TRADE AND DEVELOPMENT REPORT, 1998

Offprint from Part One (electronic version prepared by Dr. Róbinson Rojas from the .PDF version)

Chapter IV

THE MANAGEMENT AND PREVENTION OF FINANCIAL CRISES

A. Introduction

This chapter discusses key issues regarding both the management or containment of financial crises and policies that could help to prevent them. Here it is necessary to distinguish between banking crises (which frequently include runs on parts of the banking sector) and currency crises (which involve flight from the currency by residents and non-residents), although the two may be, and (as discussed in chapter III) in developing countries usually have been, closely connected.

In the case of banking crises a conceptual distinction cannot reasonably be made between management, on the one hand, and prevention on the other; strengthened financial regulation and supervision, for example, are manifestly directed at meeting both objectives. Chapters III and IV focus principally on crises where attacks on the currency were accompanied by threats to the banking system. In such cases the distinction between management and prevention is useful, and the section which follows concerning crisis management and resolution discusses policies which can be adopted in response to a currency attack for the purpose of halting or reversing it and so limiting the resulting damage to the domestic economy.

The subjects taken up under crisis management are macroeconomic policies, management of reserves and access to credit, international lender-of- last-resort financing, and international standstill and workout procedures for debtor countries. In view of the high costs and uncertain outcomes associated with reliance on domestic policies in the debtor country under attack and on external financing under arrangements which are currently in place or can reasonably be envisaged, special attention is focused on the last of the approaches mentioned, i.e standstill and workout procedures.

Measures for crisis prevention can be taken at global, national or regional levels, and the treatment here takes up policies and proposals classified under headings which broadly follow that order. However, the policies and proposals surveyed do not always fit neatly into this pattern. Global surveillance, for example, clearly belongs to the first of the three levels mentioned, and regional consultation and collaboration to the third. But in the case of other measures (such as financial regulation, controls over international lending and portfolio investment, capital controls and exchange rate policies), even though action generally takes place at the national level, in recent years such measures have been increasingly the subject of global or regional initiatives because they have significant cross-border spillovers, involve free-rider problems (arising from the advantages accruing to a country from other countries’ compliance with rules, standards or norms which it does not itself observe), or restrict national policy autonomy.

As explained below, the first two considerations have been particularly important for international initiatives regarding financial regulation and supervision, and also for multilateral cooperation at regional and global levels to prevent disorder in currency markets and competitive devaluations. The third consideration motivates the global regime for currency convertibility for current international transactions, and various regimes agreed by smaller groups of countries for the removal of restrictions on capital transactions. The WTO regime for international trade in goods and services reflects all three considerations, and its agreements contain provisions explicitly designed to deal with problems under each of them.

B. Managing and resolving financial crises

1. Self-fulfilling debt runs

While every financial crisis in developing countries is different, such crises have a common feature: the rush of investors and creditors to exit and the consequent financial panic. Indeed, whatever the proximate causes of financial crises or the events that trigger attacks on currencies, international investors and creditors of developing countries often manifest herd-like behaviour in exiting as well as investing or lending. The debt crisis of the 1980s witnessed a drastic cutback in lending by international banks to sovereign debtors, while during the 1994-1995 Mexican crisis the rush for the exits by international creditors took the form of rapid liquidation of government paper and conversion of the proceeds into dollars.

Again, in the more recent turmoil in East Asia, the refusal to roll over short-term loans together with the attempt of unhedged debtors to avoid exchange rate losses was the principal factor deepening the crisis. Creditor overreaction to debtors’ financial difficulties is often explained in terms of a collective action problem. Even though the creditors as a group are better off if they continue to roll over their maturing claims on a debtor, an individual investor has an incentive to rush for the exits. A debtor who could normally generate sufficient resources to service his outstanding stock of debt would face a liquidity problem if more than a certain number of creditors refused to renew their maturing claims.

Without access to liquidity, he would be forced to curtail operations or to resort to distress sales of assets, which in turn would lower his income and wealth, thereby further constraining his ability to service debt and hence damaging the interest of creditors as a group. In this sense, debt runs reflect the failure of markets to coordinate individual decisions so as to generate a superior outcome for the creditors as well as the debtors. The consequences of a generalized debt run by international creditors triggered by a loss of confidence are much more serious than those of the debt run by creditors of domestic debtors. Such behaviour can easily turn a liquidity problem into widespread insolvencies and defaults by altering key asset prices, interest rates and exchange rates.

In the absence of a large stock of reserves or access to international liquidity, the ability of a debtor developing country to repay its entire stock of short-term external debt on demand is no greater than the ability of a bank to meet a run by its depositors. Where external liabilities are in the form of direct securities denominated in domestic currencies, as was the case with Mexican cetes and tesobonos, the demand for foreign currency comes directly from the creditors.

In the case of bank lending, withdrawal of loans by foreign creditors could trigger a rush by unhedged debtor banks and firms into foreign currency as they seek to pay debt or cover their open positions. That would in turn drive down the value of the domestic currency and raise interest rates, making it more difficult for debtors to service their debt and forcing them to liquidate assets, thereby deepening the debt deflation process.

It is not only the international debtors that would thus be hurt; there would also be broader macroeconomic consequences, including a sharp decline in employment and output. Additional pressures on exchange rates and asset prices would arise from two other sources:

First, residents tend to flee from domestic currency assets, and can do so easily when the economy is dollarized and there is easy access to foreign exchange assets.

Second, debt runs by foreign creditors are often associated with a flight from non-debt instruments held by non-residents, notably from the equity market.

Since such investors face a decline in prices when they attempt to liquidate their holdings, the selling pressure in the currency market would be weakened. Moreover, since they would also suffer from depreciations, they may have less inducement to exit. However, investor overreaction could still amplify destabilizing feedbacks between equity and currency markets. Indeed, there has been a very close correlation between the collapse of equity prices and exchange rates in recent episodes of financial crisis in developing countries, and this linkage has been particularly strong in East Asia.

Theoretically, there are four lines of defence against a massive attack on the currency of a debtor country:

        • domestic policies, particularly monetary and interest rate policy, to restore market               confidence and halt the run;

        • hedging by keeping sufficient foreign reserves and credit lines;

        • use of an international lender-of-last-resort facility to obtain the liquidity needed;

        • a unilateral debt standstill and exchange restrictions, and initiation of negotiations for           an orderly debt workout.

The last three mechanisms affect not only crisis management but also the likelihood of emergence of debt crises by discouraging runs against the currency. The threat of a unilateral debt standstill could also dampen short-term capital inflows, thereby reducing the build-up of external financial fragility. The following sections discuss the feasibility and costs and benefits of establishing and/or using such mechanisms.

