessments, exercise authority over special-purpose holding companies and co-ordinate crisis response.[13]

Since, in the Australian context, there are rarely more than two prudential supervisors for any conglomerate, it seems practicable to manage the growth of conglomerates through such mechanisms. The Council of Financial Supervisors presently provides an effective forum for discussing conglomerate issues and devising solutions to these problems. It has recently proposed to Government legislative changes dealing with facilitating exchange of information among agencies and with regulation of special-purpose holding companies in financial conglomerates.

A detailed consideration of the organisation of product regulation is beyond the scope of this paper, but financial system developments suggest the likely greater need for reform here than in prudential supervision. This is because, while the distinctions between two or three basic classes of institution remain relevant, particular financial products and services are being offered from a widening range of sources (including firms whose business is largely non-financial). The presumption must be that, in the interests of effective competition for customers, similar regulatory treatment should apply to similar products regardless of their source. There also appear to be a larger number of agencies involved in product regulation and some rationalisation could well yield efficiency gains.[14] As far as possible, these activities might fall to an agency with economy-wide responsibility for fair treatment of consumers and dispute resolution.

As noted earlier, there is no case for a substantially different competition policy to apply in the financial sector, so responsibility should lie with an agency covering the entire economy. There does, however, need to be a mechanism for resolving a situation where a merger of financial institutions would be in the interests of financial system stability, but might be judged as potentially anti-competitive.

5. Supervisory Role of the RBA

Central banks generally have responsibility for financial system stability,[15] which is complementary to their responsibility for price stability. They consequently take a keen interest in the condition of those parts of the financial system where problems could result in systemic instability.

There appears to be no generally accepted, objective definition of financial system instability, but it may be taken to refer to circumstances involving significant disruptions to credit flows with consequential impacts on the real economy, or involving violent swings in financial prices due to disruptions in asset or derivatives markets including liquidity shortage.

Historically, concern about the former has focussed on banks because of their combined roles of maturity-transforming intermediation and payments. The collapse of a bank would have two undesirable impacts on the economy. Credit flows to its own borrowers would be interrupted and, given the information-intensive nature of much bank lending, these customers might not readily find other sources of finance. While this is clearly a greater concern the larger is the bank, the failure of even a small bank might damage other larger banks – either through its direct credit and payment system linkages with them, or through confidence effects. This could, in turn, cause broader disruption to credit flows and to the household sector's access to holdings of liquid wealth.

A standard description of banks as a potential source of systemic damage is:

Governments throughout the world regulate banks because the combination of loans financed by demand deposits has, historically, been a volatile mix, leading to costly banking panics. If the banking system becomes insolvent, potentially large costs are borne because the payments system is disrupted, borrowers become illiquid, and information about borrowers is possibly lost. Banking system insolvency is caused by a banking panic, an event in which bank depositors en masse demand cash in exchange for their deposits. Banks cannot honour these demands because markets for bank loans are not sufficiently developed. Markets for loans do not exist because of the expense of producing information about the riskiness of borrowers and the incentive problems of inducing banks to monitor borrowers if the bank has nothing at stake (having sold the loan).[16]

In Australia concerns about such systemic risk – as well as a desire to protect the liquid wealth of household savers – motivate the RBA's monitoring and supervision of banks, for which its responsibilities are specified in the Banking Act.[17] There are two related objectives. The first is ‘the encouragement and promotion of the carrying out by banks of sound practices in relation to prudential matters’, where prudential matters means, inter alia, the conduct of a bank's affairs in such a way as to keep itself in a sound financial position, and not to cause or promote instability in the financial system. The second task, which applies only to locally incorporated banks, is to ‘protect the interests’ of depositors if a bank gets into serious difficulty. For this purpose the RBA is empowered to take control of and manage a bank.

The main advantages of the central bank's being supervisor of the banking system are:

  • this gives it a direct hand in protecting the stability of the financial system for which it has ultimate responsibility, since the banking sector remains not only the largest segment in the financial system but also the most likely source of such instability;[18]
  • as supervisor, it is in a better position to assess quickly the need for lender-of-last-resort assistance to a bank which is in difficulty or suffering a loss of depositor confidence;
  • by supervising banks, it gains a first-hand knowledge and ‘feel’ for financial market conditions and for the behaviour of those institutions which are a key element in the transmission of monetary policy changes to the general economy. This can be an important input to monetary policy decisions. There are more likely to be complementarities between supervision and monetary policy than conflicts, and any conflicts which do arise will need to be resolved however the various responsibilities are allocated;
  • a bank supervisor is likely to be more effective in foreseeing and mitigating threats to bank stability if it has a close understanding of macroeconomic conditions and developing trends; and
  • banking supervision is inter-related with a central bank's policy and operational roles in the payments system.

