Speech The Role of Disclosure in Bank Supervision
Chief Manager, Bank Supervision Department
Seventh Melbourne Money & Finance Conference
Ballarat, Victoria – –
In major developed countries, regulation of financial institutions, in particular banks, is carried out by a combination of three approaches. The first involves the imposition of minimum prudential standards and the monitoring of compliance. The second involves the supervisor assessing, on the basis of qualitative and quantitative information supplied by banks, the quality of a bank's internal risk-management procedures. The third approach is to rely on disclosure of information to the public. The first two approaches are concerned with minimising the likelihood of institutional insolvency, although they recognise that this eventuality cannot be eliminated. The third is designed to assist consumer/investor choice, and to introduce an element of market discipline on institutions. This tripartite approach is widely accepted and most of the mainstream debate on regulation focuses on how supervisory techniques can be developed to keep pace with evolving trends in the financial sector.
An alternative view is that prudential supervision – that is the first two approaches outlined above – could be eliminated and that increased disclosure could suffice on its own. The argument is based on a view that regulation of banks is ineffective, that markets themselves deliver better outcomes, and that regulation generates excessive moral hazard. This approach is familiar to financial economists as the view of the ‘free banking’ school.
This paper first restates the view that adequate disclosure is a crucial part of the process of bank supervision. There is nothing new in this as it is widely accepted by financial economists and bank supervisors. Indeed, one of the reasons that disclosure has increased appreciably over recent years is the pressure applied by supervisors. Some examples of this are given in a later section of this paper. The paper then argues that while disclosure is a useful component of bank supervision, it will never be a substitute for conventional supervision as we know it. It argues that the ‘free banking’ school is naïve conceptually in thinking this could be the case, and that it does not spell out what it means by disclosure. The assumption is always made that there is a body of objective and undisclosed information which, if made available, would enable the public to make a realistic assessment of the financial position of a bank.
The structure of the paper is to start in Section 2 with a traditional view of the role of disclosure in financial regulation and some evidence on the extent of disclosure of information within the Australian banking system. Section 3 considers the limitations of disclosure, particularly when viewed as a total or partial alternative to a traditional prudential regime while Section 4 looks at the role of prudential supervision. Section 5 sets out brief conclusions.
2 The Role and Extent of Disclosure
Financial regulators have been amongst the most vocal in outlining the benefits of an effective structure of disclosure. A discussion paper issued by the Euro-currency Standing Committee of G10 Central Banks in September 1994 (the ‘Fisher Report’) was one of the clearest statements to date on the importance of and the rationale for improved levels of financial disclosure by institutions. The Basle Committee on Banking Supervision issued a similar document in November 1995, describing the role of disclosure in relation to banks’ trading and derivatives activities in the following terms:
‘The objectives of supervision can be reinforced through the public disclosure of meaningful information about how the full range of banks' trading and derivatives activities contribute to the institution's overall risk profile and profitability, and how well it manages the risks arising from these activities. Meaningful and accurate information, reported in a timely manner provides an important foundation for the decisions of market participants. Well-informed investors, depositors, creditors and counterparties can impose strong market discipline on an institution to manage its activities in a manner that is both prudent and consistent with its stated business objectives. Strong internal risk management and controls by banks, reinforced through prudential supervision and enhanced public disclosure practices provide a sound framework for fostering market stability in an environment of rapid financial innovation and increasing complexity’.
The above quote is a good summary of the mainstream view on the merits of disclosure and its relation to prudential supervision more generally defined – as indicated above, it reflects the position of most financial economists and bank supervisors. Given its importance, a good deal of effort has been put into bringing disclosure levels in banks up to a high standard. A significant part of the history of bank supervision in Australia and elsewhere over the past decade has been the encouragement given to improving both the quantity and quality of financial information provided to the public. Advocates of wholly disclosure-based regulatory regimes, however, assume that disclosure levels by banks are, in fact, relatively low and require a major upgrading. This position is difficult to support.
