In its 1999 Annual Report, APRA identified its main goals and objectives in the following terms:
- to be a world-class prudential supervisor;
- to establish effective and efficient prudential standards for the institutions it supervises;
- to operate in a flexible, open and accountable manner; and
- to be forward looking in approach, applying best-practice, risk-based techniques to the supervision of financial entities.
Achievement of these objectives rests on two pillars. The first is the delivery of high-quality supervision to the regulated financial sector. This is dependent on having proficient teams of supervisors focusing on the health of individual financial institutions and financial groups. The second is a commitment to having sound prudential policies supported by expert research and analysis.
The need for a strong focus on policy analysis, research and development is especially critical in an environment where change in the financial system is rapid. It is clear that there are many challenges ahead for prudential supervisors. For example, structural change is posing difficult questions, including how to supervise financial conglomerates that include different mixes of prudentially regulated and unregulated entities. Prudential regulators worldwide are trying to come to grips with this issue. Adding further complexity to this structural issue is the emergence of new types of financial services firms, often using new technologies (such as the Internet) as their main delivery channel to customers. Once again, prudential regulators are working to identify the issues that these developments raise. The expansion in risk measurement, management and mitigation techniques, and their implications for existing supervisory rules and procedures, pose yet another set of issues requiring close analysis and effective supervisory response.
The challenge for APRA is to deal with these complex developments, as they affect the Australian financial sector, within a single prudential framework.
As important as the policy issues being considered is the question of how supervision policy is designed. As noted above, APRA aims to be at the leading edge of international prudential developments. To achieve this requires, at its most basic level, a dedicated and professional team devoted to the study and advancement of prudential supervision practice and theory. More than that, however, it requires an analytical, non-bureaucratic approach that is aligned to APRA’s day-to-day supervisory tasks and objectives. Policy should be a means to an end - the end being better supervision of the financial sector - and not an end in itself. To be truly effective, policy initiatives must be based on a foundation of rigorous analysis and research that takes into account both domestic and international implications.
To be credible, the policy-making process must also be transparent and well understood by financial institutions and the broader community. Amongst other things, this requires special attention to coordination and communication with institutions and with the other major stakeholders that have an interest in prudential developments (industry groupings, other regulatory agencies, government and academia). In addition to face-to-face discussions with staff of financial institutions, there are many other channels by which coordination and communication can be achieved: through the regular release of draft prudential standards, policy statements, discussion and research papers and volumes of readings designed to outline research findings or issues of relevance to the financial sector, and through participation in, and organisation of, seminars and conferences focusing on financial and prudential matters. APRA uses these mediums extensively.1
In the course of its first year-and-a-half of operation, APRA has identified a range of policy matters to be addressed over the short-to-medium term. These are discussed in detail below and range from large scale projects aimed at changing the existing focus and direction of supervision, to technical areas of research associated with the measurement of risk in financial institutions.
The policy issues currently being considered can be categorised under four broad headings:
- those associated with the integration and harmonisation of supervisory arrangements, and proposed improvements to those arrangements;
- those aimed at enhancing institutional solvency or capital adequacy;
- reactions to new or emerging developments in the financial sector; and
- other activities relevant to the functions of a prudential supervisor.
As is so often the case, the boundary lines between these classifications are rough at best, and many of the specific policy projects identified below would fit well into more than one of the groupings.
One of APRA’s immediate objectives is to achieve greater integration and harmonisation of supervisory arrangements than has been the case in the past. While the system of prudential supervision in Australia was not totally uncoordinated prior to the formation of APRA in 1998,2 it is the case that the approach to prudential regulation of different financial institutions and entities had evolved along quite different paths over the years. To some degree, these differences reflected the varied activities of the respective sectors. However, the differences can also be traced to divergent supervisory philosophies that developed around the regulation of institutional sectors. As a simple example, the extent of reliance on legally enforceable and prescriptive prudential rules has varied markedly between the sectors. The supervision of the smaller deposit-taking institutions has relied to a greater degree on legal standards and prescriptive rules than has been the case where larger deposit takers have been concerned. The supervision of superannuation, based in recent years on the Superannuation Industry (Supervision) Act 1993 (SIS), is highly legalistic in approach. In contrast, supervision of the life industry, while seemingly legalistic and prescriptive on the surface, has in fact operated in a relatively flexible manner - certainly in comparison with general insurance.