The main arguments in overseas debates against central bank supervision have been the likelihood of greater moral hazard when the supervisor has its own substantial financial resources which may be called upon to fund a bailout, and the view that there are potential conflicts with monetary policy.[19]

Notwithstanding the apparent preponderance of points in favour of central banks conducting bank supervision, in many countries they are not the primary regulators of the banking system. In all such cases, however, arrangements have been formed for close liaison with the supervisory agency. Commonly, those central banks also devote substantial resources to banking system analysis, to varying extents ‘shadowing’ the banking supervisor. Often the central bank is closely involved in making supervision policy (for instance, in Germany); in some cases, it is effectively a second supervisor (for instance, in Japan).

Ongoing structural changes in the financial system are raising questions about the scope of banking and, by extension, the supervisory role of the central bank. Remembering that the main reason for its direct involvement in supervision is the responsibility for financial system stability, is the RBA's current role in relation to the banking system still appropriate? Should its supervisory focus be broader because systemic problems are now as likely to arise in other places (or, perhaps, because the RBA needs explicit authority over bank associates to protect banks effectively from contagion)?

These issues are summarised neatly in the following quote:

Should central banks be directly involved in supervision, and if so, of which institutions? The first part of this question has always been a subject of debate, with powerful pros and cons. I do not deny the strength of the moral hazard counter-argument, which points to the risk that supervision may bring with it destabilising expectations of support from the central bank. But I am closer to those who advocate bank supervision by central banks, mainly on the basis of the argument that it is difficult to draw a practical distinction between systemic and micro-prudential responsibilities. The prevention of systemic risk can hardly be effective without intimate knowledge of the participants in the market and the linkages between them. At the same time, I do recognise that the blurring of the line of demarcation between banking and other financial intermediaries raises some tricky questions. Why should central bank supervision stop with an ill-defined group of intermediaries, namely banks? If it were extended to other institutions, where should it stop? The “globalisation” of supervisory duties in the hands of central banks would not only enlarge the areas covered by the moral hazard risk; it would also put an excessive operational burden on them. There is surely a point beyond which the drawbacks would begin to offset the advantages derived from effectively ensuring systemic stability.[20]

Those who place store on the greater moral hazard risks in central bank supervision will argue for reduced involvement, and may find further support for this in the prospective relative decline of traditional banking. The case will be argued all the more strongly if such proponents are also convinced by the case for combining all supervisors into one agency.

It follows, however, from Edey and Gray's analysis that the Australian banking sector remains the appropriate, primary focus for supervision concerned with systemic instability. This sector still contains the vast majority of financial activities with the classic bundle of characteristics which have caused banks to be a particular focus of prudential supervision: maturity-transforming intermediation, opaque asset portfolios and participation in the payments system. Introduction of real-time gross settlement for high-value interbank payments from 1997 will reduce the risk of instability flowing through the payments system.[21] But a substantial proportion of domestic payments will continue to be settled on a net deferred basis, as will the large volumes of international payments for which banks are responsible.

Systemic risks do not arise to the same extent with life insurance companies whose liabilities are less liquid (and are thus less susceptible to a ‘run’) and which do not have the same linkages with each other as banks do through market trading and the payments system.[22] Even more clearly, the expanding funds-management sector does not pose the same risks. Not being capital-backed, the value of claims on these institutions is linked to the prices of their assets. If there are sharp falls in the value of assets in a fund there would be wealth effects for its investors, but none of the other consequences which could follow the collapse of a bank. There are, for instance, no credit links with other funds (as there are between banks) and, by and large, these funds invest in commoditised loan assets whose issuers would be able fairly readily to find alternative purchasers if one or more funds managers left the market.

As in most other countries, however, there are financial intermediaries now supervised as ‘banks’ in Australia whose activities are not predominantly of the classic banking kind – firms engaging in investment banking and treasury business. These tend to be regarded as banks because at least a part of their activities is usually of the traditional kind (there has not been the clear demarcation between commercial and investment banks which applies in some other countries), but also because the volume of their trading activities and settlement exposures with other banks mean they have been considered to warrant similarly close supervision for the protection of systemic stability.[23] Without adequate capital, they would be unable to continue operating and their capacity to meet immediate obligations to other intermediaries would be in question.

