Speech by Mervyn King, Deputy
Governor
20th Anniversary of the Indian Council for
Research on International Economic Relations
The International Financial System: A New
Partnership
Introduction
It is both a great honour and a real delight to make my first visit to India in order to commemorate the 20th Anniversary of ICRIER. I feel especially honoured to be invited to deliver the first K. B. Lall Lecture. The theme of my lecture will be the international financial system, and the need for a new partnership between the developed economies and the developing or emerging market economies. The international financial institutions set up at Bretton Woods over 50 years ago were designed to deal primarily with problems of current account imbalances. In recent years, however, major financial crises have originated in the capital account. Why have these occurred, and how can we make them less likely in future and improve our ability to deal with them when they do occur?
I can think of no better place to discuss this subject than ICRIER. The links between international flows of private and public finance, and the need for a restatement of the role of the international financial institutions and a clearer understanding of their modus operandi, are subjects at the heart of the work of ICRIER. Your Founder Chairman, Dr K B Lall, whom I am delighted to say is with us this evening, worked tirelessly over the years in so many areas of international economic policy, including a period as Chairman of the General Agreement on Tariffs and Trade. Your current Chairman, Dr I G Patel, served India not only as Governor of the Reserve Bank but also at the International Monetary Fund. I owe IG a great debt for his help and support at the London School of Economics during his period as Director in the 1980s. And the new Independent Evaluation Office of the IMF, an important development in improving the accountability of the Fund, will be led by Mr Montek Singh Ahluwalia, one of India's most distinguished economists, and the husband of your Director, Dr Isher Judge Ahluwalia.
India was one of the 44 countries which participated in the meetings at Bretton Woods which led to the creation of the International Monetary Fund and the World Bank. The Bank of England archives contains some fascinating material on the Bretton Woods Conference and, in particular, on the role of the Indian Delegation. A cable from the Foreign Office to the British Embassy in Washington on 1 June 1944 argued that there were strong grounds for giving India a seat on the Agenda Committee. The reply reported US opposition to this idea, in part because Canada and Australia were already represented on the Committee. The resulting compromise was the product of bureaucratic genius. An Indian representative was allowed to participate on the Committee provided that there was no public announcement of the fact. In the end, common sense prevailed and India received an official invitation to attend the Agenda Committee.
Recent International Financial Crises
Perhaps the key difference between the world of Bretton Woods and the world today is the size and volatility of private capital flows. Then, as now, it was recognised that no system could ensure the compatibility of:
| (i) | Domestic monetary autonomy; |
| (ii) | Stable exchange rates; |
| (iii) | Free capital mobility. |
This "impossible trinity" has been at the heart of the debate on the international monetary and financial system for many years. A sustainable system must sacrifice one of these three objectives. Some countries have decided to abandon the first leg of the tripod, namely domestic monetary autonomy. In Europe, twelve countries have formed a monetary union, and elsewhere, such as in Argentina and Hong Kong, currency boards, linked to the dollar, have replaced discretionary monetary policy. Other countries have abandoned the attempt to maintain rigidly fixed exchange rates, and adopted a combination of domestic monetary management based on an inflation target and a floating exchange rate. Examples include both developed economies, such as the UK and Canada, and emerging market economies, such as Brazil and South Africa. There are arguments for and against both of these approaches. But what is clear is that both in theory and practice there is now a recognition that pegged (fixed but adjustable) exchange rates do not provide a viable long-term middle course. More interesting, perhaps, is the absence of serious debate on the merits of the third position, namely the willingness to forego freedom of capital movements in order to retain domestic monetary autonomy and stable exchange rates. That is perhaps surprising in the light of the experience of the two major countries in Asia that escaped the financial crisis of 1997-98, namely India and China, which had in common the presence of capital controls.