Table 1, for example, looks in broad terms at some aspects of disclosure in the financial statements of the major Australian banks over three separate years – 1986, 1991 and 1995/96. It shows quantitative information on capital, asset portfolios, asset quality, tradeable security holdings, bad-debt expenses and derivatives. It outlines also a number of qualitative matters disclosed in financial statements relating to capital, risks and risk-management practices and derivatives, all of which might be expected to provide useful information to those dealing with banks.
|Capital||Amount of share capital. Movements in reserves. Amount and some details of loan capital.||Details of all capital types disaggregated. Details of loan capital by instrument and by maturity. Capital adequacy information: full detail of capital types, capital ratios, and risk-weighted assets.||Details of all capital types disaggregated. Movements in shareholders' equity detailed. Details of loan capital by instrument and by maturity. Capital adequacy information: full detail of capital types, capital ratios, and risk-weighted assets.|
|Assets||Separation between types of assets. Most detail on fixed assets and investments.||Balance sheet separation between types of assets – more detailed. Disaggregation of all classes of assets. Liquid/short term asset detail by origin (domestic or offshore). Amounts due from other banks separately detailed and distinguished between Australia and offshore. Details of trading and investment securities. Details of loans/advances/acceptances by type, origin, industry and maturity. Average balances and volumes and average interest rates by asset class.||Disaggregation of all classes of assets. Liquid/short term assets detail by origin (domestic or offshore). Amounts due from other banks separately identified and distinguished between Australia and offshore and detailed according to maturity. Details of trading and investment securities. Details of loans/advances/acceptances by type, origin, industry and maturity. Average balances and volumes and average interest rates by asset class. Volume and rate analysis. Maturity profile of assets. Fair value information.|
|Asset Quality||Line item on non-accrual loans.||Details of non performing loans by class of non-performance.||Details of impaired assets by class. Non-accrual loan amounts, provisions and interest income foregone separately detailed in some cases. Origin of loan and location of exposure identified in some cases.|
|Securities||Book and market values of investment securities by type.||Separation between trading and investment securities then between listed and unlisted securities. Book values and market values of investment securities disclosed.||Separation between trading and investment securities then between listed and unlisted securities. Book values and market values of investment securities disclosed. In some cases investment securities detailed by yield and by maturity with unrecognised gains/losses disclosed.|
|Derivatives||No disclosure.||Notional amounts and credit equivalents disclosed by broad derivative type.||Notional principal, replacement cost and credit equivalent disclosed by instrument type separating between foreign exchange and interest rate exposures. Counterparty profile, maturity profile by class. Closing and average fair values, realised and unrealised profits and losses by instrument type. Notional amount and market values according to purpose for use. Maturity profile of expected recognition of income/losses.|
|Bad Debt Expenses||Amount of movements in provisions.||Details of movements in provision separating between specific and general provisions. Write-offs and recoveries detailed by industry classification.||Details of movements in provision separating between specific and general provisions. Write-offs and recoveries detailed by industry classification. Provisioning ratios.|
|Capital||Generally no discussion. Some mention of capital adequacy and capital resources.||Details of loan capital. Discussion of capital adequacy framework and discussion of capital resources.||Details of loan capital. Discussion of capital adequacy framework and discussion of capital resources.|
|Risks & Risk Management||Generally no disclosure. In one case a brief discussion of credit, liquidity and balance sheet interest rate risk management.||Generally no disclosure.||Discussion of organisational framework for risk management including the role of audit/compliance function. Discussion of operational, legal, credit, liquidity and traded and balance sheet market risks, how they arise and the operational procedures used to control them. Details of the approaches for managing credit risk. Details of the methodology used to measure market risk and the average, peak, and closing amount of market risk according to the methodology.|
|Derivatives||No disclosure.||Mention of use in context of off-balance sheet assets in capital adequacy.||Discussion of the nature of derivative instruments used and description of what they are. Discussion of the purposes for which derivatives are used. Details of the risks on derivatives transactions, both credit and market, the amount of these risks according to the measurement methodology and how they are controlled. Details of the amount of income from derivatives activities. Discussion of the accounting treatment of derivatives and recoveries by industry became common. Market-related instruments were shown in greater detail with security portfolios separated into traded and investment categories, with further separation of portfolios into those with quoted and unquoted prices. Specifics of derivatives volumes or exposures remained sparse, other than information on the notional principal amounts outstanding by major derivative class. Little in the way of qualitative information had found its way into published financial statements by this stage, though, prompted by the introduction of the new capital framework, banks had begun providing descriptions of the capital-adequacy framework and their risk-adjusted assets.|
From the material presented in Table 1, the following broad observations can be made:
- A decade ago disclosures were very limited in scope, focusing mainly on traditional profit and loss and balance sheet data. Disaggregation of bank asset categories was slim and examination of asset quality was restricted largely to aggregate non-accrual loan data and movements in associated provisions. Data on market-related instruments and transactions were either non existent or extremely limited. There was little if anything in the way of discussion about the financial instruments and products offered by the bank, their characteristics or their risks. There was little or no discussion on risk-management practices in use within banks.
