Marrying the micro- and macroprudential dimensions of financial stability: six years on
Address by Mr Malcolm D Knight, General Manager of the BIS, at the 14th International Conference of Banking Supervisors, Mérida, 4-5 October 2006.
Building on a theme from an address delivered at the same event in 2000 by the previous General Manager of the BIS, Mr Knight assesses the degree to which, in the pursuit of financial stability, a macroprudential perspective has been gaining ground in regulatory and supervisory frameworks as a complement to the more traditional microprudential perspective. He argues that a welcome strengthening of the macroprudential orientation has taken place but that hurdles of an analytical, institutional and political economy nature stand in the way of a further shift. Mr Knight then considers ways in which these hurdles could be addressed.
It gives me great pleasure to have, once again, the privilege of addressing this event. This biennial conference has become a landmark in the calendar of banking supervisors from around the globe.
Two years ago in Madrid, I reflected on the powerful convergence forces that have been shaping the financial industry. I examined the shared risk management and prudential challenges that these posed in banking and insurance. And I did so primarily from what you might call a "microprudential" perspective. In other words, I considered these challenges mainly from the perspective of individual financial institutions.
Today, I would like to broaden the focus and explore more generally the challenges that supervisors face in ensuring financial stability from a "macroprudential" perspective. The focus, that is, will be more firmly on the financial system as a whole. In doing this, I would like to revisit the main themes developed by my predecessor, Andrew Crockett, in his keynote address at this event six years ago, in 2000. In seeking to identify potential future directions for prudential regulation and supervision in the new century, he argued for the need to marry the macroprudential and microprudential perspectives of the pursuit of financial stability. That marriage, in his view, called for a strengthening of the macroprudential orientation of regulation and supervision - a view which I very much share. Six years on, it seems a good time to assess where we stand in this process.
I will argue that a considerable strengthening of the macroprudential orientation in financial sector regulation and supervision has taken place, to a degree probably unimagined at the time. This strengthening has occurred against a financial backdrop that, if anything, has increased the importance of the orientation. At the same time, hurdles stand in the way of a further shift. They are of an analytical, institutional and political economy nature. Looking ahead, I anticipate a further, if possibly gradual, shift. This would be desirable to help ensure financial stability. But the task of ensuring financial stability is not one that supervisors can fulfil alone. It requires strong support from other authorities, not least those responsible for monetary policy, fiscal policy and accounting standards.
I will first recall briefly the key elements of the macroprudential orientation of financial regulation and supervision and the reasons for its importance. I will then assess the progress made before examining future prospects in the light of the hurdles that remain. I will conclude by highlighting the main messages.
I. The macroprudential orientation: key elements and importance
Although the terms "microprudential" and "macroprudential" are commonly used, there is no universal agreement on their meaning. It is universally agreed, though, that they denote elements that inevitably coexist in regulatory and supervisory frameworks.
At the BIS, we have highlighted two defining elements of a macroprudential orientation.
The first defining element is a focus on the financial system as a wholeas opposed to individual financial institutions. A macroprudential orientation highlights the cost to the macroeconomy of stress in the financial system generally; a microprudential one puts more emphasis on the losses incurred by individual financial institutions per se.
The second defining element is an emphasis on the dependency of aggregate risk on the collective behaviour of individual institutions - what some economists call the "endogeneity of risk". A macroprudential orientation highlights the fact that asset prices and the macroeconomy are themselves strongly affected by how financial institutions behave; a microprudential orientation tends to take movements in asset prices and the macroeconomic backdrop as given - as "exogenous" in economists' parlance. As a result, a macroprudential perspective makes supervisors particularly alert to the possibility that courses of action that may make eminent sense for individual financial institutions may not result in desirable aggregate outcomes. For instance, extension of credit driven by concerns with market share in good times can easily lead to over-extension in the aggregate. Likewise, retrenchment in bad times can lead to over-retrenchment, as sales of assets and the tightening of lending terms exacerbate the falls in asset prices and economic weakness. Hence the resulting unexpected rise in volatility and the high correlation across asset prices and asset classes, which typically characterise financial distress.
These two defining elements have implications for the diagnosis of threats to financial stability and for the corresponding remedies.