The policy response to a debt run has generally proved ineffective, and building up reserves to meet speculative attacks is extremely costly and barely practicable. In addition, there are serious difficulties in setting up an international lender-of-last-resort facility to provide the kind of liquidity needed to counter such attacks on a currency. An effective way of dealing with them would be to establish an international framework for debt standstills and workouts to prevent the resulting liquidity crises from leading to insolvency.

2. Monetary policy and market confidence

Interest rate differentials are undoubtedly an important determinant of international capital flows. Higher domestic interest rates, ceteris paribus, would stimulate capital inflows by increasing the profitability of arbitrage with foreign money markets. Also, they could signal the determination of policymakers to remove certain macroeconomic imbalances, such as excess domestic spending and large external imbalances, when these threaten to put pressure on the currency. Under such circumstances, restrictive monetary policy and higher interest rates can play an important role in stabilizing capital flows.

However, as the events in East Asia show, when financial markets panic the likely effects of monetary tightening and higher interest rates on capital flows are quite different, because they exert a strong influence on credit risk. The withdrawal of foreign lending and flight from the currency began in the first place because lenders and investors did not expect to receive the return on their assets. Higher interest rates simply signal declining creditworthiness and rising default risk, and the expected rate of return adjusted for risk will tend to fall as interest rates are raised. For international lenders with claims denominated in dollars, higher domestic interest rates in the debtor country do not alter the rate of return on their assets. But by increasing the financial difficulties of their debtors and reducing their incomes and net worth, they increase the likelihood of default. Thus, they provide no incentive for foreign lenders to roll over their existing loans or extend new credits.

Again, high interest rates are not always effective in stemming capital flight into foreign currency triggered by expectations of sharp depreciations. Even double-digit rates are unable to persuade people to keep their capital in domestic currency assets when they believe that such rates are politically difficult to maintain, as seen in some European countries during the 1992-1993 European Monetary System (EMS) crisis, and domestic assets have high default risks.

If persistently applied, monetary tightening and high interest rates can no doubt eventually stabilize the currency by intensifying the difficulties of the debtors and increasing bankruptcies and defaults – that is, by reducing the sales rather than by increasing the purchases of domestic currency. As debt deflation and recession deepen, debtors will become increasingly insolvent and unable to raise funds to purchase foreign exchange to service their debt or to hedge against the exchange rate risk. However, markets would be stabilized by depressing the economy and increasing defaults rather than by bringing back the foreign capital.

Quite apart from the ineffectiveness of monetary tightening in stemming self-fulfilling debt runs, there is also little economic justification in defending the exchange rate at the expense of a hike in interest rates. Devaluations tend to hurt primarily those who have currency mismatches between their asset and liability positions, which often reflect speculative behaviour. By contrast, a hike in short-term interest rates also hurts domestic investors with maturity mismatches. Moreover, traded goods sectors are hurt more by high interest rates than by devaluations; this makes it more difficult to undertake a payments adjustment based on export expansion rather than on import compression.

3. Reserve policy

It is sometimes suggested that debtor countries should maintain adequate reserves to meet their short-term obligations in order to avoid currency turmoil in the face of a massive withdrawal of foreign loans and investment. Proponents of such a policy point to the experience of economies with large reserves (e.g. China; Taiwan Province of China; and Hong Kong, China), arguing in this respect that large reserves would also deter speculative attacks on the currency. However, the consequences of building up a large stock of reserves by borrowing are quite different from when the reserves are accumulated through trade surpluses. One way of building up such reserves is to sterilize a large proportion of capital inflows, i.e. to purchase the proceeds through the issue of domestic debt instruments. However, there is a certain degree of circularity in such a strategy. In effect, it means that a country should borrow short only when it does not use the proceeds of such loans to finance investment and imports. Such a strategy can be very costly to the economy since the return on foreign reserves generally falls short of the cost of external borrowing. Moreover, the cost of sterilizing private borrowing falls entirely on the public sector. Indeed, public sector losses will exceed the foreign exchange cost of carrying such reserves since real domestic interest rates on government debt exceed by a large margin the rates earned on reserves.

There will thus be a net transfer from the public to the private sector in addition to the net cost incurred by the economy as a whole. Indeed, experience shows that such a strategy can give rise to large fiscal deficits or central bank losses (quasi-fiscal deficits). A variant of this proposal is for the public sector to fully cover the external short-term liabilities of the private sector by borrowing long and investing short abroad. However, not all governments have access to long-term foreign borrowing. More important, the cost of such an operation could be prohibitive, particularly when the international long rates exceed short rates by a large margin and the risk premium on long-term sovereign debt is high.

A similar strategy is to maintain credit lines with foreign private banks and to use them when faced with an attack, which is tantamount to arranging a private lender-of-last-resort facility. Again, however, this will work only if the amounts are small. Moreover, the costs involved can be very large and there is no guarantee that the banks will keep to such arrangements when there is a massive withdrawal of foreign lending. A further problem is that vulnerability to withdrawal of funds is not confined to short-term liabilities. In this respect, what matters is liquidity rather than maturity of liabilities. Massive withdrawal of funds from equity and/or bond markets can cause similar difficulties in the currency market, even though declines in the prices of such assets tend to alleviate the pressures on the exchange rate. When stock and bond markets are sizeable and foreign presence is significant, bearish moods in such markets can easily translate into a flight from the national currency, necessitating large-scale interventions to stabilize the exchange rate. The cost of maintaining reserves large enough to meet this eventuality would be prohibitive.

4. Bailouts and international lender-of-last-resort facilities

Provision of liquidity from an international lender of last resort to stabilize currency markets has not been the policy response to currency crises in developing countries. Rather, assistance coordinated by the IMF has usually come after the collapse of the currency, in the form of bailout operations designed to meet the demands of creditors and to prevent default.

Such operations, however, pose problems for a number of reasons. First, they protect creditors from bearing the full costs of poor lending decisions, thereby putting the burden entirely on debtors. Second, they consequently tend to create moral hazard for international lenders, encouraging imprudent lending practices. Not only do they reduce the concern of creditors about liquidity risk, but often, by securing ex post public guarantees for private debt, they also tend to reduce the perceived default risk. Third, the international financing required has involved increasingly large amounts that have been difficult to raise. However, there are also serious impediments to creating a genuine international lender of last resort to avoid such problems. The effective functioning of such a facility depends on two conditions: there should be reasonably well defined rules and conditions that the borrower must satisfy, and the lender of last resort should have the discretion to create liquidity in fulfilling its function. Amongst existing multilateral arrangements for the provision of external financing the facilities available within the EU perhaps come closest to meeting these two conditions. These facilities provide short-term support both for EU member countries participating in the EMS exchange rate mechanism (ERM) and for non-participants in this mechanism, as well as other longer-term financing.

Access to short-term external financing is guaranteed to an ERM participant for intervention in exchange markets to keep its currency within prescribed fluctuation limits; borrowing under this facility becomes subject to additional conditions only if the maturity of the loan is extended beyond an initial period, which may be as long as approximately two and a half months.