Whatever the merits of these different approaches, the emergence of APRA as a single focus for prudential supervision and policy has provided a unique opportunity to consider existing institutional based supervisory arrangements and the philosophies that underlie them, and the extent to which greater harmonisation of arrangements across the different institutional groups is justified and practicable. It facilitates change that might have been difficult to achieve under previous arrangements. At the same time, it provides an opportunity to confirm existing approaches where they can be shown to represent the optimal approach for their purpose.
Under this heading of harmonisation and integration, five projects are in train that have the potential to lead to a more unified approach to supervision than in the past.
This major policy initiative aims to bring about greater consistency of approach to the supervision of Authorised Deposit-taking Institutions (ADIs). Under pre-APRA prudential arrangements, ADIs – comprising banks, building societies and credit unions – were covered under two related, but quite distinct, regulatory regimes. Banks were subject to prudential guidelines issued by the Reserve Bank of Australia. Guidelines covered topics such as capital adequacy, liquidity arrangements, large credit exposures, funds management and securitisation, external audit arrangements, and banks’ associations with non-bank entities. The smaller ADIs, in contrast, complied with a host of prudential standards and prudential guidance and interpretative notes, often covering similar areas as the banks, issued by the Australian Financial Institutions Commission (AFIC). Ensuring adherence to these standards was the task of various State-based supervisory agencies.
With the creation of APRA, the existing prudential standards, guidelines and notes covering the ADI sector were taken over and, in the first instance, applied without significant change in approach. However, this arrangement was intended to be transitory until some reassessment of existing arrangements could be conducted. The long-term goal was the development of a set of standards which treated all ADIs equally from a prudential perspective, which drew no artificial boundaries between ADIs based on size, and which provided the flexibility to accommodate the wide range of activities to be found in the sector. Very importantly, the goal was an approach to supervision that emphasised the role of internal risk identification, measurement and management, and that provided incentives for all institutions to improve their internal risk control systems. This large project is under way and is being tackled in two stages.
Stage 1, scheduled for completion by mid 2000, will see the existing disparate arrangements covering ADIs rolled into a single set of standards to be issued under the Banking Act. The standards will cover liquidity management in ADIs, credit risk, market risk, capital and capital adequacy, large exposures, funds management and securitisation, operational risk, and relationships with related parties. Guidelines are also to be issued setting out a uniform approach to the authorisation of ADIs. While the standards being developed are intended to be legally enforceable,3 effort is being made to ensure that they are transparent and easily understood. The objective is a simple and unambiguous standard for each relevant area, directed mainly at the senior management and boards of financial institutions. The often complex information required to interpret the precise operation of the standards will be relegated to attachments and guidance notes to the main document.
Where possible, in this first phase of the project, small changes are being introduced to iron out obvious inconsistencies between the inherited guidelines and the new standards covering both large and small ADIs. However, as noted above, this process will not see significant changes by and large to the actual content of the arrangements applying to the different institutions.
Stage 2 of the process, beginning in the second half of 2000, will be more complex, involving a more thorough reassessment of the content of the standards – the aim being to revise the new standards to reflect the optimal longer-term prudential arrangements we wish to see applied to the ADI sector.
All stages of this project have involved (and will continue to involve) extensive consultation and discussion with the ADI sector. Draft standards, and final standards as they emerge, can be found on APRA’s website.
The second policy matter to fall under the broad heading of harmonisation and integration relates to the supervision of conglomerates.
Financial conglomerates have emerged as an important organisational form in the Australian financial system, as they have in many other countries. In Australia, conglomerates have evolved into financial services firms carrying out a diverse range of financial activity – from traditional retail banking, to wholesale (“institutional”) banking and traded market activities, insurance (general and life), stockbroking, funds management and superannuation. Many Australian conglomerates have extensive overseas banking and/or insurance and other operations.
The traditional approach to the supervision of conglomerates has been to focus on the separate legal entities making up the group (the bank, the insurance company, etc) and has involved application of the specific prudential standards relevant to those separate institutions. Historically, prudential regulators have not focused to a great extent on the activities of each group as a whole. Yet it is well known that the risk profile of a financial group, and the risks residing in individual institutions in the group, can be influenced greatly by its composition. It is equally well known that problems in one part of a group have the potential to spread, affecting otherwise healthy parts of the group. This is the process of “contagion”, and it represents one of the most cogent arguments for a more group-wide approach to the supervision of conglomerates than has been the case in the past.
