BIS Annual Economic Report  | 
24 August 2000
PDF full text
 |  12 pages

There seems to be a widespread perception that the global economy now stands on the brink, but the brink of what remains the question. The better than anticipated recent economic performance in many parts of the world has predictably led both private and public sector bodies to revise their growth forecasts upwards. Many now see better economic prospects than at any time since the early 1980s. Part of this is pure extrapolation, but advances in technology and continuing deregulation are further reasons for expecting rapid growth along with continuing low inflation. Indeed, as such structural developments increasingly spread throughout the global economy, the remarkable success enjoyed by the United States over the last few years seems likely to be more broadly shared. A prolonged boom in a more market-driven world economy can thus by no means be ruled out.

Yet, even if this longer-term vision is accepted, policymakers can still expect a few bumps along the way. Consistent with previous historical episodes of structural change and associated new promise, the last decade has been characterised by rapid credit expansion in many economies, and a growing appetite for risk among lenders. Concentration ratios have risen in financial markets while liquidity has sometimes fallen noticeably. These developments imply not only that the global economy may have become more exposed to macroeconomic shocks, but also that the dynamic response of markets to such shocks may be harder to predict than in the more regulated past. Finally, it must also be asked whether, with a more globalised financial system, policymakers have all the tools required both to avert problems and to manage them should they arise. This is never an easy task since liquidity injections, which may be needed to help manage one crisis, can also encourage imprudent behaviour, simply leading to the next.

The principal macroeconomic vulnerabilities are well known, not least among them the possibility of rising inflation in the countries most advanced in the business cycle. But it is the potential interactions between these vulnerabilities that may require more attention. Stock prices in many countries still seem high by historical standards, even after stripping out "new era" stocks, for which new valuation criteria could conceivably apply. The US dollar also appears to be stronger than is compatible with the stabilisation of longer-term external debt ratios. Given the increased extent to which projected returns on equity have driven international capital flows in recent years, the possibility of a simultaneous adjustment in both markets would seem greater than historical correlations might indicate. The likely implication of such an outcome would be slower demand due to wealth effects, even as inflation rose in response to both internal and external pressures. Whether the former would be judged useful or not, since it would help offset the inflationary pressures, would of course very much depend on how big and orderly the wealth adjustment proved to be. Neither a hard nor a soft landing could logically be ruled out. Finally, in such an uncertain macroeconomic environment, an outflow of capital from emerging market economies might also be anticipated. While those countries that now have current account surpluses and large external reserves might be little affected, not all emerging markets currently enjoy such a comfortable external position.

The market dynamics conditioning the response of the global financial system to a continued tightening of policy rates also deserve attention. Higher policy rates have for the most part been viewed as helpful in sustaining economic growth while heading off inflation. Growth has been positive for stock prices and for credit spreads, and low inflation has perhaps constrained the upward movement in bond rates. Yet, if the authorities were suddenly judged to be "behind the curve", this could all go into reverse, with potentially contractionary effects. In addition, equity price movements could be exaggerated by the growing use of leverage and margined debt, portfolio insurance strategies and the increased dependence of blue-chip profits on stock gains in the high-tech area. These are all interwoven elements with potential for mischief. Similarly, fixed income markets might also react uncharacteristically, given the changing status of benchmarks in both the US and European bond markets, and the growing reluctance of large firms to commit capital to a market-making function.

How any or all of these developments might feed back onto the health of individual financial institutions is another open question. A combination of continuing deregulation, heightened competition, technological change and increased concern for shareholder value may have encouraged behaviour and cross-sectoral relationships which will prove to have been imprudent only when the next downturn comes. That said, it must also be noted that financial markets have recently been extremely volatile, without any such knock-on effects, and very high volumes of financial transactions have been processed without any signs of strain.