Other short-term external financing is available to EU member countries up to certain limits after agreement has been obtained in accordance with established procedures.

Medium-term external financing is also available up to specified limits subject to similar multilateral agreement concerning the borrowing country’s need (after taking account of policies it undertakes to overcome its difficulties).

Strictly speaking, the IMF does not satisfy either of the above conditions to qualify as a lender of last resort. Indeed, that institution was not originally conceived to provide financing to its members encountering liquidity problems associated with capital flows. Article VI of its Articles of Agreement specifically precludes lending to finance persistent capital outflows. So far in its interventions for this purpose the IMF has relied on the provision of funds by its major shareholders.

A proposal was made on the eve of the Mexican crisis to create a new “short-term financing facility” (STFF) for this purpose.1 The facility was to be used by countries with close integration with international capital markets, including industrial countries and emerging markets. However, a number of difficult issues were raised by this proposal. The first issue concerns the conditions under which financing should be made available to countries facing liquidity problems. In the STFF proposal two kinds of drawing were envisaged: an automatic right to draw (analogous to the gold tranche) and a drawing subject to the approval of the Executive Board. Such a two-tier approach was thought to strike a balance between speed and risk. While automatic access would ensure a timely response to market pressures, it could also create a greater risk to the IMF and give rise to moral hazard for the borrower. By contrast, conditional withdrawal would reduce the risk to the Fund, but negotiations and approval could cause long delays and uncertainties which might in turn further undermine market confidence. The Fund paper suggested that for conditional withdrawal the request should be made at the time of the article IV consultations, and that the facility should not be made available to finance unsustainable current account deficits. In that sense, the Fund’s agreement to access would indicate a seal of approval of the country’s underlying external payments position.

In principle, access to a lender-of-last-resort facility should depend on the fulfilment of specified conditions in advance, rather than on a commitment to undertake certain actions after the crisis occurs. Such conditions may relate not only to the sustainability of exchange rates and current account positions, but also to factors that affect financial stability, such as the size and maturity structure of external debt and effective prudential regulations. The lender of last resort should have the authority and capacity to monitor the extent to which these conditions are fulfilled and to determine eligibility. There are, however, serious difficulties in implementing such a procedure. First of all, it may require considerable extension of article IV consultations regarding matters related to the capital account, and it is not clear whether this would necessitate amending the Articles of Agreement in order to give the Fund jurisdiction over such matters. Second, it may not be easy to agree on what constitutes the relevant set of policies and institutions. For instance, there has been considerable controversy over the policies demanded by the Fund as part of its rescue package for the Republic of Korea; indeed, some of the conditions imposed have been regarded as interfering “unnecessarily with the proper jurisdiction of sovereign government” rather than as technical matters for dealing with the payments problem.2

Moreover, the adequacy of national policies for exchange rate sustainability and financial stability when a country is integrated with international capital markets involves matters of interpretation going beyond those traditionally raised under IMF surveillance. Thus, considerable differences may emerge between the Fund and the member concerned during the article IV consultations over the fulfilment of eligibility conditions. Finally, while experience strongly suggests that financial crises can occur despite effective prudential regulations and sustainable macroeconomic positions, there is a tendency to assume that they are caused primarily by poor policies and the weakness of the institutional machinery.

For instance, a number of flaws in policies and institutions in East Asia came to light only after the crisis, although the policies and performance of these countries had been highly praised earlier. If the simple fact that a crisis has occurred is taken as prima facie evidence of poor policies and institutions, it may never be possible for developing countries to be eligible for recourse to a lender-of-last-resort facility without additional and as yet unspecified commitments to undertake certain actions.

A second set of problems relates to the level of access and the adequacy of funds. In the 1994 proposal these were envisaged to be commensurate with the size of reserve losses that countries could sustain, but the facility was not envisaged to finance shocks fully. Three hundred per cent of quota was considered as a possible upper limit. Such an amount would indeed be quite modest in relation to possible needs arising from sudden outflows, but it could absorb an important proportion of Fund resources. For instance, in the 1994-1995 Mexican crisis, the initial offer of IMF funding of $7.8 billion was three times the country’s quota. Even though this was subsequently raised to $17.8 billion, representing no more than one third of the total rescue package, this amount was widely regarded as unusually high and risky for the Fund. The recent intervention by the IMF in East Asia was again far above the quotas of the countries concerned, and was funded through special arrangements under emergency financing procedures established after the Mexican crisis on the assumption that “use of these emergency procedures [was] expected to be rare, and the IMF’s role [would] remain catalytic”.3 With the deepening of the crisis in East Asia, the IMF Executive Board approved in December 1997 the Supplemental Reserve Facility to provide financing to countries experiencing exceptional payments difficulties under a highly conditional Stand-By or Extended Arrangement for up to one year.4 Ideally, the SDR could play a key role in creating a lender-of-last-resort facility, so that it would become a true fiduciary asset and enhance its role and share in global reserves.

Indeed, after the outbreak of the Mexican crisis, in his statement to the Copenhagen Social Summit in March 1995, the Managing Director of the IMF suggested that an effective cure depended on “convincing our members to maintain, at the IMF level, the appropriate level of resources to be able to stem similar crises if they were to occur”, adding that this would imply a decision, inter alia, for “further work on the role the SDR could play in putting in place a last-resort financial safety net for the world”.5 Such a step would require an amendment of the Articles of Agreement and could face opposition from some major industrial countries. Since it is insisted that the IMF should remain largely a quota-based institution, funding through bond issues by that institution is also ruled out. This leaves the Fund’s normal resources, together with its borrowing facilities, as the only potential sources of funding. However, they alone would not provide financing on the scale made available by the IMF and other sources during the recent Mexican and East Asian crises.

Bailout operations by the IMF will thus continue to rely on ad hoc arrangements with major industrial countries. In view of the increased public concern over burden-sharing and moral hazard, and the constantly growing size and risk of such operations, there is no guarantee that the required funds will always be forthcoming in the future. Critics point increasingly to the non-transparent nature of such operations. Moreover, there is also concern about the risk of default to countries providing the funding for bailouts. Although Mexico was able to repay quickly its debt to the United States from the bailout operation by refinancing it in international capital markets, there is no guarantee that other distressed borrowers will be equally capable. Questions are thus raised whether such a transformation of external debt could not be achieved without going through IMF bailout operations and creating risks for taxpayers in creditor countries.6 In this respect, the application of insolvency principles, discussed in the next sub-section, may provide an effective alternative.