Any focus on the group, rather than the individual institutional activities which comprise it, immediately raises the issue of activities that can be undertaken within a conglomerate. Here, regulatory rules have led to some anomalies. For example, under former Reserve Bank prudential guidelines, groups containing an authorised bank were precluded from engaging in any significant non-financial activities. This restriction was not applied to financial groups without a banking presence. For example, for the 11 years of its existence, the Insurance and Superannuation Commission permitted significant non-financial activities to co-exist in a group containing supervised insurance companies. The significance is that in the new financial world, many “non-financial” institutions see a potential role in the financial services sector. At the very least, this puts an onus on prudential supervisors to be clear, and where possible consistent, in specifying the activities that are seen as acceptable for co-existing in a group comprising regulated financial institutions.
The conglomerates project, which began in late 1998, aims to reconsider the regulatory framework covering those conglomerates which include regulated financial institutions. Given the magnitude of the tasks involved (the supervision of conglomerates touches upon virtually every aspect of prudential regulation), the project has been divided into a number of phases.
Stage 1 involves reassessing the framework applied to those conglomerates which include an authorised ADI in their structure. Issues being addressed include:
- the use of non-operating holding companies to head such groups;
- the extent to which it will be permissible for conglomerate groups to conduct a broader range of non-financial activities than permitted at present, or for predominantly non-financial groups to conduct financial activities. Subject to satisfying any prudential concerns that arise from this proposal, this initiative may encourage new entry into the ADI sector and thus provide a boost to competition and efficiency within the financial system; and
- the scope for group-wide risk management practices to be applied more widely in conglomerate groups containing banks, and the implications of such a development for other prudential policies (such as intra-group exposure limits and capital adequacy requirements imposed on ADIs and groups containing ADIs).
Stage 2 of the conglomerates project will involve closer examination of such issues as the details for determining capital adequacy for conglomerate groups, the treatment of capital placed in subsidiaries and related entities, and more precise assessments on appropriate limits for intra-group exposures and group exposures to third parties. These issues are increasingly becoming the focus of international debate, in part as a result of the work of the Joint Forum,4 and in part as a consequence of emerging developments in countries such as Australia.
Finally, Stage 3 of the conglomerates project will consider the application of the framework being developed for groups containing an ADI to conglomerates incorporating any financial institution prudentially supervised by APRA.
To date, two Policy Discussion Papers have been issued on this subject (in March and November 1999) and there has been extensive discussion with industry on the proposals, as well as with overseas prudential regulators. Finalisation of Stage 1 of the project is imminent.
On 1 July 1999, friendly societies with benefit funds (in effect, those societies conducting some financial business) were transferred to the Commonwealth regime of prudential regulation through amendments bringing them under the Life Insurance Act 1995. In effect, supervisory responsibility was transferred from AFIC to APRA in respect of safety and soundness of the societies, and to ASIC in respect of corporations matters (in much the same way as occurred in the case of the building societies and credit unions). Friendly societies are similar to life companies in offering products such as insurance bonds and in providing these through benefit funds (which are akin to the statutory funds of the life industry). APRA's role is to maximise the likelihood that each benefit fund (statutory fund) will have sufficient assets to pay members' (policyholders') future claims. This involves the usual prudential techniques of licensing criteria, capital requirements, risk management standards, on-going surveillance and (where necessary) enforcement action.
Under the Life Act, companies are required to meet solvency and capital standards based on modern actuarial methods. These have been codified in actuarial standards issued by the Life Insurance Actuarial Standards Board (LIASB). At the time of transfer to the Commonwealth, and to ease the adjustment burden on the societies, the LIASB released a set of transitional actuarial standards for friendly societies that largely reinstated the requirements of the pre-existing AFIC regime. These were to be replaced, however, by harmonised actuarial standards for all life companies and friendly societies by the end of a reasonable transition period, generally taken to be 1 July 2001. The LIASB has subsequently commenced work on the harmonisation of the transitional standards with those applied to life insurance companies. The timetable for the work has been set by the APRA Board and APRA oversees the general direction of the work being carried out by the LIASB. The objective is to create a level playing field for life insurance products - irrespective of whether they are held in a life company statutory fund or friendly society benefit fund - while allowing existing friendly societies to maintain their traditional identity for marketing purposes.