There are also some financial institutions whose activities resemble very closely those of traditional (and non-traditional) banks which are not authorised under the Banking Act and supervised by the RBA – capital-backed intermediaries such as money market corporations, building societies, credit unions and finance companies. Many of these have exemptions from RBA supervision on the grounds that, while they carry on ‘banking business’, they do not carry on the ‘general business of banking’. As this distinction loses whatever meaning it may have had in the past, it needs to be asked whether these intermediaries should be supervised by the RBA according to the same prudential framework as applies to authorised banks.

On one view, that which emphasises moral hazard risk, the test of ‘likely damage to system stability’ should remain paramount in determining the scope of central bank supervision. A practical difficulty is, of course, knowing exactly how and where to draw the line between different intermediaries, and how to recognise early enough the need for adjustment in the line over time. A size test could be adopted on the grounds that a very small ‘bank’ poses only trivial risk to the system. If so, would this be measured by capital, assets, daily trading volumes or some combination? And would the drawing of any arbitrary line have undesirable competitive impacts? The administration of the Banking Act has resulted in clear legal distinctions being made between those intermediaries which are authorised as banks and those which are not, but these distinctions have looked particularly arbitrary since entry restrictions for foreign banks were liberalised in 1992. Some of these have taken the opportunity to apply for a banking authority, while others with virtually identical operations in Australia have chosen not to.

Another view on this question, one emphasising efficiency and competitive considerations, would have the central bank supervise all those intermediaries engaged in bank-like sets of activities. In this scenario, the RBA would assume supervisory responsibility for building societies, credit unions, money market corporations and, perhaps, finance companies. This could result in a compression in the perceived spectrum of risk among intermediaries, but that appears only to be in line with the community's preference – at least as regards intermediaries gathering household savings. There could also be an extension of moral hazard risk. To mitigate this, it would be all the more important to change the common perception that RBA supervision is an absolute guarantee against institutional failure. One useful step to this end would be to recast the Banking Act, removing the widely misunderstood references to ‘depositor protection’ and restating the RBA's dual responsibilities as prudential supervision (to reduce the likelihood of institutional failure) and crisis management (in the event that a failure occurred). The provision for depositors to have first claim on assets in Australia would be retained.

Regardless of where the RBA's direct supervisory authority starts and finishes, it needs to take a broad interest in financial system developments, both domestically and globally. While the greatest vulnerabilities are in banking, it is clearly possible for financial instability to arise in other quarters, including from overseas. This has always been so, but such risks have almost certainly become greater in the past decade with the vastly increased volumes of financial market trading. Consequently, a central bank has to be familiar with all major financial markets and to liaise regularly with other prudential supervisors. It needs to take a keen interest in the robustness of systems for clearing and settling securities transactions.[24] As noted recently by the Governor of the Bank of England:

Whatever view one takes of these particular issues [who should have prime carriage of bank supervision], two things seem very clear. First, that any central bank must monitor developments in the banking system very closely, and that will necessarily involve monitoring what is happening in individual banks. And, secondly, a central bank cannot, in the modern world, limit its view to developments in the banking system alone. Because systemic threats can originate in other parts of the financial system, and because of the speed with which they can be transmitted through the system, we must necessarily take a very close interest in the financial sector as a whole.[25]

6. Bank Supervision and Competition

The process by which certain banking services are being cherry-picked by specialist providers raises a difficult competitive challenge for the banks themselves, and poses some questions for policy makers. Technological change is facilitating this trend by giving new players access to bank customers without their having to duplicate the expensive branch network through which banks have traditionally reached depositors and borrowers. It is being aided by developments in information processing and other financial technology which promote the growth of securitised, as opposed to intermediated, borrowing. It is also being encouraged by banks' traditional pricing structures which have incorporated a substantial element of cross-subsidisation, leaving them vulnerable to specialist competitors for those services which have had wide profit margins, and by apparently greater willingness of consumers to shop around.[26]

So far, erosion of the banking sector's pre-eminent position has not run very far in Australia, but the pace of change has accelerated in the past couple of years. The Government's policy of encouraging household saving for superannuation purposes will add momentum to it. It is, therefore, hardly controversial to assume that the banking sector will become a smaller proportion of the financial system – at least as measured by assets under control – although the extent and speed will, of course, depend on the competitive responses of the banks. Particularly important in this respect will be how quickly they manage to roll back the cross-subsidisation in their pricing, cut the costs of increasingly outmoded branch networks and exploit the delivery systems made available by new technology. Banks will also, no doubt, reshape their activities by applying their competence and knowledge to the expanding types of finance, for instance in underwriting the issue of securities.

