Finansiell stabilitet 1/2003

Utskrift

Vedlegg I Norges Bank's financial stability report: A review

Norges Bank's financial stability report: A review

Alex Bowen (Bank of England)¹,
Mark O'Brien (IMF)² and
Erling Steigum (Norwegian School of Management BI)³

I. Introduction

In December 2002, Norges Bank invited the authors to draw up a review of their Financial Stability Reports. The terms of reference were described as follows:

“Although the methodology which has been built up over time in different areas has enabled us to develop our stability report further, we believe that regular expert reviews may have a positive influence on further developments. Moreover, we would like to have our reports assessed relative to those produced by other central banks: how we stand with respect to the methodology applied, in particular credit risk, and how we should focus resources in order to maintain overall high quality. At the same time, the format and the scope of the report have been very much designed to make accessible to policy makers, financial institutions and the general public issues that may be rather intricate in analytical terms as well as in terms of resolution methods.”

The review benefited from the authors’ opportunity to discuss the financial stability work of Norges Bank with staff at a seminar in Oslo in March. The main conclusions were presented to the Governor and Board of Norges Bank on 19 March. This report summarises them. It is organised in three parts: first, a discussion of what in principle a financial stability report of a central bank should aim to do; second, an analysis of the strengths and weaknesses of Norges Bank’s Report in light of that discussion, with suggestions for topics which might be covered in the future; and third, proposals for further work by the External Review Panel.

II. What should a financial stability report aim to do?

The salience of financial stability as a public policy objective has increased around the world in the past quarter of a century, as the incidence of banking crises has risen. In a recent historical study of 21 countries, Bordo, Eichengreen, Klingebiel and Martinez-Peria (2001) reported only one banking crisis in the 25 years after 1945, but 19 since 1970. Financial instability has proved to be expensive, in terms of both lost output and fiscal costs, and events have demonstrated that developed market economies can suffer at least as heavily as emerging markets (see, for instance, Hoggarth and Saporta (2001), who suggest that output losses during banking crises can amount to 15% to 20% of annual GDP). Norway itself underwent a serious banking crisis just over a decade ago, as did its neighbours Sweden and Finland. The debt problems of several Asian countries in the late 1990s and the widespread financial market liquidity problems experienced in autumn 1998 have focused policy-makers’ attention on financial stability, as can be seen from the setting up of the Financial Stability Forum and the IMF Financial Stability Assessment Programmes in 1999. Another response has been the development of financial stability reports as a means for presenting central bank analyses of threats to financial stability. Norges Bank was in the field early, but the number of central banks producing reports is multiplying rapidly. It is interesting to note that one of the conclusions of Lars Svensson (Stockholm University) in his review (February 2001) of the operation of monetary policy for the New Zealand Minister of Finance, was:

“I recommend that the Reserve Bank summarise its information about the financial system, including a number of macro-prudential indicators of financial stability, in a regular report, modelled on those published by the Bank of England and Sveriges Riksbank.”

The appropriate scope of such reports depends upon what is meant by financial stability, and what policy-makers aim to achieve by publishing for the general reader rather than doing good by stealth.

Financial stability can be defined narrowly or broadly. At one end of the spectrum, it can be defined as the antithesis of financial crises: episodes in which the banking system temporarily fails to function and the institutional underpinnings of a monetary economy – payments and settlements systems, the acceptability of bank deposits as money – are disrupted. Although such events are rare, they are very costly, so policy-makers need to assess the (usually low) risks of them occurring. Financial crises of this sort are of particular concern to central banks because they disrupt the transmission mechanism of monetary policy. At the other end of the spectrum, financial stability can be regarded as the ruling state of affairs when the efficiency of financial intermediation between ultimate borrowers and ultimate lenders is not subject to significant adverse shocks. If that is the definition adopted, the remit of policy-makers is correspondingly broader, and their analysis must range more widely, extending well beyond the core banking system, and encompassing, for example, asset price bubbles. The assignment of responsibility to the central bank for safeguarding financial stability is less clear-cut if this definition is chosen. Supervisory and competition authorities, for example, would naturally have a close interest.