As noted in the introduction,APRA currently operates within a regulatory framework that is the product of different legislation covering different parts of the regulated sector (the Banking Act
, the Life Insurance Act 1995
, the Insurance Act 1973
, the Superanuation Industry (Supervision) Act 1993
, and so on). In theory, it should be possible to envisage a situation in which all of APRA’s broad powers available to regulate these different types of institutions are enshrined in a single Act. This would greatly facilitate application of a common philosophy, where appropriate and practicable, to all regulated institutions. The impediments to such an approach would be formidable, however, given history and the range and diversity of institutions to be covered.
As part of APRA’s efforts in relation to integration and harmonisation of prudential arrangements, it intends to explore this issue in detail and, in so doing, determine what may be feasible from what is not. Outlining the issues involved in such a proposal, and drawing out the benefits and the practical constraints, would represent a major step forward. At this point, the project is in its very early stages. It should also be noted that whatever the product of the analysis, APRA can only propose new legislation. Responsibility for legislation rests with the Government and the Parliament.
Our expectation is that a policy paper on this matter would be available by the end of 2000. In considering the issue, APRA will draw on the experience of the United Kingdom, which has had to deal with the same question in establishing the new Financial Services Authority.
In supervision of the superannuation sector, our objectives are to ensure, to the extent practicable, that fund trustees manage superannuation money prudently and in the long-term interests of members, and that superannuation money is used for genuine retirement purposes. A number of factors make the supervision of this sector rather different to others:
- for the most part, superannuation funds are not backed by capital in the same way that capital is present to support the interests of bank depositors or policyholders in insurance companies.5 Solvency is not a relevant concept for most superannuation funds. They are investment vehicles and there is no institution, as such, to fail. Contributors gain when assets held against contributions rise in value, but they can also lose money if asset values decline significantly. For good reason, neither supervisors nor governments have wished to overly constrain fund trustees by imposing rules to prescribe that only very conservative (and safe) assets be held. Such rules would risk reducing the long-term gains flowing to superannuants from well-structured, diversified portfolios.
- the size, diversity and fragmentation of the sector makes it inherently difficult to supervise. In total, there are over 200,000 separate superannuation funds operating in Australia. Even allowing for the transfer of most of the smallest funds to the responsibility of the Australian Taxation Office, APRA will retain supervision of several thousand funds, some very large and some small.
There has been significant progress over recent years in improving supervision of superannuation. Our current arrangements are intended to provide more effective supervision without increasing the compliance burden for funds, and ideally reducing such costs. The primary focus of our review methodology is the key risk areas of particular funds, rather than the pursuit of a standard range of questions indiscriminately across all funds (a “one size fits all” model). Large funds are required to provide more frequent information and supervisory contact is on a more continuous basis – as occurs in other supervised sectors. The small funds are subject to supervisory review on a cycle.
We propose to refine the current risk-based approach to supervision of the sector, and to consider application of any new techniques that might usefully be applied to the large funds and to the smaller funds. This will be an on-going process.
Taken together, these five projects - the harmonisation of ADI standards, conglomerates, the integration of friendly societies under the Life Act, the exploration of “generic” APRA legislation, and the supervision of superannuation - will occupy a significant proportion of APRA’s policy resources in 2000and beyond.
The maintenance of financial solvency is at the heart of any system of prudential supervision. Only solvent firms - those with assets sufficient to meet their liabilities - can continue to trade and thus fulfil obligations to liability holders, such as depositors of ADIs or policyholders in the case of insurance companies. Of course, the tests for, and meaning of, solvency vary across different sectors. However, common to both the ADI and insurance (general and life) sectors is an emphasis on capital and/or reserves as a means of protecting against the risk of failure or loss on the part of those placing funds with the institutions in those sectors.
APRA currently has a number of projects under way that aim to strengthen and modernise existing capital and solvency regimes for financial institutions.
In 1988, banking supervisors in the major economies and financial centres of the world introduced the Capital Accord – an agreed set of definitions for bank capital and a methodology for identifying and measuring exposures to credit risk, by far the largest risk faced by these institutions. The Accord specified, for the first time, a minimum level of capital that should be held by a prudent bank to protect against possible losses arising from counterparty default. Very importantly, the Accord provided a mechanism, hitherto unavailable, to permit comparisons of capital adequacy in banks across countries and across time. This Accord was adopted quickly by many countries and, by the early 1990s, had become the foundation around which the system of prudential supervision in most countries was organised.