The willingness to impose controls on capital movements, at least temporarily, was certainly evident at the Bretton Woods Conference. Mindful of the weakness of Britain's national balance sheet, Lord Keynes, urged on by the Bank of England, argued that there should be no legally binding obligation to make the sterling balances convertible into dollars, not least those held by India. The vulnerability of countries to financial crises when private capital is freely mobile was uppermost in the mind of officials then, and has recently returned to prominence in the wake of recent crises.
Capital flows do, however, bring real economic benefits. They enable savings from around the world to move to those countries with the most profitable investment opportunities, benefiting lenders and borrowers alike. And such capital flows also transfer knowledge and expertise. The most important task of any financial system is to guide the allocation of scarce capital. As Larry Summers, the former US Treasury Secretary, said earlier this year, "If you are looking for reasons why some countries succeed and why other countries do not succeed in the new global economy, a very large part of it goes to the greater success of the successful countries in channelling capital into the right places, and then making sure that it is used in a disciplined way."
At Bretton Woods it was thought that post-war reconstruction could be financed by capital provided by the new World Bank, the twin of the International Monetary Fund. Hence it would be possible to finance reconstruction from long-term investment supplied by official institutions, without the necessity of allowing free movement of private capital that might raise the problem of the "impossible trinity". Soon, however, the demand for capital imports exceeded the ability of international institutions to supply loans. Private capital markets came into their own. The expansion of private capital flows was gradual, and, until about ten years ago, was primarily concentrated on flows among the developed economies. It is only in the past ten years that the explosion of private capital flows to emerging market economies has taken place. Data from the World Bank suggest that annual capital flows to emerging markets were less than $10 bn in the early 1970s, rising to around $300 bn in the late 1990s. At their peak, capital flows to emerging markets were around 5% of those countries' GDP.
Unfortunately, capital flows on this scale can reverse themselves as suddenly as they appear. The result of such sudden and large reversals of short-term capital flows has been a series of international financial crises, in Mexico in 1994-95, Asia in 1997-98, Brazil in 1998-99, and, more recently in Turkey and Argentina. The frequency and scale of these crises, if they were to continue, would be a serious impediment to the evolution of the international capital market.
Unless we can both reduce the frequency and severity of such crises, and improve our techniques of crisis resolution, then the demonstrators against globalisation will be provided with unnecessary ammunition.
The costs of recent crises have been large. Between 1996 and 1998, the reversal of private capital flows to the five Asian countries primarily affected (Indonesia, Korea, Malaysia, the Philippines and Thailand) was almost $150 bn, equivalent to about 15% of the pre-crisis level of GDP. Changes in the capital account imply equal and opposite swings in the current account. Inevitably, a change in the current account on that scale is likely to mean a deep recession. And during the Asian crisis, real GDP fell by 1% in the Philippines, 7% in Korea and Malaysia, 11% in Thailand and by 13% in Indonesia. Several years of economic growth were wiped out, leading in some cases to political instability. A recent study by Robert Barro shows that countries that experience financial crises can expect to return to their pre-crisis rate of economic growth only after about five years on average. Although the recovery of the afflicted Asian economies in 1999-2000 was more rapid, the fall in investment is likely to affect their level of output for several years.
It is clear, therefore, that it is dangerous for countries to sail unprepared into the deep waters of international capital markets. One of the important lessons of recent crises is that not all capital flows are equally dangerous. Most of the reversals in capital flows to the Asian countries were in the form of swings in short-term debt finance - about 80% resulted from changes in the net flows of finance from commercial banks. A build-up of short-term debt creates vulnerabilities in a country's national balance sheet. Where there are significant mismatches in either maturity or currency obligations, then a country is vulnerable to a liquidity run. In such a situation sudden reversals of capital flows can occur on a huge scale. That is the hidden cost of debt finance. In contrast, equity capital does not involve the risk of crises that are associated with the possibility of interruptions to payments on debt finance. This is because equity investment has a self-stabilising mechanism. Investors cannot withdraw from the equity market without finding a buyer to replace them. The market price adjusts in order for the seller to attract a buyer. Of course, the market price may move sharply and rapidly, and impact spending decisions. But crises resulting from payment interruptions are a feature of debt finance. Lenders whose bonds mature or who choose to withdraw their deposits do not have to find a buyer. Instead they simply exit, and, if they do so on a sufficiently large scale, countries can find themselves facing a liquidity run.