- By 1991, the volume of information disclosed by banks had grown substantially. The general trend was towards more areas of disclosure and more disaggregation within the disclosed areas than seen previously. Data on capital covered not just aggregates but details of separate capital instruments and their maturities. Similarly, information became available on asset composition by type, by region, by maturity and by industry. Average interest rates for the main asset and liability items were provided. Details of impaired assets by class and write offs.
- By 1996, further major developments were evident. Aside from information on all the areas previously shown much more quantitative information was provided on derivatives. This information included credit-equivalent values by risk type, replacement and market values by purpose of holding, realised and unrealised losses and maturity profiles of expected income and losses. Banks began presenting value-at-risk data in relation to their trading activities, covering such aspects as average, peak and closing levels of risk as measured by their in-house risk-management methodologies. There was evidence also of a significant expansion in qualitative disclosures – much more on risk-management methodology and practice including the role of internal audit and compliance, as well as discussion of financial instruments and products and the various dimensions of risk present within the bank – operational, legal, credit, market and balance sheet.
To summarise, the decade to 1996 saw major advances in the quantity and quality of information being presented in banks’ financial statements. These developments were being forced by the market, by regulators and to a lesser extent by the accounting profession. At least some of the data presented publicly by banks over this period had their origin as information collected by supervisors for the purposes of their own analysis (Attachment 1 looks at the issue of disclosure from a different perspective. It compares the disclosure by the four major Australian banks against the new disclosure requirements issued by the Reserve Bank of New Zealand).
Also significant to any examination of the role of disclosure in the regulatory framework are possible trends in future disclosures. Three messages emerge from the ongoing international discussion on this subject:
- The likelihood is that the next area for improved disclosure will be further detail on the measurement and management of market risk (including risks associated with derivatives). More detailed disaggregation of market positions and potential exposures as measured by internal value-at-risk models can be expected as well as more on the results of back-testing exercises and stress testing of internal risk-measurement models. Over time, similar techniques seem likely to be applied to the measurement of interest-rate risk on the whole balance sheet. In relation to market risk, one clear possibility is the evolution of a disclosure regime based on the output of internal models as approved by bank supervisors under the new Basle market-risk guidelines (Australia has already signalled its intention to follow this path).
- Additional disclosures on matters associated with asset quality and traditional credit risk seem unlikely. This is the area where problems within banking systems have been most pronounced, so this seems to be a disappointing conclusion. However, the main reason for this is that much is already disclosed and, as will be outlined in the following section, there is a conceptual problem in taking meaningful disclosure much further. While it would be possible to provide further disaggregation of asset categories, that path can very quickly intrude into areas involving the disclosure of proprietary information by banks (for example, disclosure of individual counterparty names).
- More detailed qualitative discussion of risk-management by banks is on the agenda. The focus here is likely to be how banks identify and manage their many risks as opposed to the actual magnitude of the risks.
The conclusion to be drawn from this analysis of the major Australian banks is not that disclosure levels are necessarily at optimal levels or that there is no room for further improvements. There clearly is, but the task of increasing disclosure in a meaningful way is a difficult and slow one. It is hard to believe that there will be a great breakthrough as some previously private information is shifted into the public arena. Having said that, efforts are still being made to improve disclosure wherever possible and each year further progress is made. That process will continue. To the extent that suggestions come forward from the public for the disclosure of new pieces of information by banks that will add genuine value without excessive compliance costs, then the Reserve Bank will explore those avenues.
An important question that flows from this relatively simple analysis is to what extent advocates of a wholly disclosure-based regime see current levels of disclosure as sufficient to produce effective market discipline. Or are current and prospective disclosure levels still regarded as incomplete? If the latter is the case, then the onus must be on those calling for greater reliance on disclosure to specify what additional information would be required to make disclosure an effective stand-alone regulatory technique.