From the outset, the 1988 Capital Accord was designed as a relatively simple set of “rules of thumb” by which some measure of credit risk could be determined by reference to a bank’s on- and off-balance sheet activities, and then translated into a minimum capital requirement. The methodology did not purport to measure risk precisely, or even closely. It was a means of determining a sensible, safe level of capital for an individual bank and, as noted above, a means of comparing capital levels geographically and over time. From a broader perspective, of course, it was also a device used by the international supervisory “community” to focus attention on capital levels in the world banking system, and it was used deliberately as a means of engineering a general increase in capital levels for the system. This broad “macro” objective emerged against the background of a long-term decline in bank capital levels around the world in the 1950s, 60s and 70s.
By the late 1990s, after a decade of experience with the Accord, it was recognised that the 1988 arrangements, while still highly relevant throughout the banking systems of the world, were becoming dated in a number of respects. Improvements in techniques for the measurement of credit risk, developed by some of the leading international banks, were an important influence. Tensions were emerging as banks estimated their true capital needs based on actual experience of credit losses, while regulators continued to impose capital requirements using the simple methods outlined in the Accord. There were also concerns that the Accord’s relatively unsophisticated approach to the measurement of credit risk may be distorting banks’ lending and capital management practices. This led to calls for a reassessment of the existing arrangements, and an upgrading of the Accord to reflect the advances in risk measurement and management practices that had occurred over the 1990s.
Considerable energy has been directed in recent years to the question of how the Accord could be made more relevant to the banking system of today.6 APRA has been active in this debate and in its research efforts. Specific projects relevant to this area include:
- a major submission to the Basel Committee on Banking Supervision (BCBS) on the new Capital Accord. APRA has made a detailed submission in response to initial proposals by the BCBS, issued in June 1999, outlining possible directions for the new Capital Accord.7 In its submission, APRA supports measures that make regulatory capital requirements more risk-sensitive, including placing greater emphasis on banks’ own risk rating systems as a means of determining capital charges, where those systems can be shown to track credit risk accurately. In addition,APRA has also supported the “unbundling” of the existing credit-based capital requirements into separate explicit, capital charges for different types of risk. In putting together its submission, APRA took account of extensive discussions with, and submissions from, Australian ADIs and other interested parties within the financial sector.
- a series of visits to the major banks, the object being to assess and research emerging capital allocation and credit risk measurement methodologies. These visits commenced in 1999 and will continue in 2000, expanding the range of institutions visited and focusing on particular issues relevant to the review of the Capital Accord. This work will be used to help shape supervisory policy in the years ahead.
- a stocktake of loan grading systems currently being used within the Australian banking system and a comparison with similar work being conducted internationally. The BCBS has recently released a paper which will assist with the benchmarking process.8 This exercise will assume increasing importance as the proposals to use internal grading systems for regulatory capital purposes take shape. Once again, APRA will be undertaking active discussions with industry (both at a bilateral level and through discussion papers) on the issues flowing from its analysis of this area.
- the creation of a consultative group on credit risk modelling, allowing APRA to discuss modelling issues with those institutions with recognised skills in this area. Not only does APRA see this as a forum for discussion on the complex issues associated with risk modelling, but as a means of enhancing and building on the work taking place in the banking system.
Other agencies, in addition to the BCBS, have been active in their attempts to improve solvency standards for financial institutions. Organisations such as the International Association of Insurance Supervisors(IAIS) have been working with groups such as the BCBS and the Joint Forum to develop greater understanding of risk in financial institutions and risk-based supervision. APRA is actively participating in these efforts through representation on the various committees of the IAIS, the Joint Forum and the BCBS.
Closely aligned to APRA’s work on the new Capital Accord are initiatives being undertaken in relation to operational risk. One of the implications of a more accurate approach to the measurement of credit risk in ADIs described above, is a new focus on other types of risk in institutions and the potential for such risks to be quantified. It has often been stated (correctly) that some of the most dramatic institutional failures have not been the product of credit risk or the consequence of sharp fluctuations in foreign exchange, security, equity or commodity prices, but the result of operational failure. The failure of Barings Bank in early 1995 (with losses of over £900 million) reflected the impact of fraud by a senior trader. Fraud of this magnitude was possible only because of poorly designed risk management systems within the bank. Much the same conclusion can be drawn from an analysis of losses experienced by many other international banks and securities houses during the 1990s.