Interest rates on lending to a country vulnerable to a run can rise to extremely high levels. If the market anticipates that a country may allow inflation to rise, or the exchange rate to fall, to alleviate the burden of domestic currency debt then interest rates on debt can rise rapidly. It is only ten years ago since overnight interest rates of several hundred percent were seen during the ERM crisis in Europe. But even on foreign currency borrowing, a country may find that the spread of the interest rate at which it can borrow over the interest rate charged to "safe" countries, such as the United States, can reach extremely high levels. And where that interest rate is markedly higher than the growth rate of the economy, the debt burden rapidly becomes unsustainable unless the interest rate can be brought down quickly. Before the Russian default and devaluation of 1998, the average spread on emerging market sovereign debt was around 500 basis points, in itself a sizeable addition to the burden of borrowing in international capital markets. That spread then rose sharply to levels of between 1000 and 1500 basis points. At these levels debt burdens were clearly unsustainable. Since then, macroeconomic reform in many of the affected countries, with the help and support of the IMF, has reduced average spreads to a range of between 600 and 800 basis points, although there has been a further rise recently following the difficulties in Turkey and Argentina.
Emerging market spreads are currently around 900 basis points over interest rates on US Treasury bonds. It is not easy to reconcile spreads at this level with the fact that sovereign debt defaults on bonds are running at historically low levels. In part, the low level of defaults may reflect the increase in the number of exceptionally large loans made by the IMF in recent years. So what does account for the high level of emerging market spreads? There appears to be a good deal of uncertainty concerning the conditions surrounding the availability of official finance. It could be that investors are demanding higher interest rates to compensate for that uncertainty. There is also uncertainty among market participants about what would happen in the event of a debt restructuring and the expected repayments that would ensue. One role for the official community is to try and mitigate these uncertainties through greater clarity about the criteria for official lending and its crisis resolution policies - a theme I return to below. If successful, one outcome of these reforms would be lower borrowing costs for emerging markets.
Investors are also starting to differentiate among borrowing countries more clearly than before. This has led to a greater dispersion of spreads on emerging market debt. Before the Russian crisis these spreads were tightly compressed with the central 50% of the distribution of emerging market spreads covered by a range of only around 100 basis points. At present, the range covered by the central 50% of the distribution is over 500 basis points. Correspondingly, there has been a sharp fall in the correlation between changes in emerging market spreads. The rolling 26 week correlation between changes in emerging market spreads reached a peak of around 0.8 at the time of the Russian crisis. Since then it has steadily declined, reaching a level of only 0.2 before the recent rise to around 0.35 following adverse developments in Turkey and Argentina. This differentiation in spreads is a welcome development. It shows that the possibility of contagion from a country affected by a crisis to others initially unaffected is less than might have been the case only a few years ago. Of course, if a country did default then correlations might rise significantly, but the focus of attention on recent crises and their causes has led investors to appreciate that many have been country-specific. Although there is no room for complacency, the lower correlation of spreads and the greater differentiation of risk assessments represents an increase in the efficiency in the way capital markets operate.
The experience of recent crises prompts two questions. First, what can countries do to protect themselves from the risk of further financial crises? Second, what should be the responsibility of the international community towards emerging market economies?
A New Partnership
We need a new partnership between emerging market economies and developed economies. It is useful to distinguish between measures to improve economic performance and prevent financial crises, on the one hand, and ways to resolve crises once they have occurred, on the other. In this section I deal only with the former. Crisis prevention should be at the heart of the policies of both emerging markets and the international financial institutions.
In terms of prevention, recent experience suggests five lessons for the future.