In my mind, despite the clear benefits of disclosure, the source and nature of problems which tend to arise within the banking sector are not of a type that can be addressed solely within a disclosure regime. This view is not based on the proposition that the general public will not be capable of understanding or assimilating the large volumes of information which banks disclose (though that is likely to be the case). Rather, the reasons have to do with the nature of banking and the structure and sensitivity of bank balance sheets which make assessing the solvency of a bank at any point in time exceedingly difficult even for the financially sophisticated. These themes are developed in the following Section.
3 Limitations to Disclosure
The debate over whether disclosure could or could not be an effective stand-alone approach to regulation of the banking system must be conducted largely on academic grounds as there is no single example of a country adopting, or at any time having adopted, a regulatory regime based solely or even mainly on disclosure. Even in New Zealand, where the supervisory approach to banking supervision is often described as one based on disclosure, it is in fact a fairly conventional approach based on a mixture of prudential regulation and a relatively demanding disclosure regime which requires that all statistical information provided to the supervisor be published and available to all market participants.
Essentially, the argument for a pure disclosure regime is a theoretical one. The pros and cons are not examined in detail in this paper, but there is one aspect that should be mentioned. Central to the argument against bank supervision or any other form of depositor protection, is the view that it sets up an expectation in the public's mind that their deposits will be wholly or largely protected in the case of an individual or systemic collapse. This, it is held, will set up a moral hazard which will encourage banks and depositors to take excessive risks. In this view, if the government clearly states that it will not protect depositors and shows this by abandoning any forms of depositor protection, such as bank supervision, then the public and the banks will adjust their behaviour away from excessive risk taking to the optimal level. In essence, they believe that an institutional change will eliminate the public expectation that gives rise to the moral hazard.
The mainstream view of financial economists and bank supervisors is very different. In this view, it is impossible in a democracy to eliminate the public expectation and moral hazard, no matter what institutional changes occur. Regardless of what the government says it will do in a future financial crisis, the public ‘know’ that it will not allow a large number of depositors (and voters) to lose their savings when the crunch comes. The history of banking in the twentieth century supports this view. In this sense, the moral hazard will always be there, and it is the job of bank supervision to limit the excesses that it might otherwise give rise to. This is the central theoretical argument for bank supervision, and one that is held by many free market economists who are, for the most part, quite unsympathetic to regulation (financial or otherwise).
The argument that supervision is required and that disclosure is not sufficient to form the basis of the regulatory structure is based on other considerations as well. As in some other industries, the presence of externalities and information asymetries in banking are viewed as important (see Diamond and Dybvig (1984), Goodhart (1988), Dow (1996) and Dale (1996)). The view is based also on an appreciation of the history of banking problems and the dynamics of bank balance sheets during periods of financial disturbance. It emphasises the complexity of valuing bank balance sheets (and thus bank solvency and depositor safety) and the tendency for valuation changes to occur rapidly and in a manner which is difficult to forecast. Taken together, these factors greatly limit the extent to which disclosure can be relied upon too heavily within the overall regulatory framework.
The problem of valuation of bank portfolios arises largely from the nature of banking business. Banks are unique in that their core activity involves the process of liquidity or maturity transformation – they write contracts with customers offering the return of a fixed nominal value over varying terms (deposits) and support those liabilities with a portfolio of assets which are largely illiquid and therefore hard to value. The usual response to the ‘imbalance’ between the asset and liability sides of the accounts is to provide more information to assist depositors and others in the task of assessing those asset values and determining whether or not the bank (and the deposits) are safe. As noted above, this approach has been pushed a fair way and calls have been made for further disclosure. The fundamental problem, however, is not that the information upon which judgements on valuation need to be made is available and undisclosed, but that the information often does not exist in the normal sense of the word.
This problem can be illustrated by reference to the measurement of banks' problem loans. In the process of preparing financial accounts, it is necessary to determine the value of loans which have gone bad (so they can be written off) and those which may have lost permanent value and/or may be in the process of going bad (so provisions can be made against them). These judgements flow directly into measurements of profit and loss and into measures of a bank's capital and its capital ratio (all important indicators of prudential soundness). Where assets are highly liquid and tradeable then markets provide accurate valuations. Valuation becomes more problematical where tradeable assets are marketable but relatively illiquid, for example, holdings of debt of emerging markets. It becomes highly uncertain and subjective, however, when markets do not exist in which to value the assets, as is the position for the vast bulk of banks' loan assets.