Hence it would be useful for Norges Bank to have a clear view of what its working definition of financial stability is, to help determine the scope of financial stability analysis and hence resource allocation, to facilitate analytical modelling, to motivate the Financial Stability Reports, and to guide the Reports’ authors. It would, for example, assist staff in deciding whether the risk of default on Brazil’s sovereign debt or the balance sheet problems of Norwegian life insurers merit analysis and coverage. (Of course, financial stability problems in other countries may warrant analysis and exposition because of the lessons for Norway rather than because they constitute a significant threat to the Norwegian financial system themselves.) It should be noted that very few institutions have in fact laid out in public what their working definitions of financial stability are, even in those countries where the central bank has been give an explicit legal or government remit to protect financial stability. There is no generally agreed analogue (yet) to the inflation target adopted in many monetary policy frameworks. Consequently, while Norges Bank should have a working definition of financial stability for its own use, this may change over time, so that it may be premature publicly to discuss it in detail at this stage.

Why should a central bank publish analyses of threats to financial stability?

There are at least two sets of reasons. First, and most important, regular publication may reduce risks to financial stability. It can do so by improving the understanding of risks to financial intermediaries in the economic environment; by alerting financial institutions and market participants to the possible collective impact of their individual actions; and by building a consensus for financial stability and the improvement of the financial infrastructure. It can add value to work undertaken by private agents in the financial sector itself, because a central bank can draw on its macroeconomic expertise and its role in payments and settlements. Also, private agents do not have as strong an incentive to assess the systematic risks in the economic environment, as they are less interested in spill-overs of their actions on to other agents. Of course, private agents will also lack sufficiently strong incentives fully to address systematic risks when such risks have been identified, because they will not expect to capture all the benefits themselves. So publication has to be combined with the promotion of measures to change those incentives or otherwise to constrain private sector behaviour. Those measures may need to be taken by the government, regulators, and / or the central bank itself. Finally, there is a need to educate the public about the costs of infrequent but catastrophic episodes of instability (analogous to the need on the monetary policy side to build a constituency for low inflation), and (if a wider definition of financial stability is adopted) about the costs of disruption to financial intermediation.

The second set of reasons concerns the impact on the central bank. Publication subjects the central bank’s analysis to scrutiny by a wide range of possible critics; it provides a discipline for surveillance work as to its quality, frequency, and timing; and it demonstrates that the central bank is fulfilling its remit. Hence publication can fulfil an important role in improving the accountability and transparency of the central bank. However, a caveat is in order here. It is possible to conceive of circumstances in which publication of a central bank analysis at a time of increasing risk to financial stability might precipitate the very shocks or crisis that the central bank was trying to avoid, by inducing liquidity problems in particular markets, or for particular financial institutions. That danger is probably reduced if the central bank has established a track record in unbiased analysis during a period of low risks to financial stability. In those circumstances, risk-reducing actions taken by private agents in response to the central bank’s analysis are less likely to crystallise liquidity risks. However, some danger remains. But this is only an argument against publication if policy-makers do not think that they have adequate contingency plans in place to deal with the consequences of private sector actions triggered by the publication of their analysis. If they are confident about contingency arrangements, publication can actually help, by reassuring private agents that measures to prevent a systemic crisis (e.g. lender of last resort facilities or government guarantees, appropriately calibrated and timed to avoid increasing moral hazard excessively) are ready to be deployed. It is possible that it is easier for central banks in countries with recent experience of a major financial crisis to decide where to draw the line between what can be discussed and what should not be.

Given the considerations above, what range of material should a financial stability report attempt to cover?

Financial stability surveillance needs to assess the shocks to which financial systems have been subjected, the likelihood of further shocks, and the vulnerability of financial systems to such shocks. Although there is no universal agreement about how to measure the impact of economic developments on financial systems, some principles are needed to assess the materiality of particular shocks. This suggests that Reports should cover:

(i) Macroeconomic developments: empiri-cally, macroeconomic developments have been central to the majority of past banking crises. Hence it is useful to assess regularly the news from the external macroeconomic environment to identify recent shocks and to determine the uncertainty of the outlook. Financial stability analysis is more concerned with downside risks than the central outlook. One way of focusing the macroeconomic work is to consider the extent to which borrowers and lenders are likely to have been surprised by economic developments, and hence to have agreed prices for loans and securities that, ex post, are inappropriate. The degree of uncertainty in the outlook may differ across agents, so the central bank’s own macro forecasts, although relevant, are unlikely to be a sufficient basis for the analysis. Indicators based on financial market prices are a potentially useful source of information about the expectations of financial market participants.