The BCBS is investigating the potential to incorporate operational risk more formally within the capital adequacy framework. Its analysis is at a very early stage, and much work is still to be done before the objective becomes reality. For its part, APRA has approached a range of financial institutions (including ADIs and insurance companies) to discuss in detail their approaches to the measurement and management of operational risk. It has discussed similar issues with some of the leading financial consulting firms and industry groups. That very preliminary analysis suggests that a number of approaches are emerging to the identification and measurement of operational risk, ranging from simple “top-down” models of risk through to sophisticated “bottom-up” statistical analysis undertaken at the individual business unit level. A few institutions (primarily the large banks) are looking at the use of models both to manage operational risk and as the basis for capital allocation (ie setting aside capital to cover possible losses arising from operational failures).
The initial product of APRA’s work in this field will be a discussion paper. The objective of the paper will be to foster debate on operational risk within the Australian financial sector and set the scene for further analysis. Specific information gained on the state of play in relation to the measurement and management of operational risk will prove useful also in developing further APRA’s ideas on the incorporation of operational risk into the ADI Capital Accord.
The general insurance industry is an important part of the Australian financial sector. Although the assets held by general insurers account for only six per cent of total financial sector assets, every household with a car, a house or a home contents or similar policy is vitally affected by potential problems or failures within this industry. Many of the loans issued by financial institutions are facilitated by the mortgage insurance provided by general insurers. Similarly, the business sector relies heavily on the protection offered by the sector, ranging from general insurance coverage through to very specialised insurance without which many sectors would find it difficult to function (shipping or construction).
Given the significance of the general insurance sector to the activities of households and businesses, it is perhaps surprising that prior to 1973 there were virtually no supervisory requirements for companies wishing to conduct general insurance business in the Australian market place. The catalyst for change was the collapse of 16 Australian general insurers in the three-year period to mid 1973. This caused major financial loss to policyholders and a widespread loss of confidence in the industry. The Insurance Act 1973 was passed largely in response to these failures. It sought to reduce the possibility (or minimise the impact) of losses to policyholders as a result of company failures.
The Insurance Act 1973 has served the Australian community relatively well over the 27 years since its introduction. During this period the insurance industry has consolidated and stabilised. However, despite the strengthened prudential regime, there have also been 21 general insurance failures over the period. The causes of failure have included fraud, unsustainable rapid premium growth, underwriting and operating loss, cash flow and capitalisation problems, poor management and underwriting practices, stronger competition in particular market sectors and a poor spread of assets to support liabilities (and, more often than not, a combination of these factors).
Many of the problems that led to the failure of these institutions could be mitigated by a supervisory framework that was more responsive to the circumstances of individual general insurance companies and their risk profiles. In addition, increased transparency of an insurer’s financial performance and position would subject general insurers to greater market disciplines and improve the market’s understanding of the business.
Against that background, APRA initiated a major review of the supervisory arrangements covering general insurance in 1999. The project, which is now well under way, aims to:
- bring the supervisory framework applied to general insurance into closer alignment with APRA’s preferred supervisory approach, namely to make regulation as risk-responsive as possible;
- make it clearer that the objective that underlies the supervision of general insurance is the protection of general insurance policyholders, not general insurance companies per se;
- improve policyholder protection by improving the responsiveness of the general insurance supervisory regime to differences in individual companies; and
- improve efficiency within the general insurance industry through increased transparency.
A discussion paper setting out the agenda for reform was issued in September 1999 and has been the subject of extensive comment by the general insurance industry. Submissions are currently being analysed. In the meantime, a new discussion paper is scheduled for release before mid year, new standards on liability valuation and solvency are being drafted, and an industry conference has been organised to discuss progress. This conference will be held in Sydney in May 2000. The aim of the proceedings will be to review the draft standards and assess them against international developments.
The entire process of reform will extend well beyond 2000. However, the bulk of the analytical work associated with the development of the new framework should be completed by the end of this year.
Concurrent with these developments is a project aimed at introducing even greater sophistication into the identification of risk by general insurers. APRA is seeking to encourage regulated institutions to focus on their individual risk profiles through the use of more sophisticated internal risk management techniques. The objective is to provide incentives for regulated institutions to develop tailored “internal models” for the identification and measurement of risk. APRA will be investigating the feasibility of forming an industry consultative group to study the options for more widespread use of risk modelling in the industry, and the potential for using resulting risk measures to determine regulatory solvency requirements.
Improved capital adequacy and solvency measures, and improved risk management practices, will not solve all the problems inevitably experienced by financial institutions from time to time. However, APRA is convinced that improvements in this field will enhance safety and soundness in the financial sector, to the benefit of the community as a whole. Work on capital standards for ADIs, and the above-mentioned proposals for the general insurance sector, represent major policy priorities of APRA in 2000 and beyond.