What can policymakers do to ensure that the global economy reaps the benefits of a more efficient production and financial structure over time? Whatever the answer, and some suggestions are made below, it must be recognised that efficiency is not everything. Fairness and perceptions of fairness must also be taken into consideration. Thus, issues of income distribution, debt relief and the protectionist policies of industrial countries, particularly towards imports of agricultural products and textiles from emerging markets, need more effective attention than they have received thus far. A further constraint on the pursuit of efficiency, particularly in the development of financial systems, must be considerations of safety and stability. The potential economic costs of sporadic financial crises must always be weighed against the ongoing benefits of freer capital markets. Unless policy measures can adequately trade off all these competing objectives, a sharp and unwelcome turn away from a market-based approach to less desirable solutions cannot be ruled out.

Imbalances and the pursuit of price stability

Uncertainty about the economic future is an inevitable complication in the formulation of monetary policy. Knowledge is limited about both the structure of the economy and the lags with which monetary policy affects ultimate objectives. Data are frequently subject to revision, as are the sentiments that often dominate expenditure decisions. While theory suggests that policymakers should at each moment simply give it their "best shot" and then systematically revise their views on the basis of incoming evidence, this strategy also has shortcomings. In particular, widespread uncertainty may lead to excessive delays in responding to new information, thus increasing the chance of large, disruptive moves later. Indeed, such delays might even call into question the authorities' commitment to meeting their stated objectives. Moreover, a policy strategy based solely on mean expectations could result in inadequate preparation in the event of extreme outturns. In the current circumstances, where many imbalances can be identified, such a shortcoming would seem particularly significant.

One point on which virtually everyone would agree is that the current rate of expansion of domestic demand in the United States is unsustainable and potentially inflationary, and that a similar if less extreme state of affairs prevails in some of the other English-speaking countries. With all talk of fiscal action in the United States moving resolutely in the other direction, the recent trend towards monetary tightening is most welcome even if some asset prices currently look quite vulnerable. Were monetary policy to back off at the first signs of declining equity prices, the risks of moral hazard would be great. In any event, if we really have entered a "new era", the likelihood of a sharp and sustained reaction in equity markets would be much reduced. And if we have not, then it could be argued that the sooner the bubble deflates, the better.

This is not to say that a significant reaction in the stock market, or in financial markets more generally, should not elicit a measured policy response. Disinflation can go both too far and too fast. This danger is not inconsequential in the United States, nor in a number of other countries advanced in the cycle. Given recent low rates of saving and heavy investment in housing and durable goods, it would now be very easy to postpone prospective expenditures. But once it has become apparent that certain investments will never yield their expected rates of return, the misguided investors should be allowed to pay the price, and quickly, so that capacity can be reduced and longer-term profitability rapidly restored. This may be the principal lesson from the 1990s in Japan.

Another area of more or less universal agreement is that the rate of growth of private expenditures in Japan is too low. The unresolved question is: how might it be encouraged to quicken? The Japanese fiscal authorities are to be commended for having been prepared to use fiscal policy counter-cyclically, yet this could also have been done more effectively. The initial grudging recourse to this expedient, accompanied by threatened and actual reversals of policy from time to time, undermined private sector confidence that a definitive upturn was in prospect. Moreover, as the scale of the spending rose, the efficiency of the expenditures declined sharply, leaving taxpayers with perceptions of increased liabilities that were not matched by worthwhile assets. While rising levels of public debt in Japan imply that overall fiscal stimulus will have to be more restrained in the future, a rebalancing of expenditures could yet pay big dividends. Cutting back on the overdeveloped public investment side (three times the G10 average) and increasing expenditure on the underdeveloped social safety net could help preserve needed confidence. It could also increase labour mobility at a time when this would be highly desirable. With old industries still plagued by overinvestment, expansion in other areas must be encouraged through deregulation and related public policies.

Whether there is any further role for expansionary monetary policy, now that short-term rates have effectively hit the zero nominal bound, remains a controversial topic in Japan and elsewhere. There are two strands to this debate. The first suggests that the Bank of Japan should have recourse to "exceptional" procedures to directly raise asset prices and sharply increase the level of reserves in the banking system, for example through unsterilised yen intervention and massive purchases of government bonds. The Bank has strongly resisted such initiatives on the grounds that they would be of limited, if any, use and might even get in the way of a needed increase in rates should the economy recover. Underlying this resistance are also legitimate concerns about the longer-term independence of the central bank. The second strand of the debate has to do with the merits of introducing a new monetary regime of inflation or price level targeting. The basic logic is that, while nominal interest rates may be bounded at zero, negative real rates may nevertheless be attainable if some means can be found to foster expectations of a rebound in prices. Proponents of a new monetary regime say that inflation expectations would indeed be shifted upwards in consequence, with price level targeting becoming the more attractive the longer prices had been dropping. Opponents, including the Bank of Japan, hold that this spontaneous shift in expectations simply would not happen in the absence of credible means to change the underlying reality of ongoing economic stagnation.