(ii) Vulnerabilities of the financial system’s major counterparties: it is important to have an idea of the scale of the financial system (and thus of the importance of financial intermediation in the economy) and a picture of the pattern of the financial system’s exposures, ideally in terms of their size, the ‘expected loss’ they entail, and the expected volatility of losses. Thus economic analysis of corporate and household balance sheets, income, and liquidity, and of the pricing of loans to firms and individuals, is helpful, as it bears on the ability of debtors to repay loans and the size of losses in the event of defaults. Other important classifications are wholesale vs retail, domestic counterparties vs foreign counterparties, domestic currency vs foreign currency, on-balance-sheet and off-balance-sheet. It is also important to have some sense of intra-system exposures (e.g. the interbank market) to assess the system’s resilience.

(iii) Risks to the financial system: given (i) and (ii), what is the likelihood of losses to financial institutions, what is the danger of liquidity problems, how strong are their buffers of profits and capital, and what is the scope for contagion (either through financial intermediaries’ exposures to each other or through their exposures to financial system ‘infrastructure’)? Another way of characterising these categories is to note that (i) and (ii) concern the assessment of the probability of a range of possible shocks to the financial system, and (iii) is relevant to the assessment of losses given the shock (i.e. ‘probability of default’ and ‘loss given default’). It is in category (iii) in particular that the definition of financial stability adopted will determine the scope of the work (e.g. does it extend to life insurers or fund managers?). In assessing the risk of externalities, amongst the key factors are likely to be the pattern of interbank links and the role of central counterparties. Possible avenues of contagion include via the balance sheets of internationally active financial firms participating in Norwegian financial markets.

III. How does the Norges Bank’s financial stability report measure up?

Overall, the Norges Bank’s Financial Stability Reports do well, judged against the criteria suggested above. They contain a wealth of relevant information, and note the progress that has been made over the last three issues, including the new emphasis on the role of capital market developments in the latest November 2002 issue. This section notes some of their particular strengths; some areas where perhaps greater coverage or deeper analysis might be warranted; and some specific topics that Norges Bank staff might wish to consider for inclusion, subject to the length constraint – the reports are currently kept to about 36 pages. Other central bank reports vary in length considerably, but the Review Panel broadly agreed with the Norges Bank staff view that the current length is about right, in view of the desire to encourage a wide, primarily Norwegian, audience to read it. Maintaining the Reports at around this length would imply that most of the suggested additional topics be considered as special issues for occasional or ‘one-off’ coverage.

III.1 Strengths

Compared with similar products, the Norges Bank’s Financial Stability Reports score well in the following respects:

(i) Application of quantitative modelling techniques.

This includes the use of Norges Bank’s macro model. Norges Bank is a leader in developing a more quantitative approach to calibrating risks, particularly credit risk in the corporate sector, and relating these risks to credit pricing.

(ii) Use of disaggregated data.

Norges Bank has harnessed some very valuable large data sets.

(iii) Offering a qualitative assessment of the overall degree of risk to financial stability and how it has changed.

The prominence given to the judgement in the Governor’s forewords is encouraging. As argued below, there is scope to develop this assessment further, but some other financial stability reports avoid addressing this challenge.

(iv) A focus on the soundness of the banking system.

The analysis of threats to the banking system’s liquidity and solvency, and its overall resilience, should be at the heart of a national financial stability report.

(v) Evidence of an integrated central bank approach.

This is demonstrated by the use of Norges Bank’s macro model and the development of analysis carried out at least in part for monetary policy purposes. The Review Panel was particularly interested to hear about Norges Bank’s internal arrangements for financial stability analysis to feed into the monetary policy-making process.

The Panel also found the Reports to be clearly written, accessible, and easy to navigate. The ability to download the report and the data underlying the charts from Norges Bank’s internet website ensures easy and quick public access. They have not dodged complex but important issues. Hence the Reports compared well with their peers.

III.2 Improving risk assessment

The general criteria outlined above do, however, suggest some broad areas that merit more or deeper coverage, and these are discussed under the eight headings below. If the objective is to increase public awareness of financial developments and thereby improve market discipline, providing even more specific information that it currently does may best achieve this objective in the long run, even if it has to be achieved cautiously and gradually, so as to limit the risk of adverse public reactions. If, however, the objective is simply to disseminate information (as opposed to Norges Bank’s interpretations and analysis), this objective may be best served by improving public access to raw data. The Reports do not yet provide a comprehensive picture of the relative importance of the various risks discussed, how they relate to each other, and through which transmission mechanisms they affect financial stability.