5. Insolvency procedures and international debt crises

(a) Insolvency principles

Commenting on the debt crisis of the 1980s more than a decade ago, the UNCTAD secretariat expressed the main dilemma facing the debtor countries as follows: The lack of a well-articulated, impartial framework for resolving international debt problems creates a considerable danger, which has in part already materialized, that international debtors will suffer the worst of both possible worlds: they may experience (and many are experiencing) the financial and economic stigma of being judged de facto bankrupt, with all the consequences that this entails as regards creditworthiness and future access to financing. At the same time, they are largely without the benefits of receiving the financial relief and financial reorganization that would accompany a de jure bankruptcy handled in a manner similar to chapter 11 of the United States Bankruptcy Code.7

Bankruptcy procedures are especially relevant to international debt crises resulting from liquidity problems because they are designed to address financial restructuring rather than liquidation. In the United States Bankruptcy Code they are based on the premise that the value of the firm as a going concern exceeds the value of its assets in the event of liquidation. No receiver or trustee is appointed to manage the debtors’ business, and debtors are usually left in possession of their property, with all the powers of a trustee.8

The aim of these procedures is to facilitate orderly workouts in three stages. At the outset such procedures allow for an automatic standstill on debt servicing in order to provide the debtors-in-possession with a breathing space from their creditors, who are not allowed to pursue lawsuits or enforce the payment of debts. The automatic-stay provision is based on the recognition that a “grab race” for assets by the creditors is detrimental to the debtor as well as to the creditors as a group. It allows the debtor the opportunity to formulate a reorganization plan and ensures that creditors are treated equally. The filing of a bankruptcy petition also fixes all claims against the debtor whereby claims for future interest on pre-petition indebtedness cease to accrue as of the petition date and may not be asserted against the debtor. In the second stage, between the filing of the petition and the exit from bankruptcy through the reorganization of the debtor’s affairs, the Code provides the debtor with access to working capital needed to carry out its operations. This it does by granting a seniority status to debt contracted after the filing of the petition. This debtor-in-possession financing does not depend on the permission of existing creditors, and is approved whenever it is judged that continued operation of the firm will enhance its value. The final stage is the reorganization of assets and liabilities of the debtor and its operations. The Code discourages holdouts by a certain class of creditors and accelerates the process towards a rapid resolution. The plan does not require unanimous support by the creditors (acceptance by 50 per cent in number and two thirds in amount of the claims is sufficient), and the debtor can obtain court approval of the reorganization plans under the “cramdown” provisions. These procedures are used not only for private debt. Chapter 9 of the Code deals with public debtors (municipalities) and applies the same principles as chapter 11. The recent successful workout of the Orange County debt was under chapter 9.

Similar arrangements exist in most other industrial countries. Although they do not always go as far as the United States in safeguarding the interests and the needs of the debtor, they do not apply a rigid and legalistic approach designed to satisfy the interest of the creditors at any cost.9

(b) International application

International private debtors may enjoy in-solvency protection subject to provisions in their contracts with the creditors even though the application of such provisions involves a number of complex legal questions such as the determination of the relevant law and forum, and enforcement.10 However, under debt runs such protection does not offer much relief to the country concerned even if the bulk of the external debt is owed by private banks and firms. If there are numerous debtors, it is very difficult to simultaneously initiate insolvency procedures in respect of them all so as to halt the “grab race” by the creditors. Moreover, as in East Asia, most private debtors may indeed be solvent and hence unwilling to file a petition for insolvency, but the country may not have the reserves to meet the demand for foreign exchange.11

However, as noted above, debt runs can make such debtors insolvent, and this danger is greater when external debt is owed by the private sector and exchange controls have been dismantled. With sovereign debt a “grab race” on the currency is limited, and exchange controls can help contain the flight of residents from domestic assets. The task falls on the government to take action to secure the kind of protection provided under the insolvency procedures, particularly debt standstill. However, current judicial practices and government policies in the major industrial countries do not allow debtor governments to benefit from debt standstill provisions in the case of external obligations (see box 4). In this context, a question arises as to whether the relevant provisions of the Articles of Agreement of the IMF can provide a statutory basis for action by debtor governments through exchange controls. The most relevant provisions are in article VIII, section 2(b):

    Exchange contracts which involve the currency of any member and which are contrary to       the exchange control regulations of that member maintained or imposed consistently with     this Agreement shall be unenforceable in the territories of any member. In addition,               members may, by mutual accord, co-operate in measures for the purpose of making the         exchange control regulations of either member more effective, provided that such                 measures and regulations are consistent with this Agreement.

This article has given rise to a number of different and conflicting interpretations.12 On one view, it allows governments to take unilateral action for standstill on debt payments, since under article VI, section 3, members are free to impose capital controls without IMF approval. The courts of the member countries cannot refuse to recognize such controls if they are consistent with the Articles of Agreement. It therefore follows that any suspension of debt servicing introduced in the context of exchange controls approved by the IMF would render debt contracts unenforceable in the courts of any IMF member.

On another view, this was not the original intent of the clause. Indeed, there are considerable ambiguities regarding concepts such as exchange controls and exchange contracts, allowing different interpretations. While the courts in France appear to favour a broader interpretation, those in the United States and the United Kingdom tend to define exchange contracts to include only contracts having as an immediate objective the exchange of international means of payments, rather than any contract that affects a country’s foreign exchange reserves. Consequently, on this interpretation, international loan agreements are not “exchange contracts”, and hence do not fall within the ambit of the article.13

In practice, governments are reluctant to resort to unilateral suspension of debt servicing and exchange controls even in the extreme event of financial panic. The reasons put forward by the IMF are that: Because there exists no well-defined and accepted legal process that is applicable in such cases, the process of debt resolution by involuntary restructuring is necessarily ad hoc with an uncertain outcome. Bond holders may try to seek redress, on an individual or coordinated basis, by attempting to seize the assets of the borrowers or by threatening to disrupt their trade and pay-ments systems ... “Free riders” may also undermine any negotiated solutions by trying to attempt to enforce their individual claims. In addition, involuntary debt restructuring will damage creditworthiness provisions allowing countries to take emergency actions. There would also be a need to combine debt standstills with debtor-in-possession financing in order to replenish the reserves of the debtor country and provide working capital. This would mean IMF  “lending into arrears”. The funds required for such emergency lending would be much less than the scale of bailout operations.