A final point on which virtually everyone would agree is that, by default, the primary challenge for continental Europe is structural reform. Major macroeconomic imbalances of the sort that increasingly confront the United States and Japan are neither evident nor in prospect. Current account balances for the region as a whole do not pose a problem. Equities, which have seen sharp price increases in many countries, still constitute only a small (if rising) proportion of household wealth. And while the continuing weakness of the euro clearly has the potential to raise inflationary pressures, a forward-looking and vigilant monetary policy would seem sufficient to allay most concerns. Fiscal restraint might also be recommended, particularly in those smaller European economies which are already pushing the limits of potential, but in most countries policies of restraint have already been in place for some time. Of course, all of these macroeconomic prescriptions for holding a steady course presume that the current European expansion will continue uninterrupted by any dramatic events elsewhere.

The enthusiasm for structural reform in Europe has clearly been stimulated by the introduction of the euro, with elements of both carrot and stick coming into play. A single currency could catalyse enormous efficiency gains in product markets, if only other reforms would let it. Conversely, failure to effect labour market reforms, which would allow more rapid response to market signals, could lead to rising unemployment given asymmetric shocks in a single currency area. Much has already been done, as indicated by the recent unusual drop in the EU unemployment rate at a very early stage of the recovery, but the lessons from the United States and "best practice" in Europe indicate that further progress is possible. The heads of government pledged to pursue structural reforms actively at their recent meeting in Lisbon but, as always, the difficult task will be to push this commitment through at home.

Exchange rate movements among the major currencies have generally provided support for cyclical stabilisation over the last few years. Recently, however, the strength of the yen and the persistent weakness of the euro have been less consistent with domestic objectives, raising the question of what might be done about it, aside from moving policy rates. The current answer appears to be, not much. However, circumstances could conceivably change. Unilateral intervention by the Japanese has the disadvantage that its signalling function has limited credibility, since rates already at zero can be lowered no further (although the possibility of "unconventional" measures remains open). Bilateral intervention with European involvement has also been eschewed to date, even though, prima facie, it would seem potentially useful in the right market conditions. One reason might be concerns in Europe that intervention would be misread as signalling a dilution of the commitment of the newly created central bank to domestic price stability. And as for multilateral intervention, involving the United States as well, concerns have been expressed that this might be interpreted as the harbinger of a managed global exchange rate system, for which there currently seems to be little official enthusiasm.

Looking further ahead, the biggest policy challenge could be coping with a sudden reversal in the fortunes of the dollar. The additional disinflationary impact on Japan would clearly be unwelcome, although much less so in Europe, where inflationary concerns have mounted. Even here, however, problems could arise if a rebound of the euro went too far and too fast. Given the extent to which the recent decline in the euro was unexpected, and that momentum-related factors could reverse, this possibility should not be dismissed. Of course, were the euro to rebound sharply due to renewed optimism about structural reforms in Europe, accompanied by both increased consumer and investment spending, the dangers posed for the current expansion would be significantly mitigated. This would seem to be another very good reason for proceeding with structural reforms.