Australian banks are also relatively free to establish subsidiaries for ‘non-bank’ financial activities – funds management, superannuation, securitisation, and so on. Competition policy can, of course, impose some constraints here, and there are some supervision requirements designed to protect banks from reputation and other risks arising from the activities of their non-bank associates. Banks have, however, begun to respond to intensified competition from funds managers by establishing their own subsidiary operations, and some have been conspicuously successful. More recently, they have set up their own mortgage origination and securitisation vehicles.

Whatever progress is made in these areas, more intense competition will draw attention to the competitive costs of bank supervision. Traditionally, bank regulation has been seen as conferring monopoly advantages on banks, protecting their profitability if not their market share. Although the evidence is not clear-cut, it appears that the average profitability of banks has declined a little following deregulation in the first half of the 1980s. Deregulation allowed the banks more freedom in pricing and in lines of activity, but also exposed them to greater competition both with each other and with new players taking advantage of liberalised rules for banking authorisation and participation in the Australian financial system in other ways.

Experience with deregulation and more competitive markets, and the heavy banking losses early in the 1990s,[27] have resulted in an intensification of prudential supervision to protect banking system stability, both in Australia and internationally. Like the earlier regime of regulation this confers certain benefits on banks, as well as imposing various restraints.

The benefits include:

  • enhanced market standing, which reflects in average funding costs (both domestically and overseas);
  • higher general status in the community;
  • participation in the payments system though exchange settlement accounts with the RBA; and
  • opportunity to establish operations in foreign markets which would otherwise be denied.

The value of the first and second aspects probably varies over the cycle, with banks seen as a particularly safe haven when the economy is in recession and some financial institutions are in difficulty through loan-losses.

The restraints of prudential supervision include minimum requirements for capital and holdings of prime assets, limitations on large individual credit exposures and open foreign exchange positions, and so on. There are also conditions to be observed by banks regarding their involvement with non-bank activities, including securitisation and funds management. These requirements do not impinge on the funds-management operations in their own right, other than on representations about the nature of bank support for investment performance. There are also costs associated with data compilation and transmission, executive time devoted to consultation with supervisors and the cost of additional work by external auditors at RBA request.[28]

Neither the costs nor benefits of supervision can be estimated with any precision, although the gross costs do not appear to be particularly significant at present. They are almost certainly a considerably less important source of competitive disadvantage than banks' high operating costs, inappropriate pricing structures and shifts in consumer preference. Most banks are, for instance, holding capital clearly in excess of the RBA's requirements and large exposure limits are rarely approached, except by the smaller foreign-owned subsidiary banks. It is, however, likely that those structural changes in the financial system which are facilitating competition from new players are eroding the value of the net benefits – or exacerbating the impact of the net costs – of supervision.

Should prudential supervision be made less onerous to help banks maintain their competitive position? This would be very difficult to justify if it left the banking system more vulnerable to instability. Some commentators have, indeed, suggested that bank capital standards could have to be raised if their balance sheets become riskier when denuded (by securitisation) of relatively high quality housing and corporate loans. It might, on the other hand, be possible to justify less restrictive supervisory requirements if banks eventually become a much smaller component of the financial system and therefore a less significant potential source of instability. It is too early to judge what the relative importance of these factors might be.

I also note that, realistically, bank supervision could not be modified significantly as long as Australian banks aspire to extensive involvement in international markets. The trend is to international conformity in supervision standards, rather than the reverse.