Valuation of illiquid assets, therefore, may be better thought of in terms of forecasts or predictions of value than as objective indicators of market value. Those subjective measurements must still be reflected in the financial accounts. What cannot be readily measured in a quantitative sense, however, is the myriad of assumptions and predictions which must be made not only about the intrinsic quality of the loan in question, but about the likely economic cycle, movements in interest rates, exchange rates, commodity or equity prices, and asset prices generally, that may influence its ultimate value.
History shows that macroeconomic cycles or events are the predominant influence on bank asset values and on the soundness of banks and the banking system. The situation of one bank becoming insolvent while the rest of the banking system remains in good health is a rare one. (It is even rarer for a bank big enough to have systemic implications to find itself in this position.) When banking problems come, they come in waves – capital ratios of individual banks and the entire banking system shift rapidly, not so much as a response to scientifically estimated changes in value but as a result of changes in sentiment of bank management in relation to the factors driving underlying asset values. This was the experience of the banking sector in Australian over the late 1980s and early 1990s as well as in a range of other countries. The sudden recognition that asset portfolios were worth much less than had previously been estimated, due largely to changing perceptions about the commercial property cycle, saw a significant and rapid write down of asset values and a direct flow on to measures of capital.
Perhaps the most dramatic example of this process was the experience around the turn of this decade by the State Bank of South Australia, where the State Government owner was called upon to provide support for losses which eventually reached around $3 billion. An initial assessment that losses of the bank could amount to around $50 million was revised over a very short period to an estimated loss of $1 billion. As noted above, the eventual loss was a significantly more than that. In other words, changed perceptions over the volume of problems loans, and thus the need for write offs and provisioning, saw what had been viewed as a ‘solvent’ institution transformed over a relatively short time into an ‘insolvent’ institution.
Disclosures which provided early warning of such problems would be highly valuable. There is, however, little evidence that the relevant information becomes available until it is too late to respond. Corrigan (1989) notes that when Continental Illinios and a range of Texan banks got into difficulty in the 1980s, they were rated highly (investment grade) over the very period that the problem loan portfolios (as they subsequently became) were being built up. There is similar evidence from a range of other countries, including Australia. Disclosure regimes by their very nature generate static, not dynamic forward-looking information. Early assessments that problem portfolios are developing within an institution call for highly subjective judgements, taking into account factors internal and external to the institution itself.
Securitisation, through its ability to ‘liquefy’ otherwise illiquid assets, has the potential to reduce the severity of this valuation problem in the future – in effect, increasing the proportion of the balance sheet that can be marked-to-market. The difficulty is that the exposures most amenable to securitisation are also those which are easiest to value. The more complex valuation issues arise where exposures are non-homogeneous (loans to small and medium business particularly) and these are the most difficult assets to securitise.
It is the structure of banks' balance sheets, therefore, and the difficulty of accurately valuing assets, especially in times of stress, that leads to the conclusion that disclosure-based regimes are insufficient on their own, and that reliance on such an approach would compromise stability of the financial system. The argument is well understood. In a non-regulated system without deposit protection or like arrangements, it becomes rational for depositors to flee from institutions at the very first sign or suggestion of problems (real or imagined). Those departing early from suspect banks maximise their chances of fully recovering their funds. Given the broad similarities between the balance sheets of most banks, suspicions of problems in one bank can lead to doubts about others. This is especially likely given that serious or system-wide problems in the banking sector are linked invariably to macro events. The market response to the suspicion of banking problems, therefore, tends to be abrupt and inconsistent with system stability. Supervisory measures seek to address those issues (see Section 4 for details).
The nature of banking problems, and some of the limitations of point-in-time disclosures, suggests that there may be scope for increased emphasis on qualitative rather than quantitative disclosure – how banks manage risk as opposed to how much risk is present at a particular point in time. As outlined in Section 2, banks have been providing more and more information on the broad area of risk management and that trend is set to continue. History shows, however, that it is very difficult to provide valuable qualitative information within a disclosure regime. One of the most disturbing aspects of the Barings collapse in 1995 was, not how much could be lost over a short time period, but how effective their risk-management process could appear on paper, yet how flawed it could be in practice.