Moreover, the Fund could also help arrange emergency lending from private capital markets with seniority status. As regards sovereign debt to private creditors, reorganization could be carried out through negotiations with the creditors, and the IMF could be expected to continue to play an important role

Box 4

COURT RULINGS IN THE UNITED STATES ON THE APPLICATION OF CHAPTER 11 OF THE BANKRUPTCY CODE TO INTERNATIONAL DEBT

In 1982 payments difficulties prompted the Costa Rican Government to suspend debt servicing by three state-owned banks. Initially, the case opened by the creditors in the District Court in New York in 1983 was dismissed on the grounds that the action by the Costa Rican Government constituted an act of State – i.e. that it was “governmental” (as opposed to commercial) – both in nature and in purpose.1

The Court of Appeals upheld this ruling, though on different grounds; namely, that the action was consistent with the law and policy of the United States, with reference in particular to chapter 11 of the Bankruptcy Code. It ruled that Costa Rica’s action was “not a repudiation of the debt but rather was merely a deferral of payments while it attempted in good faith to renegotiate its obligations”, and was “in entire harmony with the spirit of bankruptcy laws, the binding force of which, upon those who are subject to the jurisdiction, is recognized by all civilized nations”, prompting such remarks in the financial press as that New York was “unsafe for loan agreements”.2

However, after rehearing the case the same court reversed itself in 1984, when it was “bluntly told by the US Government that the court’s earlier decision had incorrectly interpreted US policy as supporting the enforcement of the Costa Rican decrees”.3 The court ruled that the Justice Department’s brief clearly established that the Government’s policy was to support “the debt resolution procedure that operates through the auspices of the International Monetary Fund”, and that “Costa Rica’s attempted unilateral restructuring of private obligations ... was inconsistent with this system of international cooperation and negotiation, and thus inconsistent with United States policy”.

This final ruling in effect established that for foreign governments to enjoy insolvency protection in United States courts, their actions should be in conformity not only with United States law, but also with the policy of that country with respect to international debt restructuring. Indeed, this ruling gave rise to such remarks as “existing US legal doctrines ... could not easily be stretched into creating what amounted to a code of international bankruptcy practice when there was no statutory or other basis for such a result ... Absent some guidelines as to what constituted a good-faith rene-gotiation of sovereign debt, the suspension of creditor legal remedies might empower a foreign sovereign to act unilaterally and arbitrarily in matters directly affecting US banks and indirectly affecting the stability of the US banking system”.4

1 For an extensive discussion of this case see L.C. Buchheit, “Act of State and comity: Recent develop-ments”, in Sassoon and Bradlow, op. cit.; and TDR 1986, box 6. The quotations below from court rul-ings are taken from these two sources.

2 Financial Times, 24 May 1984.

3 Buchheit, op. cit., p. 103.

4 Ibid., p. 102.

by providing a forum for bringing all creditors into negotiation with the debtor government. Special arrangements might be needed for bonds, which are often more difficult to restructure. For private debt, negotiations could be launched with private creditors immediately after the imposition of debt standstill. Judicial procedures might also be applied to individual debtors according to the law and the forum governing the contracts at issue. Their application would be greatly facilitated by the existence of proper bankruptcy procedures in debtor countries.

In past episodes of debt crisis, negotiated settlements often resulted in the socialization of private debt when the governments of developing countries were forced to assume loan losses.20 This leads not only to a regressive redistribution of wealth in the debtor country, but also to moral hazard for both private debtors and creditors. The introduction of automatic stay, together with debtor-in-possession financing, could help relieve such pressures. Certainly, a number of issues would need to be addressed in establishing procedures that would protect the debtors from the consequences of self-fulfilling debt runs and allow them to carry out their operations without creating moral hazard and opportunities for abuse of exchange controls.

The recent East Asian crisis has shown once more that there is a need to safeguard debtor countries from the overreaction of financial markets, “in entire harmony with the spirits of bankruptcy laws, the binding force of which, upon those who are subject to the jurisdiction, is recognized by all civilized nations” (see also box 4).21 Adoption of the principle of automatic stay for international creditors and investors is certainly one of the most helpful steps which might be taken in that direction.

C. Prevention of financial crises

1. Global surveillance

Global surveillance has not been successful in preventing international financial crises. In part this failure reflects belated, and so far only partial, adaptation of existing procedures to the problems posed by large autonomous private capital flows. But perhaps more fundamentally it is due to the unbalanced nature of these procedures, which give too little recognition to the disproportionately large global impact of monetary policies in a small minority of OECD countries.

In view of the growing size and integration of financial markets, every major financial crisis now has global ramifications. Consequently, preventing a crisis is a concern not only for the country immediately involved, but also for other countries which are closely integrated into the global trading and financial system and which can be affected in a number of ways. As already noted, contagion can occur through various channels, including those resulting from liquidity and credit interdependencies among major financial institutions and markets in the world, from expectations of competitive exchange rate adjustments, and from changes in perceptions regarding risks associated with a certain class of markets.

Global surveillance of national policies is thus called for, with a view to ensuring stability and sustainability of exchange rates and external payments positions. However, financial crises are not always home-grown. As noted in the preceding chapter, international financial crises are typically connected with major shifts in macroeconomic indicators external to the countries where the crises first manifest themselves. This is true of the debt crisis of the 1980s and of the Asian financial crisis. The origins of the former are to be found in shifts in the macroeconomic policies of major OECD countries in response to inflationary pressures. The inconsistency between contractionary monetary policy and expansionary fiscal policy in the United States, combined with the overall deflationary stance of macroeconomic policies in other major industrial countries, resulted in a sharp rise in interest rates in the United States and the appreciation of the dollar, both of which played a crucial role in the developing country debt crisis. 22

The Asian crisis was influenced by similar factors. The large capital flows before the crisis to East Asian countries (which over-financed their current account deficits) began in the early 1990s to a significant extent in response to an easing of monetary conditions in major OECD countries, on the one hand, and high interest rates and relatively stable exchange rates in the Asian countries, on the other. Again, as discussed above, the reversal of these flows was closely connected with the swings in exchange rates and monetary conditions in the United States and Japan.

Various other recent examples of external influences on capital movements and currency markets come easily to mind, such as the fluctuations in private external financing for Latin American countries, an important determinant of which has been shifts in monetary conditions in the United States.

Indeed, econometric research indicates that internal and external factors were about equally important in the surge in capital flows to Latin America during the early 1990s.23 The objectives of IMF surveillance, as formally stated, are limited to exchange rate policies, focusing primarily on the sustainability of exchange rates and external payments positions and on the appropriateness of the associated economic policies of individual countries. However, its scope has tended to broaden over time. For instance, the guidelines established in 1977 for surveillance made an explicit reference to the obligations of a member to avoid manipulating exchange rates or the international monetary system to gain an unfair competitive advantage over other members.24 Again, in the 1980s the major members of the Fund came to favour a broader interpretation and recognized that “to be effective surveillance over exchange rates must concern itself with the assessment of all the policies that affect trade, capital movements, external adjustment, and the effective functioning of the international monetary system”.25

However, the modalities of IMF surveillance do not include ways of responding to and dealing with unidirectional impulses resulting from changes in the monetary and exchange rate policies of the United States and a few other OECD countries which exert a strong influence on international competitiveness and capital movements. In the absence of incentives and enforcement procedures linked to the process of peer review under IMF surveillance, countries elsewhere in the world economy lack mechanisms under the existing system of global economic governance for redress or dispute settlement regarding these impulses. In this respect, governance in the area of global finance lags behind that for international trade, where such mechanisms are part of the WTO regime.26