Another response to the emergence of new competitors for banks would see prudential regulations extended to them. If these new suppliers of financial services were performing the same roles which have given banks their distinctive character, they would meet the ‘threat to system stability’ test described earlier and the case to bring them under bank-like supervision would be strong.[29] By and large, however, the new entrants are not offering the same package of financial services which distinguishes banks – they are unbundling that package and offering individual elements such as mortgages or capital-stable liquid investments.[30]

To illustrate the issues here, a topical example of new competition for banks is housing loans from mortgage originators, at least half of which are funded by mortgage-backed securities issued through special-purpose vehicles. There have been suggestions that mortgage originators or the associated securitisation vehicles should be subject to prudential capital requirements similar to those applying to housing loans on a bank's balance sheet. But these requirements are to protect banks from insolvency, an objective which is not relevant for the originators or their securitisation vehicles, neither of which has deposit-type liabilities. Funding for their loans comes through marketable securities issued in professional capital markets, while the originators themselves are agents with no significant balance sheets of their own. The failure of an originator (or several of them) would have no significant ramifications for investors or the broader financial system. There would, at most, be some inconvenience until a new servicing agent was appointed for existing loans. Investors in a securitisation scheme might, of course, find that returns are less than expected, or even negative, but such losses do not have the same potential impacts on continuing credit flows or on the payments system as may flow from losses in a bank.[31]

The appropriate course for bank supervisors appears to be to maintain the current broad framework of supervision,[32] while accepting that banking groups will increasingly diversify into ‘non-bank’ activities. This will increase the prominence of the supervisory challenges posed by conglomerates. There is no doubt that, in particular, the supervision policies aimed at protecting a bank from reputation damage due to the poor performance of its funds management associates will need to be vigorously enforced, including disclosure standards and restrictions on the extent to which a bank may use its own capital to support the performance of an associated funds manager. If these policies cannot be made to work effectively, and they could well create some friction with banks' commercial preferences, there would be a major expansion in the range of investments for which the community might look to official support if performance expectations are not realised.

Another policy issue is whether it matters that banks are likely to become a smaller part of the financial system. Would there be an adequate supply of relatively safe deposit-type savings products, or will savers be ‘forced’ to more risky investments? Will there be an adequate supply of loans to small and medium businesses?[33] As long as the reshaping of market shares is being driven fundamentally by market forces and not by officially imposed costs and restrictions, there seems little reason for concern on this score.

7. The Approach to Bank Supervision

The forces of change in the financial system will pose numerous challenges for banking supervisors.

As already noted, these will include adapting supervision in response to banks' increasingly operating in conglomerate structures with insurance companies and funds managers. This will require more communication with other regulators and greater attention to limiting contagion risks.

Another issue will be dealing with the trend for banks to outsource components of their operations. This will no doubt increase as banks seek to spread the costs of expensive technological infrastructure, build mutually beneficial alliances with telecommunications companies and other specialist suppliers, and concentrate their own resources on those activities where they have comparative advantage. Supervisors will need to be satisfied that this trend does not compromise the capacity of banks to manage their risks effectively.

Supervisors are also having to deal with the increasing complexity of banking. This complexity is associated in large part with the widespread use of derivative products which hedge, but can also create, risks for banks. Often the dimensions and the sources of these risks are difficult to assess, as are the channels through which problems might flow between institutions and across markets. At the same time, the linkages among markets (including internationally) and modern communication mean speedier transmission of disturbances. Banks are also relying on more sophisticated and complex techniques to determine asset pricing, capital allocations and loan provisioning. These developments are in part a response to intensified competition, but also reflect the availability of computer-based techniques and data processing capacity.

Effective supervision will require supervisors to be familiar with these new techniques, particularly as banks contend that supervisory standards should place more reliance on their internal risk-management systems and less on ‘broad brush’ rules of thumb (such as the Basle capital adequacy framework for credit risk) which have the advantage of simplicity, but which are less finely tuned to risk-management needs than are the more sophisticated banks' own systems. This need is illustrated by the debate over capital adequacy standards for market risk where banks argued, with substantial success, for supervisory capital charges to be based on their internal models rather than a standard model, as originally proposed. Certain minimum criteria have been specified to ensure a degree of consistency across banks and countries.

If similar approaches can be adopted more widely, they offer the joint prospects for supervision to become more effective while also being less prescriptive and costly for banks. This might help, in turn, to offset some of the competitive disadvantages flowing from bank supervision. For instance, moves in this direction might be possible when the capital standards for credit risk are revised, with the possibility of recognising bank-developed systems for credit scoring and dynamic provisioning and avoiding the need for parallel internal and supervisory systems of capital allocation.[34]

Another benefit in this approach of placing more weight on a bank's own management systems is to emphasise clearly the responsibility of managers and directors for its prudent operation. Some would argue, however, that arrangements which entail supervisors effectively ‘signing off’ on banks' internal systems also increase moral hazard risks by making it harder for a supervisor to abstain from providing support for a bank which subsequently gets into difficulty. Supervisors need indeed to be alert to such risk, but even if there is a moral hazard price it might still be worth paying for a more market-oriented form of supervision entailing a lower regulatory burden for banks.