As regards diagnosis, the orientation highlights various elements. These include: the distribution of risk across the financial system as a whole and the nature of shared exposures to macroeconomic factors; the mutual interaction between the real economy and the financial system; and the need to be particularly watchful in good times, because it is then that, disguised by overly buoyant economic conditions, over-extension in balance sheets ("financial imbalances") builds up, often sowing the seeds of subsequent financial distress.
As regards remedies, the macroprudential orientation emphasises the importance of encouraging financial institutions to build up buffers in good times, as risk and financial imbalances grow, so as to run them down in bad times, as imbalances unwind and risk materialises. For buffers to act as such, they should be allowed to be run down. But, to avoid a situation in which financial supervisors exercise forbearance, they need to be raised to proper levels beforehand. Their build-up strengthens both individual institutions and the financial system as a whole, both in individual countries and globally. And it may help to restrain the build-up of financial imbalances themselves.
There are at least two overarching reasons why a macroprudential orientation is an important complement to a microprudential one.
First, experience suggests that the dynamics of distress throughout history argue for the relevance of the mechanisms I have just outlined. The financial crises that have exerted the most significant costs on the real economy have not typically arisen from the contagious spreading of problems incurred by individual institutions owing to idiosyncratic factors. Rather, they have resulted from shared exposures to macroeconomic risks. And these financial crises have been exacerbated by the behaviour of financial institutions themselves, both in the build-up of the financial imbalances and in the blow-out of distress. This was true of financial crises as far back as in the gold standard period prior to the First World War; it was true of the Great Crash of 1929 and the subsequent Great Depression; and it has been true of the episodes of widespread distress over the last quarter of a century or so, including the banking crises in the Southern Cone countries of Latin America in the late 1970s and early 1980s, those in the Nordic countries in the late 1980s and early 1990s, Japan in the 1990s and the Asian financial crisis countries in 1997 and 1998.
Second, changes in the structure of financial systems have increased the relevance of the macroprudential orientation. New financial instruments have made it much easier to transfer risk across the financial system at the same time as the risk management practices of large private sector financial institutions have tended to converge. Think, for instance, of the exponential growth in credit risk derivatives. Strengthened regulation of traditional financial institutions has been a factor contributing to a migration of risk onto the balance sheets of institutional investors, other asset management vehicles and the household sector. Financial distress with potentially serious consequences is now more likely to arise outside the banking sector; the LTCM problem in 1998 was an obvious example. And the sheer magnitude of cross-border flows and the internationalisation of local financial markets complicate matters further. The "macro" dimension has become more global.
II. The macroprudential orientation: progress made
It is useful to consider the progress made in strengthening the macroprudential orientation under three headings: awareness of this dimension of the pursuit of financial stability, the identification of vulnerabilities, and the calibration of policy tools.
There is little doubt that the awarenessof the importance of the macroprudential orientation for financial supervision has grown substantially. The very term "macroprudential", hardly known outside some narrow central banking circles in 2000, has become a familiar word. The Asian financial crisis of the late 1990s did not just highlight the crucial role played by a strong financial infrastructure in addressing foreseeable and unforeseeable threats to financial stability - and, as you know, the Core Principles for Effective Banking Supervision are a key element of such an infrastructure. It also drove home the significance of the nexus between the macroeconomy and the financial sector. Subsequently, the debate surrounding the procyclicality of Basel II underscored the same message. The value of the macroprudential orientation has been strongly supported by international financial institutions, such as the IMF and World Bank, in their surveillance and technical assistance work. And it has found some resonance in the risk management practices of financial firms. Firms have begun to think of ways to better incorporate macro factors in their risk assessments. They are also keenly aware of the need to manage the endogenous risks linked to the evaporation of liquidity at times of stress, arising from excessive search for yield, crowded trades and retrenchment.