Exchange rate issues loom even larger from the perspective of emerging market economies, as they are relatively more open and even more prone to sentiment-driven swings in capital flows. For the moment, prospects of improved economic performance in Asia, Latin America and eastern Europe have encouraged inflows, of direct investment in particular. Nevertheless, concerns remain that external shocks or internal policy failures could suddenly put those flows into reverse. Current account deficits and structural fiscal problems are potential worries in Latin America, while a heavy dependence on the technology sector and slow progress in corporate and bank restructuring may pose risks in Asia. Although most emerging market countries nominally have floating exchange rates, it is also true that many, especially in Asia, have been intervening heavily to stop a loss of competitiveness through currency appreciation. While touted as being consistent with the need to build reserves, and better than capital controls, which have evoked surprisingly little interest to date, such policies could eventually lead to excessive credit creation. This is just another way to lose competitiveness. Perhaps more pernicious is the danger that borrowers will revert to a fixed rate mentality and be encouraged once again to borrow in foreign currency at lower cost. While this might seem unlikely in the light of recent crises, it is astonishing how quickly people forget even hard-learned lessons.

Some emerging market countries have been clearer in their advocacy of alternative policy regimes. In just the last year, for example, South Africa, Brazil, Poland and the Czech Republic have all announced the adoption of inflation targeting regimes. In making this choice, they have joined a club of industrial countries which set policy to reach preannounced targets for inflation within the framework of a floating exchange rate. Whatever the merits of such a framework in industrial countries, the balance of advantages and disadvantages may be somewhat different when adopted elsewhere.

Among the advantages for emerging markets would be its potential transparency. Thus, under an inflation targeting regime, policymakers might be less subject to accusations that they had personal interests at stake when setting both interest rates and exchange rates. A further advantage would be the possible anchor provided for inflation expectations in countries, like many in Latin America, with a particularly poor track record in controlling inflation. However, in targeting inflation, emerging market economies commonly face some offsetting disadvantages as well. These include a relatively poor capacity to undertake sophisticated economic (including inflation) forecasting, unreliable data and the problem of ongoing structural change. In addition, many emerging markets, most recently in Asia, have proved vulnerable to asset price bubbles, which are not easily handled in an inflation targeting framework. Finally, the fact that food and imports make up such a large part of the consumption basket, and that price deregulation is often pervasive and ongoing, may drive a significant wedge between headline inflation and the index that the central bank actually feels it has a chance of controlling.

The upshot is that emerging market countries face a sharply exaggerated version of the trade-off confronting industrial countries that target inflation. If they set their targets too ambitiously, they will lose credibility if targets are missed. Conversely, if they set more realistic goals, these may be judged to be unambitious and the regime will enjoy no credibility in the first place. As with all regime choices, no one size fits all. Each country must look at its own circumstances and its own history and make a judgment about the best course to take. When it comes to the need for sound fiscal policies, however, no exercise of judgment is required. Without such policies, history teaches us that any regime aimed at controlling inflation is doomed to failure.

Structural change and the prevention of global financial instability

Financial failures in the 1930s seriously aggravated the economic downturn in many industrial countries, and led to a sharp tightening of the regulations governing financial activity. The postwar period witnessed a progressive liberalisation as the memories of earlier difficulties faded and the potential benefits of freer financial markets became better recognised. However, recurring financial crises during the last three decades, in both industrial countries and emerging market economies, have focused renewed attention on three issues. How might future crises be avoided? How might they be better managed when they occur? And how might crises ultimately be resolved, including through debt reduction? With respect to each, the actual progress made has been substantial, but is dwarfed by what remains to be done. On some questions there is still no international consensus as to what constitutes sensible policies. And in virtually all cases, the practical challenges involved in actually implementing agreed proposals remain daunting.

Measures directed to preventing financial crises increasingly recognise how microeconomic deficiencies can interact with macroeconomic phenomena with insidious effect. There are three possible market manifestations of such problems: extreme short-term price volatility in some markets, medium-term price misalignments, including those leading to asset price bubbles and excessive capital flows, and contagion across markets and countries unwarranted by the underlying fundamentals. Preventive attention must then focus on each of the three pillars supporting both the domestic and international financial systems: the good health of financial institutions, the proper functioning of markets, and the establishment of a sound infrastructure including legal and judicial processes, payment and settlement systems, and accounting standards. With respect to each pillar, a further trio, this time of incentive systems, can be identified to help promote prudent behaviour. The starting point must be internal governance, based fundamentally on self-interest and the preservation of private capital. To this can be added adequate supervision and oversight. And finally, while recognising its periodic limitations, market discipline must increasingly be assigned a role in a more market-driven world. Given three problems, three pillars, and three prescriptions for each, it is perhaps not surprising that so much still remains to be done.