The other element of the trend toward more ‘market-friendly supervision’[35] involves encouraging more informative public disclosure about banks' activities, thereby enhancing market disciplines as a supplement to official oversight. Such efforts should certainly be pursued in co-operation with the accounting profession and some significant progress has been made, particularly in relation to derivatives trading and asset quality. But the same complexity in banking which renders more challenging the supervisor's task, together with the fact that exposures in large trading positions can change very quickly, would seem to militate against heavy reliance on disclosure as an effective substitute for on-going prudential supervision – including assessments of the integrity of risk-management systems – by an official agency.


Llewellyn, in this volume, describes the trends affecting banking in a more general context; see also Borio and Filosa (1995). [1]

These broad objectives are described at more length in many places, including by Dale (in this volume). [2]

Moral hazard refers to the possibility that, in the presence of official supervision, banks will adopt riskier business strategies in the expectation of a bailout if problems occur, or that depositors and other creditors will be less discriminating in their choice of institution. [3]

Council of Financial Supervisors, Annual Reports. [4]

An alternative classification system is proposed by Goodhart (1995), in which a ‘system stability’ objective is distinguished from ‘investor protection’. A practical problem with this in thinking about organisation is that prudential supervision of financial firms not threatening system stability is placed in the latter category, when the techniques and skills involved for this are virtually identical to those required for system stability supervision and quite different from those required for other sorts of investor protection. See also Taylor (1995). [5]

Note that there can be some trade-off between prudential supervision and product regulation, in that regulation to do with information for savers might be tougher for institutions which are only lightly supervised. For instance, finance companies – which are not prudentially supervised – have to issue prospectuses with more detail than is normally made available to intending bank depositors. [6]

See Dale in this volume. [7]

This generalisation leaves aside some fringe institutions, such as friendly societies, which are supervised by State agencies. AFIC, in fact, sets prudential supervision standards for building societies and credit unions which are then implemented by agencies in each State. [8]

No prudential supervisory system can guarantee against institutional failure. It can only reduce the likelihood of failure and facilitate the resolution of one institution's failure (whether through merger or liquidation) in a way which limits the damage to other parts of the financial system. Goodhart (1995) argues that an optimal regulatory system should involve occasional institutional failures, because avoiding these would be too expensive in terms of limiting risk-taking and innovation. [9]

A third, rather special, category comprises defined-benefit superannuation funds, where the fund has an obligation to make a future payment which is determined by criteria such as the member's final salary, age and years of employment. Such funds, which are declining in importance, are provided mainly as part of employment contracts rather than on a ‘public offer’ basis and ultimate payment depends partly on the ongoing viability of the employer. [10]

Glading (1995) argues, ‘There is no likelihood that the prudential rules for banks and life insurance companies will coalesce – at least in the foreseeable future … we should understand that a mega regulator would not remove the imperative that banks and life companies need to be regulated in quite different ways, because of the differing nature of their balance sheets’ (p. 28). [11]

Conglomerates pose greater difficulties when major components – such as non-financial firms – are not subject to supervision. Such conglomerates are, however, rare in Australia. [12]

Tripartite Group (1995). [13]

The RBA, for instance, has become involved with monitoring codes of practice and security standards in electronic funds transfer, activities which do not contribute at all to its prudential supervision role, or other roles, and should probably be located elsewhere. [14]

For the RBA this responsibility derives from its broad charter which extends to the ‘economic prosperity and welfare of the people of Australia’. [15]

Gorton (1994), p. 108. [16]

Note that not all intermediaries have been obliged to become ‘banks’, with the Banking Act providing for an exemption from the requirement to be authorised as a bank and supervised by the RBA where ‘a person desires to carry on any banking business in Australia but does not desire to carry on the general business of banking’. As discussed later in this paper, there is a real question about the ongoing tenability of this feature of Australia's system. [17]

Significantly, it is uncommon (though not exceedingly rare) for central banks to supervise insurance companies, securities firms, funds managers and so on. I am not aware of any countries where the central bank supervises such institutions without also having responsibility for bank supervision. The value of central bank involvement in supervision has been recognised in reallocation of responsibilities in recent years to the monetary authorities in Finland and Hong Kong. [18]