The need for strengthening IMF surveillance in response to conditions produced by greater global financial integration and recurrent financial crises has been recognized by the Interim Committee. For example, at its meeting in April 1998, the Committee agreed that the Fund “should intensify its surveillance of financial sector issues and capital flows, giving particular attention to policy interdependence and risks of contagion, and ensure that it is fully aware of market views and perspectives”. It made special reference to the risks posed by abrupt reversals of capital flows and to the need for efforts by the Fund and the World Bank to help member countries to strengthen their financial sectors, and for an improved communication process between the IMF and member countries, requesting the Executive Board to develop a “tiered response” involving increasingly stern warnings to countries believed to be following policies seriously off course.27

However, despite the reference to interdependence, it is not evident that these proposals extend to weaknesses arising from the lack of balance in existing procedures. The focus of attention continues to be on the impact of domestic policies in generating financial fragility and crisis rather than on external influences produced by monetary and exchange rate policies of the major industrial countries. Moreover, even within the current limits of surveillance, the IMF has a mixed record of diagnosis of build-up of financial fragility and external vulnerability.

Thus, various questions emerging from recent experience can be posed regarding the direction which should be taken by more concrete guidelines for article IV surveillance as a follow-up to the Interim Committee’s Communiqué:

• In the context of such surveillance can confidence be placed in the improvement of capacities to identify factors likely to cause such crises in a world of increasingly liberalized capital flows?28

• In the absence of such capacities might it not be more prudent to place greater reliance as a matter of course on capital controls and other measures at the national level directed at external assets and liabilities (such as those discussed in subsection 5 below)?

• If the latter approach is adopted, should new guidelines for IMF surveillance not specify circumstances in which the Fund should actually recommend the imposition or strengthening of capital controls?

• How far should IMF surveillance be extended to cover subjects such as financial regulation and standards for financial reporting and accounting?

• What should be the relations between IMF policy surveillance and the consultation and collaboration procedures of regional bodies, which in future are likely to include not only existing agreements such as those of the EU but also new ones among developing countries, an example of which is described in subsection 7 below?

• Finally, how can more effective implementation of the policy recommendations put forward as part of surveillance be achieved?

These are clearly delicate questions involving not only formulation of an appropriate framework and development of technical competence, but also powers and responsibilities in areas where multilateral bodies other than the IMF already exist.

2. Information and transparency

The Asian financial crisis has accelerated initiatives to improve the timeliness and quality of information concerning key macroeconomic variables as well as the financial reporting of banks and non-financial firms. The first of these subjects was accorded by the IMF’s Interim Committee in April 1998 an essential position in its proposals for strengthening the architecture of the international monetary system. The central element of the IMF’s own initiatives in this area is the Special Data Dissemination Standard (SDDS), established in April 1996 to guide member countries in the public dissemination of economic and financial information in the context of seeking or maintaining access to international financial markets. At the time it was hoped that the new, more stringent rules associated with the SDDS would serve as an early warning system that would help to prevent future financial crises. However, in the event the rules did not make such a contribution in the case of the Asian crisis.

Countries subscribing to the SDDS commit themselves to certain standards regarding data dissemination in four areas: coverage, periodicity and timeliness; public access; integrity of the data; and quality of the data. The subjects to be covered comprise national accounts, production, conditions in the labour market, prices, the determinants and principal features of the government’s fiscal balance and debt position, the accounts of the central bank and of the financial sector (which include monetary aggregates and credit), interest rates and stock prices, the balance of payments and international reserves, international investment, and spot and forward exchange rates. In April 1998 the Interim Committee proposed a broadening of the SDDS, clearly inspired in part by what it considered to be the role of informational deficiencies in the Asian crisis, so that the system would also cover additional financial data such as net reserves (after allowance for central banks’ liabilities under forward or derivative transactions), the debt (especially the short-term debt) of economic agents, and other indicators bearing on the stability of the financial sector.

While initiatives such as the SDDS are capable of furnishing additional, more timely and reliable information to investors and policymakers, emphasis on inadequate information as the major reason for failure to forecast the Asian crisis appears misplaced or exaggerated. Data were generally available concerning key variables in the countries concerned, such as their balance of payments, both their short- and longer-term external debt and net external assets (in particular in the periodic reports of the Bank for International Settlements [BIS] concerning international bank lending), their capital inflows, the exposure of banks and other financial firms to different sectors or categories of economic activity, the problems of the property sector, and (in the Republic of Korea) the precarious balance sheets and low recent profitability of many non-financial firms. The crisis has pointed to weaknesses in available information pertinent to governments’ ability to manage capital flows and external debt: for example, in some cases existing data systems provided inadequate indications about the scale and nature of the exposure of Asian banks to other countries in the region, and about the country of ultimate risk in international inter-bank lending involving such banks. But these weaknesses were not an essential part of the failure to forecast the crisis. Rather, what was missing was adequate evaluation of the implications of available information for countries’ ability to continue to obtain funding from the international financial markets.

Furthermore, it should be noted that quicker access to macroeconomic and financial information may also be a source of instability. General dissemination of certain up-to-date data (including some bearing on unfavourable developments affecting countries’ external assets and liabilities) is capable actually of increasing the volatility of capital flows. If, in consequence, a decision were to be taken to restrict the availability of such information in the interest of avoiding volatility, a difficult and perhaps invidious choice might have to be made regarding the parties to whom disclosure would be made.

The Asian crisis has also focused special attention on standards of accounting and financial reporting. Efforts in these areas were already under way before the crisis as part of the upgrading of financial markets not only in Asia but also in other regions. But the crisis has provided additional impetus to the process, particularly as part of the strengthening of bank regulation and supervision, of which adequate accounting and reporting are integral components.

3. Domestic financial regulation and supervision

Weak credit evaluation and speculative lending, as well as failure to control currency risk among banks and other financial firms, contributed both to the outbreak of the Asian financial crisis and to its amplitude. The growth of doubtful and non-performing loans, accompanied in some countries by widespread insolvencies in the financial sector, will create major problems for government budgets and be a drag on the availability of lending for a considerable time to come. There is general agreement that regulatory reform is an essential part of the strengthening and restructuring of the financial sectors of most countries affected by the crisis.

However, such reform is not a fail-safe way of preventing financial crises, though it can reduce their likelihood and help to contain their effects. In recent years there has been widespread reform and strengthening of financial regulation at the national level, accompanied by a proliferation of international initiatives to raise regulatory standards and to improve cooperation among supervisors. These processes have been largely driven by concerns raised in relation to financial liberalization and global financial integration. On the one hand, the diversification of their services and the increased competition that are associated with liberalization have exposed financial firms to new levels of risk, which have necessitated overhaul not only of financial regulation but also of firms’ systems of internal control. On the other hand, global financial integration has brought in its train much greater exposure among countries to each other’s financial and macroeconomic conditions and increased possibilities for the cross-border transmission of destabilizing influences.