The range of problems associated with the valuation of assets, the conceptual problems of price determination where no markets exist, the speed and lack of predictability with which exposures can change, and the widespread as opposed to sporadic nature of banking problems, all carry important implications for disclosure-based regulatory arrangements.
On this broad subject, Alan Greenspan (1994) has noted:
A generation ago, a month old balance sheet was fairly indicative of the current state of an institution. Today, owing to the proliferation of transactions, a day old balance sheet can be obsolete.
Corrigan (1989) has argued along similar lines:
it is sheer fantasy to assume that individual investors and depositors – and perhaps even large and relatively sophisticated investors and depositors – can make truly informed judgements about highly complex financial instruments and institutions.
Crockett (1995) argued in the following terms:
Disciplinary forces have probably become weaker because changes in the structure and activities of the banking firm have made it more opaque. Greater opaqueness has resulted from the increasing functional and geographic complexity of the operations, from the sophisticated nature of new instruments and from the unprecedented speed with which exposures can change.
The disturbing possibility is that the increasing complexity within banking and financial markets, typified by the growth in derivative instruments which can be used quickly to vary risk, might lead to less reliance on public disclosure than may have been the case in the past when banking was less complex. This position is interesting in that it runs counter to those advocating a greater role for disclosure, in part, because of the greater complexity and sophistication in the financial sector. More generally, arguments outlined by Greenspan, Corrigan and Crockett do not dismiss the role or desirability of high-quality disclosure from financial institutions. Their conclusion, rather, is that in an increasingly complex financial system, disclosure alone cannot represent a viable regulatory option for the financial system.
4 The Role of Prudential Supervision
If disclosed information is so difficult to interpret, why should we expect supervisors to be able to do it better than the market? Does a role for the prudential supervisor arise because of his or her special ability to read and interpret bank data? The answer is that they find interpretation of banking data as complex and difficult as most other people. As a result, they have developed approaches that reduce obstacles imposed by the imprecision of statistical data on the banking sector. These approaches are, or relate to:
The setting of standards
Supervisors have played a leading role in the setting of minimum prudential standards within the banking system. These include the broad range of criteria used to assess entry to the banking system, capital adequacy standards for credit risk and market risk, standards set for defining and measuring problem loans and impaired assets and standards covering risk-management practices. Many of those standards have been widely accepted, even outside traditionally supervised areas. Within the finance sector, many supervisory standards have come to be adopted as industry benchmarks. Some supervisory standards have also come to be incorporated into accounting rules and regulations (the impaired asset arrangements devised in Australia in 1993 are now incorporated into Accounting Exposure Draft 63 which relates to the disclosure of information by financial institutions).
An interesting and on-going issue is the extent to which it is the role of supervisors to encourage banks towards ‘best practice’ in relation to standards or whether, as is currently the case, the role of supervisors should be to prescribe regulatory minimums. Another is the range of matters that should be captured by supervisory standards. At what point does the setting of supervisory standards intrude on the rights of bank managements and boards to run their bank?
More generally, it is not clear how the process of standard setting would emerge outside the traditional supervisory framework. While self regulation within the finance sector is one possibility, it is not evident that any standards which may arise would take wider system-stability considerations into account.
Accessing and interpreting information
Irrespective of the nature of the disclosure regime, supervisors will always be capable of accessing information that banks would not be prepared, or not be in a position, to disclose to the public. This includes quantitative information that banks would regard as being held in confidence, such as customer details, individual counterparty exposures and specifics of risk-management methodologies. It also includes a large body of qualitative information, on matters to do with the risk-management framework and how it works in practice, on operational risks within the institution and issues associated with the quality and competence of staff and management. None of these areas are amenable to disclosure in the traditional sense. Yet, as noted in Section 3, this body of qualitative information about how the bank actually functions can ultimately be far more important than quantitative ‘point in time’ statistics on an institution.
Comparing one bank with another
Using information to which they have access, supervisors can compare banks and assess relative risk and risk-management practices. They can identify risk ‘outliers’ or exceptions and use this information to question bank management. Just as there are economies of scale involved in the assessment of single institutions, there are economies involved in such system-wide assessments of the banks. It is difficult to envisage any private mechanism emerging which could reproduce that function in as efficient and cost-effective a fashion.