(1) First, it is important that borrowing countries, especially those without a track record of international borrowing, monitor and manage the maturity and currency composition of their national balance sheet. This is not a trivial matter. Most countries have inadequate information on the composition of their external liabilities, especially those of the private sector. Nevertheless, monitoring and managing the exposures of the public and financial sectors are important to avoid a build up of potential vulnerabilities. In this respect, the IMF can play a helpful supportive role by providing assessments of vulnerability as part of the Article IV process, the new joint IMF/World Bank financial sector assessment programmes and debt management guidelines, and technical assistance on the data requirements implied by the need to monitor national balance sheets.
(2) Second, limitations on official finance mean that countries should think carefully about the provision of self-insurance against a liquidity crisis. A simple, but often expensive, way to do this is to build up large foreign currency reserves, a strategy taken to heart by a number of emerging markets, including China and also Korea following its crisis. A potentially superior alternative is the creation of contingent credit facilities with both official and private sector creditors. So far, even at the high spreads on emerging market debt, these facilities have not proved attractive, and the CCL facility created by the IMF has lain dormant. The next few years will be a test of the value of such facilities.
(3) Third, experience has shown the value of borrowing countries establishing good relationships with creditors well before any possibility of difficulty in repayment arises. The creation of investor relations programmes and the regular briefing of creditors about developments in economic policy can play a role in providing the information which the market requires to assess the riskiness of sovereign loans. It is never too early to build a relationship with actual or potential creditors. The IMF, in collaboration with the private sector, have recently drawn up a set of guidelines that countries might usefully follow when setting up an investor relations programme. The second aspect of relationships with creditors is the insertion of collective action clauses in sovereign debt contracts. This proposal, advanced originally in a G10 Deputies' report in 1996, has gradually become accepted as a sensible step forward. The UK has introduced collective action clauses into its foreign currency debt instruments. And just last month, the G7 Finance Ministers agreed on the importance of introducing collective action clauses into debt contracts to facilitate crisis management. Again, time may be on the side of reform.
(4) Fourth, in the long run, the best way to avoid the problem of liquidity crises is for the composition of capital flows to emerging markets to move away from debt, both bank and bond, finance towards portfolio equity and direct investment. Shocks to the borrowing country would take the form of a fall in equity prices, not a liquidity run with its associated risk of a financial crisis and the need for external finance. Encouragingly, the pattern of capital flows to emerging markets is already evolving in that direction. Bank lending represented around 80% of capital flows to emerging markets during the 1970s. But by the end of the 1990s, FDI accounted for around 80% of emerging market capital flows, with portfolio equity accounting for much of the remainder. As Ken Rogoff, the new Economic Counsellor to the IMF, pointed out in 1999, there are still several biases towards debt rather than equity finance in capital flows to emerging markets. One of these is deposit insurance in both creditor and debtor countries which makes it more difficult for the authorities to avoid being seen as providing some implicit support to international loans by domestic banks. There is no easy answer to this problem, but a shared concern in both borrowing and lending countries is the implicit insurance which both sides are giving to large parts of the financial system. The moral hazard so created is not restricted to international lending, but it does affect the incentives for the form of investment in emerging markets. We need also to guard against institutional or regulatory mechanisms - both international and domestic - which favour short-term over longer-maturity capital flows. In the long run, the solution is for emerging markets to create legal structures and a stable economic policy environment that provide the confidence to support inward equity investment in their economies.
(5) Finally, greater transparency allows better informed decisions by both borrowers and lenders, and reduces the risk of contagion by allowing markets to differentiate among borrowers. Much has been said, and, more importantly, achieved in the area of transparency in recent years. There has been an explosion of codes and standards on different aspects of economic and financial policy in recent years. So much so, that some countries are claiming that the process needs to slow down. As part of the new partnership transparency must be based on three principles.