(i) How have the macroeconomic risks changed?

The Reports have not always made this clear. That may be because of a reluctance to duplicate material in Norges Bank’s Inflation Report, but, as argued above, the relevant view point is different from a financial stability prospective. As Norges Bank already has some effective mechanisms for bringing together monetary and financial stability analysis, this should not be too difficult to address. The Financial Stability Report could start with an overall update of Norway’s financial environment, highlighting the main sources of risks and discussing the main transmission mechanisms through which risks are most likely to affect Norway’s financial stability, cross-referencing the Inflation Report as appropriate. For example, both the Bank of England and the National Bank of Belgium begin their discussions on financial stability (respectively, in the December 2002 Financial Stability Review and the 2002 Financial Stability Review) with a review of the main sources of risks that are likely to affect their respective financial systems. A second aspect of the same issue is the lack of comment on the assessments of macro uncertainty by financial market participants.

Two particular macro questions have not been pursued as far as they might be. First, who is bearing foreign exchange risk and over what horizon? This is important for Norway as a small, open economy with a flexible exchange rate, large gross capital inflows and outflows, and a net accumulation of assets abroad. It is evident from reading the Reports that Norwegian savings institutions diversify into foreign assets while foreign banks lend to Norwegian companies, so there is a demand for currency hedging in both directions. But do the hedges offset each other exactly and are any important financial institutions left carrying currency risk? Second, what are the consequences of changes in the level of and uncertainty about the price of oil? Both these are examples of questions concerning the linkage of Norway with the international economy. The Report provides extensive analysis on developments in the international context. For example, in addition to discussing international financial markets, the 2002 May issue contained a one-page discussion on the Enron bankruptcy (on page 11) and another discussion of equal length on the state of the Japanese banks (on page 14). While interesting tin themselves, these analyses offered little discussion of how they related to Norway’s financial stability, and it is not obvious that the links were significant. It would be useful to clarify how Norway’s banking sector is linked to the rest of Europe and the rest of the world and to identify and explore the most important possible contagion transmission channels.

(ii) What is the overall pattern of Norwegian banking system exposures?

This important if the reader is to understand better the main transmission mechanisms through which developments in the financial environment are most likely to affect Norway’s financial stability. Many of the elements have been discussed in past Reports, but it is not easy to piece together the overall picture. A comprehensive snapshot would be a very useful way of putting individual risks in perspective. An example of this approach was presented in the December 2002 Financial Stability Review of the Bank of England, which provided information on foreign and domestic exposure of its financial institutions, the latter being further broken down into exposure to the households, corporate sector and other financial institutions (see chart).

Chart 1

It would be even better if exposures could be weighted by expected loss and/or the expected volatility of loss. Some of the major exposures of Norwegian banks – measured by the stock of lending - may be relatively low risk (e.g. domestic mortgages). In any event, the Report could discuss in more depth the nature and extent of the banks’ exposure to the household and corporate sectors, with special attention to the sectoral breakdown for the latter⁴.

(iii) How vulnerable is the Norwegian banking system to hypothetical shocks?

The Reports do not always convey the likelihood and severity with which identified vulnerabilities may affect Norway’s financial stability, or discuss how a particular vulnerability may affect the stability of the banking sector. For example, the stress tests focus primarily on broad macroeconomic implications and provide little information on their effects on banks’ balance sheets. Similarly, the Report discusses banks’ counterparty exposures, but with little reference to the degree to which counterparty risk may be a source of systemic risk for Norway. In the November issue, a box discussed the results of a survey on banks’ counterparty exposure. But the discussion remained at a level of aggregation that did not give much insight into the vulnerability of those financial institutions with particular relevance to systemic stability. The box informed the reader about the general nature of counterparty exposures, and indicated that Norges Bank has access to detailed information on these exposures and notes that these exposures have risen since the last survey. It also revealed that several banks in the survey have exposures to the same counterparties and to each other. However, it neither indicated the severity of these risks nor interpreted them in the context of Norway’s financial stability.

The Report could discuss
a) the ownership structure of financial institutions and the role of foreign-owned institutions;
b) the relative importance of interbank exposures and exposures to the same counterparty;
c) how these exposures develop over time;
d) how these exposures relate to the stability of the financial system.