Prudential responses to developments in the financial system and financial markets can take numerous forms. They may be proactive initiatives of the supervisor to potential market developments, or more reactive in nature, representing a response to issues or problems that highlight the need for a certain capability or technique. New supervisory questions, necessitating regulatory response, can also arise as a result of technological or market developments that lead to changes in the way that financial institutions conduct their activities, or in the emergence of new and often hard to quantify forms of risk.
A number of projects are currently in train that seek to analyse new or expected developments in the market and formulate an appropriate regulatory response.
There has been rapid development in electronic commerce in recent years. It has taken place against a background in which banks and other financial institutions have sought to lower costs, access a wider range of customers than is possible through their more traditional means (such as branches, shop fronts,etc) and provide greater flexibility in services available to customers. A growing number of financial institutions now offer services through electronic channels, enabling customers to obtain account balances, transfer funds between accounts, pay bills, purchase insurance policies and invest funds via theInternet. This process has been facilitated by improvements in technology and security, and by the rapid growth in Internet usage within the community.
Prudential regulators must understand the impact of these new activities on the risk profile of financial institutions and assess whether current supervisory policies adequately address the risks. Whilst many of the risks inherent in electronic commerce are already addressed in respect of existing operations, such as ATMs, EFTPOS and phone banking, newer forms of electronic commerce, including through the Internet, raise new issues and risks. A major focus of attention to date has been Internet banking and the operational and security risks that this poses – for example, a system malfunction or failure, security breach or computer virus could result in substantial reputation and legal issues. Electronic banking is also associated with a heavy reliance on third-party service providers, with many financial institutions forming strategic alliances with information technology and telecommunications providers, which often extends to partial ownership or an equity investment. This is an area of increasing importance to APRA.
APRA is examining these developments, in the light of experience both in Australia and overseas. It needs to ensure that financial institutions have in place adequate processes to identify, assess, manage and control the risks associated with electronic commerce and that rapid expansion by institutions in this area does not give rise to increased risk within the financial system. At the same time,APRA is mindful not to stifle innovation in this developing area through the application of unnecessary rules and regulations.
Over the course of 2000, we will be issuing a consultative document to stimulate debate on this increasingly important activity.
There has been rapid expansion over the past decade in the range of techniques used for the mitigation of credit risk. One important and relatively recent development is the use of credit derivatives (financial instruments that transfer credit risk between counterparties, much like guarantees, but which lend themselves more readily to trading in financial markets). Worldwide, the credit derivatives market has been growing strongly. In Australia, the market is still in its infancy, however, it is developing and the use of credit derivatives by Australian institutions is becoming more prevalent.
The evolution of traded credit-linked instruments raises many important challenges for the users, for the market, and for financial and prudential supervisors. Many of the key questions revolve around the certainty with which the transfer of credit risk can be achieved. In particular, there are obvious difficulties associated with quantifying the amount of protection or exposure generated by instruments that have a payoff contingent on the credit qualities of individual entities. This problem is accentuated in Australia by a lack of good quality data on creditworthiness. There are also concerns surrounding the absence of standard market practices and legal documentation for many types of credit derivative products. Additionally, the impact of credit derivative transactions on the transparency of individual credit portfolios, and on credit markets more generally, is an important consideration.
Notwithstanding these challenges, credit derivatives offer significant potential benefits to financial institutions, in particular ADIs, and to the wider economy. Like other credit risk mitigation techniques, credit derivatives enable financial institutions to reshape their portfolios in unlimited ways and can play a key role in prudent credit risk management. For this reason, APRA is developing guidelines that integrate credit derivatives into the existing ADI capital adequacy framework. APRA’s objective is to develop a set of guidelines that capture the exposures created by credit derivative instruments, while also recognising their potential to significantly reduce credit risk and hence the benefits they offer as risk management tools.
In April 1999, APRA released to the market for comment a Policy Discussion Paper detailing these proposed guidelines. The comments received were diverse, reflecting the relative immaturity of the credit derivatives market in Australia and the differing levels of experience with credit derivatives amongst institutions.
After taking into account industry comments, draft guidelines were issued to the industry in December 1999.9 It is APRA’s intention to release more definitive guidelines in the near future, although any guidelines are not intended to be final or all encompassing. Further change is likely to flow from the review of the Capital Accord, alluded to above.
Moreover, as more complicated credit derivative structures become commonplace, APRA will be endeavouring to accommodate these new products within its credit derivative framework. The guidelines will provide a degree of regulatory certainty for ADIs wishing to begin making use of these products.