First, the appropriate codes and standards for a country at one stage of economic development may not be appropriate to countries at other stages. I say "may" advisedly because in each individual case the argument has to be made. But codes and standards should reflect different stages of development.
Second, countries must make clear to which codes and standards they are actually adhering. That is why countries should not be able to opt out of "transparency about transparency". Following the production of pilot transparency reports on a number of countries, including the UK and Argentina, the IMF has now made rapid progress in producing reports on the observance of standards and codes (ROSCs). As of April 2001, 110 ROSC modules had been completed for 43 countries, of which 76 have been published covering 31 countries. India has recently had a ROSC published on fiscal transparency. It is critical that monitoring of the observance of standards and codes be fully integrated into IMF surveillance under Article IV. Here implementation is urgent.
Third, in the area of transparency, ownership is all. Transparency cannot and should not be imposed on any country. Countries themselves benefit most from being transparent and releasing the reports of assessments by the IMF about their financial systems. And the enthusiastic embracing of transparency by a number of emerging markets has paid off in terms of better relations with creditors.
In the field of transparency, the key elements of the new partnership should be, first, a commitment by emerging market economies to implement transparency about transparency by publishing ROSCs, and, second, new opportunities for emerging markets to engage in the process of constructing and developing codes and standards. There is encouraging evidence of greater collaboration and consultation between developed and developing countries in the design and implementation of the core standards - for example, on banking, securities and insurance regulation, data, payment systems, insolvency and transparency of monetary, fiscal and financial policies. That is all to the good and there is further to go. Ultimately, the most successful route to enhance the influence of emerging markets on the development of standards and codes is to strengthen the role of the IMF and the World Bank, the twin institutions that can claim legitimacy through the membership of 183 countries. Under British Chairmanship, efforts have been made to increase the effectiveness of the meetings of the IMFC which represents all countries around one table.
Private Finance and Public Funds
In terms of the resolution of crises less progress has been made. The problem would be easier to solve were it possible to distinguish between two rather different sources of crises - a liquidity-based problem caused by a currency or maturity mismatch in a country's national balance sheet despite a sustainable macroeconomic and debt position; and a fundamentals-based problem which means that the debt burden is unsustainable or the exchange rate or other key macroeconomic policies need to be altered. In the former case, the provision of liquidity support by the international community might help to bridge to a position in which the country could re-engage with its private creditors. In the latter case, the main requirement is not liquidity support but a change in macroeconomic or debt management policies. Recent crises have seen examples of both types of problem.
Liquidity runs typically occur because of a co-ordination failure among creditors. There are two solutions to this co-ordination problem. The first is a lender of last resort that is able to provide liquidity support quickly and on a large - indeed, potentially unlimited - scale for a short period to enable the affected country to meet its obligations. Such loans would normally be short-lived, and should be made available at an above-market interest rate such that this sort of finance is seen as last not first resort. The second solution to the co-ordination problem is for the borrower to impose a temporary suspension of payments to create a "time out" during which the borrower can negotiate directly with the creditors, and so arrange a new profile of repayments of debt. Both approaches, if understood and implemented consistently over time, can provide an efficient solution to the co-ordination problem and eliminate the incentives for a liquidity run.
But, as Ken Rogoff pointed out in 1999, the lender of last resort approach carries with it the risk of introducing significant moral hazard into the loan market. If lenders believe that sovereign borrowers are likely to be bailed out, then their incentive to assess the riskiness of their loans will diminish. Equally, borrowing countries will find it more attractive to claim that the measures necessary to continue servicing their debt are "politically impossible" if they believe that there is an international deep pocket willing to extend loans and defer the moment when the national balance constraint is binding.
There are two good reasons for the IMF not being able to play the role of an international lender of last resort, at least for the foreseeable future. First, the moral hazard created by both lenders and borrowers cannot simply be assumed away. It is not easy to quantify, but it is noteworthy that the number of sovereign defaults has declined quite sharply during the 1990s. If sovereign risk is mis-priced by private capital markets, this sows the seeds of future crises. The increased provision of official finance would proceed hand-in-hand with an increased incidence of crises.