The Report tends to rely on a relatively narrow set of assumptions. Assuming that one of the objectives of the Report is to assess Norway’s capacity to withstand shocks, the analysis could include a few more extreme assumptions, that tend towards, and in the case of the stress tests include, worst-case scenarios. This does not seem to be always the case. For example, the May 2002 issue concluded that banks would continue to have a good capital buffer, provided they were able to earn a satisfactory level of earnings in the future. By relying on this – possibly optimistic – assumption, the Report seemed to avoid commenting on the impact that less than satisfactory corporate earnings might have on the stability of Norway’s financial system.

Many central banks and supervisory bodies have been developing macroeconomic ‘stress tests’ of banking systems (see, for example, the article in the forthcoming June 2003 Bank of England Financial Stability Review on the stress tests carried out by the Bank of England and UK Financial Services Authority in connection with the IMF’s Financial Stability Assessment Programme for the United Kingdom). The Panel is aware that Norges Bank staff are developing such stress tests further. The Reports would be a good way of publicising their results and calibrating the change in risks to bank’s capital and profits over time. One of the major issues in designing macro stress tests and then reporting them on a regular basis is how to assess the likelihood of the scenarios chosen and to maintain a similar likelihood in successive tests.

The Reports downplay the systemic risks associated with a possible reversal in real estate prices. So far, rising property prices have buttressed financial stability by increasing financial wealth and boosting consumer confidence. But rising property prices may also put Norway’s financial stability at risk if they start to diverge significantly from fundamentals. An unanticipated and substantial correction in property prices could put pressure on the private sector, especially if this price reversal reduces sharply the value of lending collateral. The May 2002 Report, which highlighted the positive wealth effect of rising property prices, tended to tone down the likelihood that a negative shock to the real estate sector, especially if combined with a rise in interest rates or unemployment, might put at risk Norway’s financial stability. To the authors’ credit, the November 2002 issue included a box discussing the commercial property market. Nonetheless, the box sheds little light on the impact of a negative shock in the real economy on the debt servicing capacity of property companies and hence, on the quality of banks’ loan portfolios. A more in-depth analysis of the real estate sector would be useful to assess more clearly the ability of the financial system to withstand a sharp reversal in property prices⁵.

iv) Why are Norwegian banks comparatively lowly rated?

Financial Stability Reports have alluded to this question, but without offering an entirely convincing answer. Possible explanations include: the size distribution of Norwegian banks relative to the threshold at which ratings agencies start to take an interest; obstacles to risk-based pricing of credit; cost inefficiencies, perhaps related to inadequate scale; and moral hazard, particularly as a result of relatively generous deposit insurance. It would be useful if future Reports could give a clearer view of Norges Bank’s analysis of these explanations.

(v) Can more use be made of ‘benchmarking’ financial sector developments against history and the experience of other countries?

For example, it would be useful to put recent equity price movements in historical perspective, and it would be interesting to know if Norwegian households’ debt-income ratios are high compared with those of other OECD countries. Part of the point of benchmarking is to avoid attempting to ‘over-analyse’ what in fact are relatively insignificant movements in the data. This becomes more important as the range of financial market prices available to monitor expands. In benchmarking data from banks, it may be helpful to define peer groups of institutions with similar business mixes, to make it easier to identify which business lines or classes of exposures are problematic.

(vi) Are life insurers relevant to financial stability?

This is an issue with which many policy institutions around the world have been struggling recently. Behind it lies the question of whether Norges Bank chooses a broad or narrow definition of financial stability. Life insurers are important in their own right as financial intermediaries, but it is not clear that idiosyncratic failures pose a substantial risk to the integrity of the banking system, payments and settlements. That depends on life insurers’ links with banks, through ownership or counterparty exposures. Whichever choice is made, the Financial Stability Report would be a good place to explain it, using the concrete example of this category of financial institution.