Others argue that any supervisor will periodically be judged incompetent because banks will inevitably fail or suffer large losses, and that if the central bank is the supervisor its credibility in monetary policy will suffer. This may have some substance, but very few central bankers with responsibility for bank supervision consider it sufficient justification to seek to discard that role. [19]

Lamfalussy (1992), p. 13. [20]

Goodhart (1987) argues that, even without any role in payments, the other classical features of banks would make them critical to financial system stability. [21]

Although life offices offer some liabilities with functional similarities to deposits, these account for only 5 per cent of aggregate deposit-type savings. [22]

As well as having 47 per cent of total financial system assets and over three-quarters of intermediaries' assets, Australian banks are responsible for some 80 per cent of trading volumes in derivatives. [23]

Regulators of securities – in Australia, the ASC – also have an interest in such systems through their concern for high standards of competence and integrity in the conduct of these markets. See Large (1996). [24]

George (1996), p. 10. [25]

See Edey and Gray (1996) in this volume for a detailed exposition of these developments. [26]

See Thompson (1993). [27]

Banks have to hold 1 per cent of their liabilities (less shareholders' funds) in non-callable deposits earning a below-market rate of interest. This is not a prudential supervision requirement, but a revenue measure. When the interest rate was reduced by a 5 per cent discount in 1995, it was also described as being in the nature of a payment for the benefits which accrue to banks from being authorised by the Government. [28]

Non-bank financial intermediaries also compete in bank products, but their competitive advantages over banks are not so great because, as capital-backed intermediaries, they are mostly subject to bank-type supervision, either directly or through parent firms. [29]

Similarly, it is hard to see significant threats to systemic stability in the activities of new players in the retail payments market. [30]

The need to align capital requirements for similar risks more closely across capital-backed institutions such as banks and life insurance companies should be investigated. [31]

Likely directions in implementing this framework are discussed in the following section. [32]

Gorton (1994), p. 109. [33]

Greenspan (1996). [34]

Padoa-Schioppa (1995), Crockett (1995). [35]


Borio, C.E.V. and R. Filosa (1995), ‘The Changing Borders of Banking: Trends and Implications’, in The New Financial Landscape: Forces Shaping the Revolution in Banking, Risk Management and Capital Markets, OECD, Paris.

Council of Financial Supervisors, Annual Report, various issues.

Crockett, A.D. (1995), ‘Market Regulation and Level Playing Fields’, Remarks to Session 1: Trends in Banking IMC Conference, Seattle, June.

George, E.A.J. (1996), ‘The Bank of England: How the Pieces Fit Together’, The Bank of England at the London School of Economics, 36(1), London, February.

Glading, R.G. (1995), ‘Life Insurance Distribution and Marketing’, Insurance and Superannuation Commission Bulletin, September.

Goodhart, C.A.E. (1987), ‘Why Do Banks Need a Central Bank?’, Oxford Economic Papers, 39(1), pp. 75–89.

Goodhart, C.A.E. (1995), ‘Some Regulatory Concerns’, LSE Financial Markets Group Special Paper No. 79, London, December.

Gorton, G. (1994), ‘Bank Regulation When “Banks” and “Banking” are Not the Same’, Oxford Review of Economic Policy, 10(4), Winter, pp. 106–119.

Greenspan, A. (1996), Remarks at the 32nd Annual Conference on Bank Structure and Competition, Federal Reserve Bank of Chicago, May.

Lamfalussy, A. (1992), ‘The Restructuring of the Financial Industry: A Central Banking Perspective’, SUERF Papers on Monetary Policy and Financial Systems No. 12, March.

Large, A. (1996), ‘Systemic Risk and the Securities Regulator’, Speech to the Japanese Securities Dealers Association, Tokyo, March.

Padoa-Schioppa, T. (1995), ‘Co-operation Between Banking and Market Regulators’, Banca d'Italia Economic Bulletin No. 21, October.

Taylor, M. (1995), ‘“Twin Peaks”: A Regulatory Structure for the New Century’, Centre for the Study of Financial Innovation, London, December.

Thompson, G.J. (1993), ‘Financial Stability and Bank Supervision – The Australian Experience’, CMBF Paper, Conference on Financial Stability No. 7, Centre for Studies in Money, Banking and Finance, Macquarie University, Sydney, October/November.

Tripartite Group (1995), ‘The Supervision of Financial Conglomerates: A Report by the Tripartite Group of Bank, Securities and Insurance Regulators’, Basle, July.