Second, to be effective, a lender of last resort must have the ability to extend sufficient resources that the market has no doubt whatsoever about the ability to provide whatever it takes to deal with the immediate crisis. The IMF is not in that position. There is no political commitment to provide the IMF with unlimited funds. As the finance ministers and central bank governors representing all IMF member countries said in their communiqué of the IMFC last September in Prague: "The Committee notes that Fund resources are limited and that extraordinary access should be exceptional …".
In practice, however, exceptional access has often been more the norm in recent years. Normal access is typically defined as 300% of IMF quota. During the Asian crisis, Korea's programme was almost 2000% of quota and Thailand's over 500%. More recently, Turkey's programme was over 1500% of quota and Argentina's 500%. If creditors and debtors continue to believe that exceptional access is readily available, then international credit will be over-extended and the incidence of crisis will increase.
One reaction to these extremely large packages - "bailouts" - and the accompanying moral hazard is simply to say that the official sector should have no part to play in what is essentially a private international capital market. Official lending is now small relative to private capital flows. Over the last three years, private flows have been around 7.5 times greater than official flows, according to IIF data. Against that backdrop, some have argued that the IMF should be abolished. This would be to throw out the baby with the bath water.
What is needed is a "middle way" between full IMF insurance and no insurance at all. This middle way would comprise IMF lending but within strong presumptive limits. A key principle underlying this approach is that the international community needs to set out as clearly as possible the criteria that will govern the size and scope of IMF lending. Since most agree that there are limits on IMF lending, there is merit in explaining those limits to both potential borrowing countries and their private creditors. This would enable debtor countries better to plan their policies. It would also allow creditors to assess risk more accurately. Indeed, put more controversially, how can sovereign risk be accurately assessed without clarity about the Fund's role? A lack of clarity about the likely response of the international community to potential crises is a recipe for inaccurate assessments of risk. Such uncertainty would add to the (already high) cost of borrowing by emerging markets.
So far, the "middle way" seems a statement of the obvious - namely that there are limits to IMF lending and that there is merit, for debtors and creditors, in having clarity about those limits. The other side of this coin is that, on occasions, there will be countries that have run up unsustainable debt burdens, or face severe liquidity pressures, and who have little alternative but to restructure or reschedule their debt. Perhaps this is why the international community has moved significantly in the direction of giving "private sector involvement" a greater role in the resolution of financial crises than was typically the case in the late 1990s. The IMF communiqué in Prague last year, the G7 Finance Ministers statement last month, recent joint work by the Bank of England and Bank of Canada, and, significantly, speeches by the new Managing Director of the IMF, Horst Kohler, have all emphasised the need to move further in the direction of greater private sector involvement. As one example the Report of the G7 Finance Ministers to the Heads of State and Government only last month stated that, "While the IMF has an essential role to play, official resources are limited in relation to private financial flows. The engagement of private investors is thus essential for the resolution of payment imbalances in crises… the official sector needs to avoid creating expectations that private creditors and investors will be protected from losses or that official resources would be used to finance large, sustained capital outflows. … We underscore the need for further progress".
The opponents of this approach raise two important questions. First, is it possible to define limits on IMF lending? Second, is it acceptable for the official sector to countenance default by a sovereign borrower when the consequences for both the country and the international financial system could be devastating? I shall try to provide answers to both of these questions.
It is true that recent crises have been the result of developments in the capital account rather than the current account. As capital flows have grown, so too have the potential demands on the official community as they attempt to fill capital account financing gaps. That is why so many more programmes have involved "exceptional" access in relation to quota. So perhaps, in a world of capital account crises, exceptional access should become the norm?