Such exposure has been dramatized by various events since the beginning of the 1970s. For example, the insolvencies of two international banks in 1974 (Bankhaus Herstatt and Franklin National Bank) pointed to the danger of cross-border spillover effects from the failures of financial firms, and provided the initial impetus for international initiatives regarding financial regulation and supervision. Subsequent efforts to strengthen standards and international cooperation in this area have also been partly a response to, and their substance has been influenced by, such events as the developing-country debt crisis of the 1980s and failures of individual financial firms such as Banco Ambrosiano (1982), Bank of Credit and Commerce International (1991) and Barings (1995), each of which in their different ways exposed weaknesses in banking regulation and in cross-border cooperation among banking supervisors.

The main vehicles for international initiatives regarding financial regulation and supervision have been the Basle Committee on Banking Supervision and other bodies with close links to the BIS, other groups of financial supervisors, associations of exchanges, and organizations concerned with accounting standards. The initiatives of the Basle Committee have included the adoption of principles designed to ensure that no international bank escapes adequate supervision and the prescription of levels of capital commensurate with the risks that banks run: agreements under the latter heading were reached concerning credit risks in 1988 and concerning market risks in 1996. 29

The Basle Committee has also devoted considerable attention to the improvement of banks’ systems of internal control and, together with the International Organization of Securities Commissions (IOSCO), has developed guidelines for the disclosure by banks and securities firms of their trading and derivatives activities.

Furthermore, the Committee on Payment and Settlement Systems has made several proposals designed to reduce the risks due to financial firms’ exposure to the possibility of non-payment by their counterparties in international transactions. Membership of the various bodies linked to the BIS which are concerned with different aspects of banking supervision is limited to a small group of countries. However, efforts have been made to promote Basle standards through contacts with other groups of banking supervisors, and special attention has recently been paid to regulation and supervision of emerging financial markets.

One important outcome of these efforts is the recent release of the statement entitled Core Principles for Effective Banking Supervision, the drafting of which involved extensive consultations with parties in developing countries. The coverage of these principles includes the permissible activities of banks, licensing criteria, the vetting of banks’ controlling interests, capital and risk management, guidelines on lending to related companies and individuals, “know your customer” procedures intended to prevent the criminal use of banks, the information and methods required for effective supervision, the powers of supervisors, and consolidated supervision of international banks.

However, the introduction of improved standards in this field takes considerable time and the full benefits of international initiatives so far are unlikely to be experienced soon. Moreover, the coverage of international regulatory and supervisory cooperation is incomplete, thereby restricting its effectiveness: offshore financial centres and several increasingly important actors in international capital flows such as investment funds are still only partly included or not included at all. And the networks of cooperation and information exchange among financial supervisors required for the effective implementation of international agreements are still being developed.

Strengthened financial regulation can at best reduce the probability of financial crises. But the periodic incidence since the beginning of the 1980s of banking crises in industrial countries such as the United States, the United Kingdom and parts of Scandinavia exemplifies its inability to eliminate them. This inability stems partly from imperfections in the regulatory process itself, such as its tendency to lag behind changes in financial firms’ practices, and the difficulty of imposing regulatory transparency on such firms.30 Perhaps more fundamentally for the assessment of what regulation can and cannot do, no loan or other asset on a bank’s balance sheet should be classified generically as “good”.

However reasonable the original managerial decision to make a loan and however justified its initial classification as low-risk by banking supervisors, the loan is vulnerable to the possibility of an eventual deterioration in its status. Unfavourable changes in macroeconomic conditions (of external as well as domestic origin) are a factor frequently cited here. Arguably, the deterioration in the status of many loans is in fact an intrinsic feature of the boom-bust process often associated with financial crises. During this process risks take time to build up and to become widely evident. Indeed, for a time the quality of a loan can be validated or even enhanced by the effects on values of the very financing boom of which it is a constituent part. Thus, during booms the incentives for herd-like behaviour are not limited to speculative lenders. As a result, “risk-based competition propels the entire system towards excessive levels of indebtedness”,31 but excess capacity generated by the boom itself (widely exemplified during the savings and loan crisis in the United States discussed in the annex to chapter III) as well as the over-extended positions of financial firms do eventually make themselves felt, often in conjunction with rises in interest rates or downturns in economic activity.

The limits on the crisis-preventing potential of financial regulation are generally recognized by specialists in the field,32 so that its primary objectives are regarded as having more to do with reducing financial firms’ liquidity and solvency problems, protecting depositors, and preventing systemic risks due to contagion effects. This is not to deny that beneficial connections among regulation, incentives and internal controls are capable of enhancing the safety of financial firms.

Capital requirements appropriate to the credit and market risks run by these firms can improve the quality of their lending and their portfolio management, and lead to better pricing of the services which they supply. But as should be evident from the argument above, some of the risks faced by financial firms arise from circumstances over which they have little or no control. Against such risks robust financial regulation provides cushions both to individual firms and to the financial system. However, the protection thus afforded has repeatedly been shown to be only partial. If absence of complete protection from crises is characteristic even of financial sectors subject to relatively developed regimes of regulation and supervision, then unsurprisingly the same is a fortiori true of those subject to weaker regimes in the great majority of developing countries, whose vulnerability has been graphically illustrated by some of the examples discussed in chapter III above.

Moreover, the financial sectors in the latter countries frequently have to withstand more severe macroeconomic shocks than their counterparts in industrial countries.33 The severity and frequency of such shocks cannot always be reduced by macroeconomic policy. True enough, financial regulation and supervision can be improved until they attain the levels of prevailing best practice (though, as already suggested, that will generally take several years), but even then financial crises will remain possible.

4. Tighter control of international lending and portfolio investment

It could be argued that in a well functioning world economy no separate rules or restrictions would be required for international lending and portfolio investment beyond those associated with national prudential regulation of financial firms in both the source and recipient countries, and with the regulation of issuance and trading procedures for organized exchanges and other markets for financial assets and instruments. In such a world international capital flows would be closely related to payments and financing in international trade and investment, and driven by the economic fundamentals of firms and other recipients. But reality is otherwise. Much bank lending and portfolio investment in short-term debt securities is a response to interest rate arbitrage margins, which reflect the exigencies of monetary policy and can persist for long periods (frequently being eliminated by large eventual devaluations), or to differences among countries in the regulatory and tax treatment of external borrowing.

Furthermore, much international portfolio investment responds less to the long-term economic prospects of individual firms than to expectations of short-term capital gains and losses, of which a major determinant in many cases is the ebb and flow of international portfolio investment itself, because effects on equity prices reflect the disparity between the limited capitalization of many stock markets and the large size of funds at the disposal of investment institutions of major industrial countries.