Work on both of these topics is likely to be on-going over the next few years as the growth and, no doubt, complexity of electronic commerce expands, and as the market for instruments to mitigate risk increases in size and sophistication. APRA will monitor developments in these and related fields with a view to ensuring that supervisory practice keeps pace with market developments.
In addition to the specific policy issues discussed, APRA maintains a strong research program aimed at advancing its state of knowledge in other areas. Research focuses on technical issues linked to projects as diverse as credit modelling and assessing capital adequacy in ADIs, to programming work associated with improving APRA’s access to data on financial institutions.
One important project over the next 12 months will be developing a framework in which the benefits and costs of prudential regulation can be considered more rigorously. This is an extremely difficult issue conceptually, and some other prudential regulators are turning their attention to the same question (most notably the Financial Services Authority in the United Kingdom). Increasingly, however, financial regulators are being called upon, quite reasonably, by government and industry to demonstrate the value which they add in a modern financial system. At this stage, the tools necessary to reach views on this matter are poorly developed worldwide. APRA hopes to be able to make some contribution to research in this field.
Ensuring effective training is a constant challenge in a relatively new organisation undergoing significant structural change. This is particularly so with regard to technical training (knowledge of the industries being supervised, knowledge of techniques of supervision, etc). Traditionally, the “policy side” of APRA has played a role in developing internal training programs and seeking to expand knowledge in some of the more technical areas of supervision. For example, the Consulting Services area in APRA’s Policy, Research and Consulting Division contains experts in credit, market and operational risk. One objective of this group is to ensure that this knowledge base is available and transferred as far as practicable to allparts of the organisation. Similarly, the policy initiatives discussed previously, such as the harmonisation project and review of solvency and capital adequacy arrangements, are ultimately being applied by staff in the operational divisions of APRA. Policy staff, therefore, have an important task in ensuring all APRA staff are up-to-date and informed about supervisory initiatives and developments.
As well as internal training,APRA has many requests for training from external sources. Not all requests can be accommodated, however, we do undertake to provide assistance to a number of overseas supervisory bodies, with a particular emphasis on those in the Asia-Pacific region. APRA is currently providing a major training and technical assistance program to the Bank of Thailand and, in recent times, has provided assistance to Malaysia, Vietnam and China, amongst others.
The challenges confronting APRA as Australia’s new prudential regulator are exciting. The task of effectively supervising institutions requires a heavy on-going commitment to both on-line supervision and to policy development. A good start has been made and progress on the work program outlined in this paper will make a further significant contribution to improving the soundness and the efficiency of the Australian financial system.
1 A range of policy-related material issued to date can be found on APRA’s website (www.apra.gov.au/policy).
2 For example, the Council of Financial Supervisors (COFS) was formed in 1992. It comprised the Reserve Bank of Australia, the Insurance and Superannuation Commission, the Australian Financial Institutions Commission, and the Australian Securities Commission. Its main goal was to ensure appropriate liaison and coordination between the regulatory agencies. Since 1998, the newly formed Council of Financial Regulators (COFR) has comprised APRA, the Reserve Bank of Australia and the Australian Securities and Investments Commission.
3 Meaning that breaches can be used by APRA, at its discretion, to trigger formal “directions” which are enforceable in the Federal court.
4 The Joint Forum is a tripartite group of banking, insurance and securities regulators sponsored by the Basel Committee on Banking Supervision, the International Association of Insurance Supervisors and the International Organisation of Securities Commissions, which coordinates international responses to cross-sectoral issues.
5 Retail superannuation typically has some capital backing eg approved trustees generally require $5m capital, and retirement savings accounts are backed by the capital of any ADI or life company involved.
6 Recent publications by the Basel Committee on Banking Supervision include:
- Credit Risk Modelling: Current Practices and Applications, April 1999;
- A New Capital Adequacy Framework, June 1999;
- Industry Views on Credit Risk Mitigation, January 2000. These and other papers can be found on the Bank for International Settlements website (www.bis.org).
7 APRA’s submission is available on the website.
8 Basel Committee on Banking Supervision, Range of Practices in Banks’ Internal Ratings Systems, January 2000. This paper can be found on the Bank for International Settlements website.
9 This took the form of Guidance Notes to Prudential Standard C1(Capital Adequacy of Banks) and Prudential Standard C3 (Market Risk). Further comment was sought from the market by the end of February 2000.