This argument is superficially quite attractive. But its implications need to be assessed carefully. It would mean that IMF resources would need to increase in line with private capital flows even for the IMF to maintain its current role. And since 1970, capital flows have grown around four times as fast as world incomes. The share of world GDP devoted to resourcing the IMF would grow rapidly over time. In fact, as private capital markets came to understand this, the scale of private capital flows could increase to an even greater extent. Private creditors and debtors would accumulate ever larger bilateral debts, safe in the knowledge of a multilateral insurance mechanism. The logical end point of this game is that the international community would be locked into providing ever increasing sums of money to countries in difficulty - in short, an international lender of last resort would be created by stealth. There is no evidence that anyone wishes to go down this route.
To guard against this, it is crucial that there be some clearer presumption about the scale of "normal" access. That scale may well be higher than was the case in a world of current account crises. And the approach of defining limits in terms of multiples of quota may be unsatisfactory because the size of quotas in some cases needs to be revisited. But that is not an argument against the principle of presumptive limits; it is a case either for reconsidering quotas or, in a more practical vein, relating access to finance to some other metric.
The key to limiting lending is not strict rules but stronger presumptions. These presumptions then provide the backstop for debtor-creditor negotiations and help condition expectations in financial markets. Exceptional lending above this presumptive limit would be possible in order, to provide operational flexibility in extreme cases - for example, those threatening systemic stability. That is why the framework is one of presumptive limits rather than strict rules. But granting exceptional access should require much greater ex ante justification and ex post accountability. For example, exceptional access programmes should be automatically referred to the new Independent Evaluation Office of the IMF. This would raise the hurdle for granting exceptional access and provide greater clarity to debtors and creditors about the support countries could expect from the official community.
The logical consequence of limited official finance is that inevitably there will be times when a reprofiling of sovereign debt may be necessary for some countries. Some have argued that sovereign debt restructuring or default is potentially too disastrous to contemplate, for the country or indeed for the world economy. The Russian default in 1998, and the disruption to world markets that followed from it, is often cited as evidence for the prosecution.
More careful analysis suggests two rather different conclusions from the Russian experience. First, the Russian default was disruptive in part because it came as a surprise to market participants. Private creditors had planned on one assumption - exceptional IMF financing, or the "moral hazard" play as it was labelled by the market - and were surprised when their comfort blanket was removed. Greater clarity and stronger presumptions about the size and form of Fund financing would have reduced the surprise and the accompanying contagion. Expectations of debtors and creditors would have been conditioned ex ante and the severity of the crisis thereby reduced ex post.
Second, the Russian experience illustrates the importance of having orderly rescheduling mechanisms in place. In Russia, it was not that they defaulted but the way that they did it which generated costs. The default was disorderly - neither efficient, nor equitable, nor expeditious. And that contributed importantly to the disruption to Russia and more widely. But rescheduling need not be like that. There are dead weight costs to disorderly default. So there are lump-sum gains - to both debtors and creditors - to having orderly mechanisms in place.
This is where the official sector has a role to play. A key principle of any crisis resolution framework is that decisions on a sovereign's debt are the responsibility of the borrowing country, in consultation with its private sector creditors. Neither the IMF nor any group of countries should tell a country to restructure its debt. There are a range of options open to debtors in dealing with their creditors at times of crisis. Countries with a good track record of repayment and long standing relationships with their creditors may be able to borrow more from the international market. Others, facing more severe liquidity pressures, may seek to undertake voluntary rescheduling or rollover of debt by bringing together all or some of their major creditors, as in the cases of Korea and Brazil. Those facing unsustainable debt positions may seek to negotiate market-based write-downs of their debt, as in the cases of Ukraine, Pakistan and Ecuador. And in yet another set of cases, a country may find it necessary to impose a temporary timeout on payments to all creditors, to give themselves some breathing space to address macroeconomic or co-ordination problems. The decision on exercising any of these options must rest with the debtor country.