Promoting healthy financial institutions, especially banks, is a crucial prerequisite for financial stability. The largest number of crises still arise, be it in emerging market economies or industrial countries, from financial institutions overextending themselves when times seem good and then retrenching violently afterwards. Governance would first benefit from a greater internal focus on risk-adjusted rates of return, particularly when rewarding traders and credit officers. The relentless pursuit of shareholder value, without this crucial adjustment, could prove a very dangerous strategy. To this recommendation might also be added greater attention to the way in which public safety nets may encourage imprudent behaviour, not least through institutions thought to have a state guarantee. The recent explosive growth of such institutions in the United States reinforces long-standing concerns about similar state involvement in both Japanese and continental European banking.

The proposals put forward by the Basel Committee on Banking Supervision for improving the 1988 Capital Accord, to be modified in the light of the recently concluded consultation process, will link the minimum capital requirements for banks with well developed rating systems more closely to the banks' own internal assessments of credit risk. Such a linkage is clearly desirable, provided of course that these assessments are not themselves biased in any way. Eligibility for this approach will thus be subject to sound practice standards and guidelines aimed at ensuring the integrity of the rating system output and process. The standardised approach being proposed for less sophisticated banks perhaps promises fewer benefits, but may also entail fewer risks. It obviously needs to be well thought through since the vast majority of banks in emerging markets are likely to be governed by it.

The proposed new Capital Accord also recognises that supervisors have an important role to play in ensuring that the need for capital is being adequately assessed and that capital requirements are consistent and comparable across institutions. In this latter regard, simple rules ensuring forward- looking provisioning might play a bigger role than is currently the case. The Basel Committee has pointed out that market discipline could also contribute to prudent behaviour on the part of financial institutions. Credit spreads and share prices are traditional mechanisms in this regard, but subordinated debt might also be considered. A necessary but not sufficient condition for the application of market discipline is that the market has enough reliable information to allow it to make a sound judgment. Ensuring the provision of such information, along with a clear explanation of the accounting principles on which it is based, should continue to be of the highest priority.

Of course, one problem with market discipline is that it too might be subject to the same swings of optimism and pessimism that could affect internal ratings. Since such phenomena can have systemic repercussions, much more attention should be paid by the public sector to monitoring developments and to developing analytical procedures for evaluating the risk of systemic problems. Indeed, using stress tests as a corollary to such forecasts also has a lot to recommend it. Whether analyses of this sort should be done primarily by supervisors or by other bodies (commonly central banks) charged with overall responsibility for systemic stability, or by both, needs to be clarified to ensure that this important function does not simply fall between the cracks. One argument for involving central banks is that there may be a useful complementarity between their "top down" approach and the "bottom up" approach more commonly followed by the supervisory community. It is a simple but important insight that many recommendations supporting prudent behaviour at the level of a single firm can have undesirable effects if a large number of firms have simultaneously to alter their behaviour in the same way. Fallacies of composition of this kind are well known in the macroeconomic literature.

While healthy financial institutions remain key to financial stability, especially in emerging market countries, the events of autumn 1998 underlined the growing importance of well functioning financial markets as well. A central point is that markets can provide alternative sources of credit when bank sources run dry, and vice versa. This is one important explanation for the recent interest in developing liquid government bond markets in many emerging market countries. With time, such government bond markets could provide benchmarks to support the issue of private sector bonds as well. Governments in many industrial countries are also facing similar issues pertaining to market functioning, as rising government surpluses reduce the amount of bonds available, potentially complicating the problem of risk management under stress.

Even if more complete financial markets are a public good, it must be recognised that such markets do not always function as efficiently as might be hoped. One remedy, recognising the self-interest of individual market participants, is greater transparency. If participants knew more about the financial condition and exposure of counterparties, they would be less likely to disengage when confronted with uncertainty. In the same vein, if lenders had better information about the debt profiles of sovereign borrowers they might be less likely to suddenly develop fears and head for the exits. A number of initiatives are under way to further transparency of both sorts (see the chapter on the Activities of the Bank) and these should be vigorously pursued.