The moderate tone used when commenting on unfavourable developments in the insurance sector, which culminated in a government intervention in the autumn of 2001, provides an illustration of the sometimes limited discussion of weaknesses in the financial system. Between June 2000 and September 2001, the buffer capital of life insurance companies fell from NOK 38.1 billion to close to zero. Although these developments may not have put the stability of the financial system risk, it is interesting to consider how the Financial Stability Report addressed these developments. The November 2001 issue, which must have been prepared during or right after the financial market upheaval, mentioned these developments only very briefly and in very moderate terms. It was not before the May 2002 issue that Norges Bank acknowledged that the insurance sector went through a ‘turbulent’ period. In the end, the problems in the Norwegian insurance sector did not put at risk the stability of the financial system. However, Finansiell stabilitet 1/200343it is not clear from the Report whether Norges Bank downplayed the problems of the insurance sector because (a) it never analysed them; (b) it analysed these problems but then considered them not likely to be of systemic relevance; or (c) it judged that the problems could have threatened the stability of the financial system but were resolved successfully through the timely intervention of the government. In any case, the 2002 November issue discussed in greater length the financial strength in life insurance companies.

(vii) Have there been changes in financial system infrastructure of major significance to financial stability?

Several such changes – such as Continuously Linked Settlement – have been reviewed in past Reports, but there is a case for a more systematic stock-taking of developments in key financial system institutions (e.g. regulators, central counterparties) and in the legal and professional framework governing financial intermediation and affecting its riskiness (e.g. capital market practices; corporate governance arrangements post-Enron; key legal decisions; implications of Basel II). Given a regular stock-take, the Reports themselves need only cover the material developments for financial stability; ongoing public policy debates that have not yet affected the risks facing financial sector agents need not be reported (unless Norges Bank has some other reason for intervening publicly in the debates).

(viii) Mitigating the FS risks identified.

Past Reports have correctly attempted to weigh the different risks discussed and to weave them into a coherent story. This is the main challenge for central banks that publish financial stability reports. Calibrating risks is not easy, not least because of the absence of a generally agreed framework of economic analysis and definition of the public policy objective(s). And if risks are thought to be increasing, there is always a danger of exacerbating them by publicising them, as already noted (page 38). In such cases, it is beholden on the central bank to explain how the risks can be mitigated. One way might be through policy actions. Norges Bank might therefore also consider further how it might draft and use the conclusions of its Reports in order to identify possible areas for policy review, and ultimately to promote risk-reducing changes in regulatory policy, the legal framework, and financial market and institutional practice.

This touches on the communications strategy lying behind the regular publication of Reports and meetings with financial sector institutions. This strategy should be comprehensive, encompassing Norges Bank’s other publications (e.g. Inflation Report, Payment System Report), and taking into account other bodies’ publications (e.g. those of Kredittilsynet). The Financial Stability Reports need to be clearly focused on Norges Bank’s systemic financial stability objective, drawing on the other publications as necessary, but without duplication⁶. If the suggestions above have any merit, the editors will have to be ruthless in excluding discussion of interesting but peripheral issues. They may also decide that a publication of the current length cannot serve both as a flagship of financial stability analysis and as a vehicle for explaining Norges Bank’s financial stability role to the general public. If that is the case, they may wish to consider the scope for an additional very concise and more ‘populist’ publication for a wider readership.

Some points of detail:

Subject to space constraints, the authors may wish to explore occasionally some of the following narrower issues.

• The macro environment
  - External balance sheets by broad sector
  - Scope for hedge fund destabilisation?
• Household sector exposures
  - Unsecured borrowing, especially credit cards
  - Mortgage equity withdrawal
• Corporate sector exposures
  - Bond spreads
  - Corporate liquidity
  - Defined-benefit pension schemes
  - What have firms been doing to address balance-sheet problems?
• The financial system
  - Foreign ownership
  - Quality of bank capital
  - Forecasts of bank capital and / or losses
  - Basel II
  - Banks’ Payment and Central Clearing House

IV. Proposals for further work by the external review panel

In the light of the discussions held at Norges Bank during March, we propose that the Panel meets staff again in one to two years’ time to discuss recent Reports, in the light of (i) this initial review and (ii) questions for further consideration identified in the dialogue following its receipt.

References

Bank of England: Financial Stability Review, various issues.

Bordo, M, Eichengreen, B, Klingebiel, D, and Martinez-Peria, MS (2001): ‘Is the crisis problem growing more severe?’Economic Policy, No 32, pp 51-82.

Hoggarth, G, and Saporta, V (2001): ‘Costs of banking system instability: some empirical evidence’, Bank of England Financial Stability Review No10, June, pp 148-169.

Svensson, L (2001): ‘Independent review of the operation of monetary policy in New Zealand: report to the Minister of Finance’, executive summary, February.

National Bank of Hungary, Financial Stability Report, various issues.

Sveriges Riksbank, Financial Stability Report, various issues.