The role of the official sector is to ensure that the full menu of financing options is made known and available to the debtor, from which it then chooses. This menu should include both the easier options - such as raising new private sector money - as well as the harder ones - such as suspending payments. Each of these options is backstopped by limits on IMF lending, so that the "pure bail-out" option is heavily circumscribed. Indeed, it is this backstop which helps provide the incentive for creditors and debtors to seek alternative, market-based solutions sooner.
The IMF should stand ready to assist countries, whichever of these options debtors choose to exercise, provided the appropriate prior conditions are satisfied. For example, should a country facing severe liquidity pressures decide to suspend payments temporarily, then the IMF should be willing to support that decision while remedial policy measures are put in place. This support could take the form of bridging finance - so-called IMF lending-into-arrears. The pre-conditions of lending-into-arrears could be designed to ensure that payments suspensions are handled in an orderly fashion - for example, that they are time-limited and equitable, and the debtor is negotiating in good faith. This would reduce uncertainty on the part of both debtors and creditors as to how the end game would be played out, thereby reducing borrowing costs.
These sets of procedures represent evolution in, rather than revolution of, the international financial architecture. They are about ensuring the official sector's own actions are clear, consistent and accountable, so that they contribute effectively to the resolution of financial crises. These procedures are fully consistent with the principles of private sector involvement outlined in the IMFC communiqué last September and more recently by the G7 Finance Ministers last month. The framework set out here is an attempt to begin to add some operational meaning to those overarching principles.
To sum up, both borrowing countries and private sector creditors must expect that, except in exceptional cases of systemic concern, the limits on official finance means that they and they alone will be responsible for dealing with a resolution of problems concerning debt repayment. Standstills and debt restructuring will be only one of many options open to negotiation between debtors and creditors. The IMF should not attempt to impose a solution on borrowing countries. It should be willing to lend into arrears in circumstances where countries have chosen the route of a standstill and its associated conditions. But it should not create expectations that exceptional access is the norm.
Conclusions
Progress can be made only by closer co-operation between the developed and developing countries. The development of standards and codes, the design of IMF lending and the wider agenda of trade liberalisation and international co-operation are all part of the new partnership of which I have spoken. The closeness of the relationship between Britain and India is a compelling reason for our working together in the various international fora to improve the international financial system. In his final report to the British Government on the creation of the Bretton Woods system, Maynard Keynes wrote that "the excellence and closeness of our relations with the Indian delegation deserves special comment. Sir Chintaman Deshmukh (Governor of the Reserve Bank of India) handled his case with high dignity, ability and reasonableness; we always supported him on his interests and he always supported us on ours". Perhaps our joint work on international financial architecture will recall the common architectural heritage of the Bank of England and official buildings in New Delhi. Herbert Baker, who, with Lutyens, was responsible for the design of early New Delhi also rebuilt the Bank of England in the inter-war period. In the upper storeys of the Bank he placed pavilions which are derivatives of the ends of the Secretariat blocks in New Delhi.
As Lord Keynes said in his speech at the closing plenary session of the Bretton Woods Conference on 22 July 1944, "it has been our task to find a common measure, a common standard, a common rule applicable to each and not irksome to any. We have been operating, moreover, in a field of great intellectual and technical difficulty. We have had to perform at one and the same time the tasks appropriate to the economist, to the financier, to the politician, to the journalist, to the propagandist, to the lawyer, to the statesman - even, I think to the prophet and soothsayer. … We have shown that a concourse of 44 nations are actually able to work together at a constructive task in amity and unbroken concord. If we can continue in a larger task as we have begun in this limited task, there is hope for the world." The IMF is still the only international body with the legitimacy, as well as staff and expertise, to build and defend a successful international financial system. It is important, therefore, that we not accept uncritically the way it has developed, and we must examine closely how far changes in international financial markets require us to understand the consequences of limits to IMF lending and the implications of those limits for borrowing countries and private creditors alike. We need neither a grandiose new plan nor another Bretton Woods Conference. But we do need greater clarity and less fudge about how the present system is supposed to operate.