This increased awareness has been translated into the greater emphasis on macroprudential analysis in the identification of vulnerabilities. There has been a more holistic view of financial systems. Examples include: the growing attention paid to currency mismatches; the work done on the transfer of risk across the financial system, not least through credit derivatives; and the questions asked about the apparently extensive risk transfer to the household sector, on the back of the shift from defined benefit to defined contribution pension systems, rapid debt growth and the surge in house prices around the world. Macroprudential analyses have become increasingly common. They have combined judgmental and quantitative aspects, based on the development of new analytical tools such as early warning indicators and macro stress tests. The policy venues carrying out such analyses have multiplied. This has been so both at the national level, encouraged by FSAP exercises and manifested in the growing number of Financial Stability Reports, and at the international level, such as at the Financial Stability Forum and the Committee on the Global Financial System at the BIS. And macroprudential elements have also been more formally incorporated into the analyses of private sector monitors, such as the credit rating agencies.
The stronger macroprudential orientation has also been reflected in the calibration of policy tools, in the form of both built-in financial system stabilisers and discretionary adjustments by supervisors.
As regards built-in financial stabilisers, I would note two developments. The first is the introduction by Spain's bank supervisory authority, the Bank of Spain, of a system of prudential "statistical" provisioning, which anchors more tightly the level of current loan loss provisions to the historical experience over varied business conditions. The system is intended to mitigate the danger that low realisations of risk in good times lead to an excessive fall in loan provisions. The second relates to the modifications made to the original Basel II proposals in order to reduce the potential procyclicality of the arrangements. These have included a flattening of the risk curve, greater emphasis on longer horizons and downturn loss-given-default estimates, and the possibility of using the supervisory review pillar of the framework to tighten requirements in the light of the results of stress tests of macroeconomic conditions. As a package, Basel II may be less procyclical than Basel I, not least by encouraging an earlier recognition of the potential future deterioration of asset quality. The improvement in risk management culture it has supported is likely to remain its most enduring achievement. Recall that banks used to throw away key historical data on their loan loss experience!
As regards discretionary adjustments, a range of policy actions has been taken in a more informed and deliberate way. Several authorities have tightened prudential standards in the light of the perceived build-up of financial imbalances. The instruments have included adjustment to loan-to-value ratios, provisioning rules and capital requirements. This has been the case in a number of countries in Asia and in Europe, including central and eastern Europe, which have been faced with rapid credit expansion and asset price inflation, not least in the residential property sector. In addition, in the United Kingdom, in 2002 the FSA relaxed stress test requirements on life insurers to prevent distress selling of equities, which risked exacerbating the sharp fall in equity markets at the time.
III. The macroprudential orientation: prospects
Welcome as this progress has been, a number of hurdles stand in the way of a further strengthening of the macroprudential orientation. They have to do with limitations in our knowledge, with the realities of institutional setups and with the political economy of implementation.
First, our knowledgeof the workings of the economy, especially of interactions between the financial sector and the real sector, is inevitably limited. While the contours of the genesis and dynamics of financial instability are broadly understood, our ability to identify vulnerabilities and effectively calibrate the policy instruments remains unsatisfactory.
In identifying vulnerabilities, analytical work is still in its infancy. The quantitative tools developed so far, including at the BIS, are a useful first step. But their performance is still too uneven and their ability to identify problems well in advance with sufficient reliability remains an open issue. As a result, they fall short of providing a reference framework that is tight enough to discipline the inevitably more qualitative assessment of vulnerabilities. The result is that the efforts can lack concreteness and bite. For instance, is the fact that the apparent financial imbalances we currently observe do not unwind evidence that they are not imbalances after all? Or is it simply a sign that the risks are, in fact, growing larger? Moreover, the same limitations in our knowledge make it harder to calibrate policy instruments.
Second, the complications arising from institutional setupsare multifaceted. Sometimes the objectives of prudential authorities are not conducive to a strengthening of the macroprudential orientation. In particular, this is so for those prudential authorities with strong depositor/investor protection elements in their mandates, more consistent with a focus on individual institutions as opposed to systemic risk per se. Such consumer protection elements are stronger among insurance supervisors but are also present in banking. Sometimes the control over the relevant instruments is dispersed across different types of authority with, naturally, quite different objectives. For example, accounting standard setters and tax authorities have had reservations about dynamic provisioning rules with strong forward-looking elements. Sometimes, the professional background may be less supportive of the use of the instruments. It is not uncommon for supervisory agencies to have a comparative advantage in legal and possibly accounting issues. This coexists uneasily with a macroprudential orientation, which puts a premium on understanding the links between the macroeconomy and the financial system as well as market dynamics.