Improved Crisis Management
The ability to look across the banking system permits greater flexibility and responsiveness to potential problems should they arise. This helps in deciding whether the appropriate response to institutional insolvency or near insolvency is a re-capitalisation, a takeover or merger, or an orderly exit from the market. In contrast, the market response to financial disturbances in banks (once identified) tends to be abrupt and potentially at odds with overall stability within the financial system.
It will never be possible, of course, to ‘prove’ that prudential supervision of the banking system is effective and it will always be possible to point to instances of regulatory failure, big and small. It is possible to show, however, that prudential supervision has the potential to improve significantly on the market solution, through its role as the standard setter in relation to prudential matters in the banking system and its ability to assess otherwise unquantifiable information about the banking sector and to draw comparisons across the system.
5 Summary and Conclusions
The case for a well-developed structure of financial disclosure within the banking sector is widely accepted and nothing in this paper should be read as a rejection or a questioning of the benefits of disclosure. As emphasised throughout, supervisors have been among the most vocal supporters of better disclosure by banks and financial institutions. In calling for total or substantial reliance on a disclosure-based regulatory regime, however, advocates must be very specific about the extra information that they see as necessary for such a system to work effectively. It is not enough simply to call for more information, broadly defined. To the extent that specific and realistic suggestions can be made, then there would be strong support by supervisors to see that information made available. For all the reasons outlined in this paper, however, we believe that this approach would not be enough. In particular, we would see wholly disclosure-based systems as inconsistent with the fundamental objective underlying the current structure of prudential supervision covering banks – the maintenance of financial system stability.
See Dowd (1995) and some of the ideas outlined in Chapter 2 of the Treasury Submission to the Wallis Inquiry. 
See Basle Committee (1995), pp 4. 
The Governor of the Reserve Bank of New Zealand, Dr Brash, described the new arrangements in the following terms:
While the emphasis on market disciplines through disclosure lies at the core of the new approach, it is important to note that the Bank continues to have responsibility for supervising the banking system
He goes on to note that the Reserve Bank of New Zealand:
- continues to register new banks:
- subjects them to minimum capital requirements consistent with the Basle Accord;
- monitors banks (via a Bank Supervision Department);
- conducts regular consultations with banks' senior management, and
- retains the capacity to respond to financial distress or bank failure where there is a threat to the stability of the financial system.
See Brash (1996).
See Report of the Auditor General (1993). 
Basle Committee on Banking Supervision, ‘Public Disclosure of Banks' Trading and Derivatives Activities’, Report of the off-balance sheet sub-group, November 1995.
Brash, D (1996), ‘Briefing of the Reserve Bank of New Zealand’, Reserve Bank of New Zealand, October 1996.
Corrigan, EG (1989), ‘A Perspective on Recent Financial Disruptions’, Federal Reserve Bank of New York Review, Winter 1989/90.
Crockett, A (1995), ‘New Horizons in Banking – the Challenge of Global
British Bankers’ Association Inaugural Annual Lecture London, 13 September 1995.
Dale, R (1996), ‘Debt, Financial Fragility and Systemic Risk’, (Second Edition), Oxford University Press, Oxford.
Department of the Treasury (1996), ‘Submission to the Financial System Inquiry’, Canberra.
Diamond, D & Dybvig, P. (1984), ‘Bank Runs, Deposit Insurance and Liquidity’, Journal of Political Economy, 91, June, pp 401–419.
Dowd, K. (1996), ‘The Case for Financial Laissez-Faire’, The Economic Journal, 106, May, pp 679–687.
Euro-Currency Standing Committee of G10 Central Banks ‘A Discussion paper on Public Disclosure of Market and Credit Risks by Financial Institutions’, September 1994.
Goodhart, C (1988), ‘The Evolution of Central Banks’, MIT Press, Cambridge.
New Zealand Department of Internal Affairs ‘Registered Bank Disclosure Statements’, New Zealand Gazette, Issue 135, November 1995 and Issue 21, March 1996.
The Auditor-General of South Australia (1993), ‘Report of the Auditor-General on an Investigation into the State Bank of South Australia’, Volume 1.