As for contributions by market overseers to better market functioning, there is evidence that markets are becoming less atomistic, and potentially more subject to herding behaviour particularly at times of stress. Growing concentration among market participants, common risk management and regulatory schemes, increasing use of benchmarks and index tracking, and the exploitation of common and instantaneously available information may all be contributing to this. However, what supervisory authorities might do about these underlying structural trends is significantly less obvious. Finally, there is the most fundamental issue of all. Why do markets overshoot, in effect failing to discipline themselves? In an ideal world, those who pushed prices away from "equilibrium" levels would quickly lose money as prices reverted to the mean. However, in the real world, this often does not happen. Since such phenomena have been seen since time immemorial, it may be that market failures of this sort are simply one of the costs to be borne when reaping the overall benefits of a market-based economic system.

The third pillar of a properly functioning financial system is the underlying infrastructure; in particular, payment and settlement systems, legal and judicial frameworks and proper accounting. In the area of payment systems, a great deal has been done since the late 1980s. Last year, the Basel-based Committee on Payment and Settlement Systems published a consultative document entitled "Core principles for systemically important payment systems". The main task for the future is to refine these principles in the light of comments received, and to see them successfully implemented in all major financial markets. Given that such systems in the industrial countries alone are now processing around $6 trillion of transactions a day, it is clearly imperative that they function flawlessly. As for the legal infrastructure, its crucial importance was brought home by the practical implications of inadequate property laws in Russia and some other transition economies. The crisis in East Asia also revealed certain legal shortcomings which have been only partly addressed. And in many countries, the application of existing law needs to be speedier and much less arbitrary than is currently the case. Finally, significant progress has been made towards agreement on a common set of global accounting standards. This vital work needs to be carried forward, completed and implemented.

The management and resolution of financial crises

While the incidence of financial crises may be reduced through preventive measures, they will never completely disappear. How best to manage a crisis in order to reduce its macroeconomic and social costs will depend very much on how it manifests itself in market processes. Nevertheless, common to virtually all financial crises is a sudden loss of market confidence and the need to quickly evaluate liquidity requirements in the system and how they might best be met. In contrast, crisis resolution is a longer-term endeavour, often requiring some degree of debt reduction. The management and resolution of financial crises can have both a domestic and an international component, and will normally involve private sector creditors as well as public authorities. However, if moral hazard is to be avoided, the latter should be drawn in only as a last resort, when there is clear evidence of some market failure or externality.

Nor should it be automatically assumed that such market failures are likely. Private financial markets, which are increasingly interlinked, have the potential to defuse crises as well as to propagate them. For example, in autumn 1998, many borrowers found it difficult to obtain credit via the short-term paper and derivatives markets. However, the interbank markets expanded enormously and played a crucial role in redistributing credit to those that needed it. Private sector participants, particularly large ones whose behaviour can affect the whole system, will tend to internalise those externalities (indeed, as in the LTCM case, they could be encouraged to do so) and this should lead to more stabilising behaviour. Of course, this is not to say that explicit public sector involvement will never be required, but the public sector itself must retain the right to make judgments in this regard. This discretion must apply not only to possible emergency lending to banks, but also, and even more so, to the possibility of supporting markets suddenly made illiquid for some unforeseen reason.

The nature of any public sector involvement must also be conditioned by two aspects of modern financial markets: the speed with which changes can now occur and their international dimension. With respect to the former, it will be particularly important in the future for the various parties involved to have open lines of communication and agreed procedures for using them. At least among themselves, government bodies should also be clear about who is responsible for what. Such clarity is obviously necessary at the domestic level, including within the euro area, but may apply at the international level as well. Individual financial institutions active in a given country may well have international roots, and responsibility for their behaviour would then lie with the "home" supervisor. Moreover, the liquidity required might well be in foreign currency, which would add a whole new dimension to the problem. Even if a country's central bank were prepared to increase liquidity for the system as a whole, and thus reduce interest rates, this would be of little help if the problem were a shortage of foreign currency. Indeed, such a policy could easily prove counterproductive if the domestic currency depreciated in consequence and the servicing requirements on foreign currency denominated debt rose appreciably.