The globalisation of the financial system has exacerbated the limitations in institutional setups, as increasingly seamless financial markets meet, not to say clash, with regulatory and supervisory arrangements anchored to national jurisdictions. For example, when cyclical conditions and financial pressures vary across countries, they can raise dilemmas for host authorities. Tightening prudential standards on domestic banks may unlevel the playing field wherever foreign branches, outside their control, are present. And home authorities may have little incentive to tighten their own standards for those banks that operate internationally, especially when their business out in the host country accounts for only a small fraction of their overall activities. Moreover, increasingly mobile financial flows may undermine the effectiveness of joint action.
Finally, political economy considerationscan seriously complicate implementation. In part, this reflects the presence and importance of narrowly interpreted depositor protection or investor protection elements in the mandates that I have already mentioned. More generally, though, measures designed to tighten standards in good times can easily run against the ingrained resistance of financial markets, the body politic and the public at large. We know all too well from other policy fields, such as monetary and fiscal ones, how famously unpopular it is to spoil a party when it gets going!
The diagnosis of the three hurdles that the macroprudential orientation must confront naturally suggests three broad directions for policy responses.
The first is to strengthen analytical efforts to improve our knowledge. Admittedly, progress may well be slow. Not least, instances of serious financial distress are likely to be rare, especially if policymakers do their job well! This complicates the validation of empirical evidence. At the same time, I remain cautiously optimistic. The state of our understanding of financial stability reminds me of that of monetary policy in the late 1960s and early 1970s. As the subsequent successful battle against inflation testifies, central banks have made much progress since then, based on sound analysis of the monetary system. Why should progress be beyond reach in the fight against financial instability?
The second is to address the hurdles arising from institutional setups. This calls for strengthening cooperation among authorities with different statutory objectives so as to pool knowledge and expertise and to better activate the available instruments in support of the macroprudential goal. At a minimum, this involves cooperation at three levels. One is between different prudential authorities within national jurisdictions, so as to improve the monitoring of system-wide risks and develop more consistent regulations across sectors. The second is between prudential authorities in the financial sector, on the one hand, and monetary, fiscal and accounting authorities, on the other. The comparative expertise of monetary authorities in understanding the nexus between the financial sector and the real sector of the economy can provide invaluable input. An intensified dialogue with accounting and tax authorities can help to render prudential instruments more effective, by removing obstacles to their deployment. Enabling more forward-looking provisioning is a case in point. Looking ahead, the trend towards more extensive marked to market accounting makes such cooperation all the more important. The dialogue with tax authorities could also help to remove tax distortions that encourage risk-taking, such as by promoting excessive indebtedness. A third level is cooperation across borders, underpinned by greater convergence in prudential frameworks and supervisory practices.
The third direction involves addressing the political economy obstacles to implementation. This is no easy task. But the analytical and institutional efforts that I have just mentioned should help. The former can provide the necessary intellectual basis for action. After all, this is precisely what happened in the case of monetary policy: it was the recognition of the absence of a long-run trade-off between inflation and unemployment that laid the basis for central banks' successful fight against inflation. The latter can generate the necessary consensus among policymakers. Complementary educational efforts targeted to financial markets and the public at large are also essential.
To conclude, much has been done in recent years to strengthen the macroprudential orientation of prudential frameworks. Progress has gone furthest with respect to the awareness of this dimension of the pursuit of financial stability, it has been substantial with respect to the assessment of vulnerabilities and somewhat less advanced with respect to the calibration of policy tools. This shift has been welcome. Looking forward, I expect it to continue, although significant hurdles remain. Further progress calls for additional analytical, institutional and educational steps. It also puts a premium on cooperation among prudential authorities and other policymakers, both nationally and internationally. As the macroprudential orientation underscores, ensuring financial stability is not a task that can be fulfilled by prudential authorities alone. It requires purposeful action by a broad set of policymakers within an agreed framework, based on a shared diagnosis, a clear allocation of responsibilities and a coordinated use of available instruments.