How international liquidity might be provided to emerging markets suffering from such problems has been widely discussed over the last few years. Yet it is not clear that much concrete progress has been made with respect to the two central issues. The first is how the private sector might be induced to provide needed liquidity both by rolling over maturing debt and by providing new money, for example to help finance an ongoing current account deficit. The former can be effected through a temporary suspension of payments, which has the advantage of treating all creditors equally, but is hardly likely to encourage new credits. Another possibility is moral suasion, as was recently applied with regard to Korea, but ultimately such arm-twisting comes close to coercion. A growing problem, moreover, is how to coordinate such financing. As borrowers in emerging markets and lenders in developed countries become ever more heterogeneous, it becomes increasingly difficult to determine who, if anyone, has the authority to make promises and then follow up on them. In this context, the more issuers of bonds that can be persuaded to include collective action clauses, the greater the likelihood of orderly debt reschedulings.

The second issue is the appropriate modalities for the involvement of the public sector. A balance needs to be found between the provision of liquidity by the IMF and the moral hazard that such assistance engenders. One view is that access to the IMF should normally be for far smaller amounts than recently provided, and through a reduced number of facilities. According to this essentially "rules-based" approach, only limited access to public sector money is consistent with the spirit of burden-sharing, and the terms and conditions of that access should be known to everyone in advance. Advocates of this approach would not deny that there might be cases, presumably of systemic import, in which the public sector might consider it appropriate to make far larger loans. However, it would be argued that, in such cases, the approval process ought to be both more formal and more difficult. The principal counterargument to these views is that all crises are different and ways to manage crises need to be found very quickly. This rather argues for more "discretion" on the part of the IMF. In contrast to these opposing standpoints, there seems to be a growing consensus that official financing should be linked more closely to crisis prevention efforts, and that longer-term and repeated borrowing from the IMF should somehow be discouraged.

Ways to resolve crises through explicit debt reduction are no less controversial. Two sets of problems currently occupy centre stage: bank restructuring in a large number of emerging market countries, and debt reduction for an even larger number of very poor and highly indebted sovereigns. In both cases, two principles should apply. First, if the fundamental issue is who is to pay for losses already incurred, it should be those who contributed to such losses. Taxpayers should not bear the losses incurred by others, unless their governments also have creditor status. Of course, governments and the international financial institutions are the largest creditors of the poorest countries, and therefore bear a special responsibility for alleviating their debt burden. And second, crisis resolution measures must be accompanied by efforts to avoid future crises. Thus, banks must not only be restructured but made efficient and profitable. In the case of debt reduction for sovereigns, steps to ensure good governance and investment in the health and education of ordinary people are crucial. Debt reduction may in some cases be a necessary condition for crisis resolution, but it is by no means sufficient for ensuring future prosperity.

Experience now seems rich enough to hazard a few more suggestions about bank restructuring. Perhaps most important, private sector solutions should always be tried first. Even technically insolvent banks may have franchise values that could attract takeover interest, and sometimes even small capital injections can make a big difference. However, takeovers and mergers must make financial sense; size in itself is no guarantee of survival. And foreign banks should be part of the process, as they bring know-how and technology as well as capital. Also, there seems to be growing agreement that clear criteria for public sector intervention may be useful in avoiding forbearance, and that such intervention should be subject to certain principles. It should be timely, comprehensive, non-political, transparent and, above all, definitive.

Whether in the realm of promoting financial stability, or of managing and resolving crises, making recommendations and setting standards of good practice is the easy part. The actual implementation of good policies is much more difficult. There are many impediments to getting the right thing done: inertia, vested political and oligopolistic interests, often inadequate laws and judicial follow-through, and undertrained (and sometimes underpaid) public servants. These influences are at play in all countries, though most obviously in emerging market economies. Against these factors, incentives to do the right thing must be put in place: market discipline, surveillance, external technical and financial support and, above all, a heightened sense of self- interest. The benefits that derive from an efficient yet safe financial sector can hardly be overestimated. They are well worth the substantial efforts which might be needed to attain them.