The frequently tenuous connections between the forces influencing international lending and portfolio investment, on the one hand, and the fundamentals of economies and firms, on the other, have led to booms and busts in such financing which bear many similarities to the analogous fluctuations in bank lending at national level described under domestic financial regulation. Risks of loss in this system are unevenly distributed.

While during financial crises large losses may be incurred by external investors in stock markets, banks are often protected from losses on their international lending in various ways – by formal or informal protection against insolvency provided by the governments of borrowing countries to their domestic banks (often large recipients of funds borrowed from abroad),34 in some cases by explicit guarantees extended by governments on foreign deposits in their banks, and (as noted in chapter III) by the IMF bailouts.

Controls over international lending and portfolio investment can be imposed at source, by the recipient, or at both levels. Controls by the recipient belong under the heading of those over capital transactions and are discussed above. The motivation of proposals for control at source is the belief that not all the responsibility for, and the costs of, such controls should be borne by recipients, and that even when controls by recipients are in place, controls at source are capable of further reducing the probability of potentially destabilizing capital flows and financial crises.

Many ideas for controlling capital flows at source have been put forward in recent years, several of those directed at international bank lending having originally been a response to the developing-country debt crisis of the 1980s.35

Proposals for checking excessive international bank lending typically involve mechanisms for capping external indebtedness which could not be expected to emerge through the operation of competitive financial markets, such as cartel-like arrangements among banks to impose country credit ceilings or the acceptance by lenders of guidelines regarding a country’s sustainable level of borrowing set by a multilateral institution. Unsurprisingly, the Asian financial crisis has served as a stimulus for new proposals, and greater attention has been given to portfolio flows, which were not prominent in the debt crisis of the 1980s. The general conclusion of the discussion of proposals which follows is that the more ambitious ideas have features which are an obstacle to their adoption, while the contribution of ameliorative changes which seem more likely to be within reach is not such as to remove the need for capital controls imposed by recipient countries.

One proposal, which would lead as a by-product to better control of international lending, is for a radical strengthening of existing supervision of financial firms through the establishment of an international body – the Board of Overseers of Major International Institutions and Markets – with wide-ranging powers for the oversight and regulation of commercial banking, securities business and insurance (activities now bestrode in some cases by financial conglomerates). For this purpose, it would be “empowered to set mutually acceptable standards for all major institutions, to establish uniform trading, reporting and disclosure standards for open credit markets, and to monitor the performance of institutions and markets under its jurisdiction”.36

This proposal would address problems associated with the significant differences which still characterize national regimes for financial regulation. Such differences, as mentioned above, are one of the causes of capital flows in the form of international bank lending with often only a limited connection with real economic activity. But despite progress under recent international initiatives concerned with financial regulation and supervision towards objectives which include both the raising of standards and greater convergence among national regimes, the proposal seems utopian.

Another proposal, which focuses more narrowly on international bank lending, is for the establishment of an International Credit Insurance Corporation (ICIC) “as a sister institution to the IMF”.37 This body would guarantee international loans for a modest fee but would set a ceiling on the amount of borrowing by particular countries which it was willing to insure. The ceiling would be based on evaluation of data concerning all of its borrowings, which a country would be obliged to furnish to the ICIC. In consequence, countries would be able to borrow at low rates of interest up to their ceilings, but beyond them lenders would be much more cautious and money would be available only at rates of interest incorporating a substantial risk premium (or not at all). The likelihood of excessive credit expansion would thus be reduced, as would that of the financial crises which can follow in its wake.

This proposal poses questions concerning feasibility, the quality of credit rating and the powers which would be conferred on such an institution. Feasibility does not appear to be an insurmountable problem: more widespread application of well-established modalities for the provision of credit insurance would be involved. These modalities comprise establishment of risk criteria, decisions about particular borrowers’ creditworthiness as measured by these criteria which enable the setting of insurance premiums, and the administration of the insurance facilities. Such tasks are already carried out by export credit agencies (ECAs) of OECD countries. The major departure under the heading of an ICIC would be conferring on a single body the responsibilities regarding the risk criteria and creditworthiness indicators for the lending covered by its insurance facilities.38 Administration might actually be carried out by existing institutions such as ECAs. However, it is questionable whether generally acceptable indicators could be developed by an ICIC, given the current state of the art in this area. The record of credit rating agencies, for example, in assessing the creditworthiness of developing-country borrowers exemplifies the difficulties entailed by the evaluation required, although an ICIC would not have to depend, as the agencies sometimes do, largely on published information.39

Surveillance by the IMF, on the other hand, which involves evaluation similar in some respects to that carried out for credit rating, not only has been characterized on occasion by failures to identify weaknesses which could result in financial crises but also illustrates the political sensitivities associated with the disclosure of the evaluations of an official multilateral body bearing on countries’ creditworthiness (even when, as in this case, the disclosure is less directly linked to access to borrowing than it would be under the proposal for an ICIC). Nevertheless, better information concerning borrowers could be expected to lead in time to the possibility of improved evaluation of countries’ creditworthiness. 40

On the final question posed above concerning an ICIC’s powers, however, the prospect of international agreement to confer such powers on either a new or existing international agency seems remote.

In debate about ways to exercise better control over international bank lending attention has also focused on inter-bank flows. There is widespread agreement that improved monitoring of such flows could contribute to better decision-making by participants in financial markets and better management of the international financial system.41 But it is also believed that inter-bank lending is often associated with weaker credit assessment and with levels of bank capital which do not adequately reflect the credit risk involved. This has led Alan Greenspan, for example, to suggest that international inter-bank lending is an area requiring regulatory changes which would have the consequence of raising the cost to banks of such lending so that they better reflected its risks.42 Steps in this direction would represent a reversal of long-term tendencies to reduce the costs associated with international bank lending.43 One possible starting-point for action here would be the 1988 Basle Capital Accord, under which claims incorporated on banks outside the OECD area with a residual maturity of up to one year and all claims on banks incorporated in the OECD area are attributed a low (20 per cent) risk weight for the purpose of calculation of capital requirements. Yet the short-term exposure of international banks has been a major feature of recent external debt crises. Thus one way of causing tighter control to be exercised over banks’ international inter-bank exposure would be to increase the risk weight for such exposure in the setting of capital requirements.

Such a step should lead to better internal accounting by banks for the risks of this type of international lending but would none the less be a crudely calibrated method of dealing with problems caused by banks’ inter-bank exposure to countries with deteriorating creditworthiness. A more flexible approach might be based on existing country-specific procedures for monitoring banks’ external exposures as part of bank supervision. These procedures frequently provide supervisors with authority to determine the levels of reserves appropriate to banks’ external exposures, and could easily be used for the purpose of a more rigorous treatment of their inter-bank exposures to riskier countries (though, again, the effectiveness of the measure would depend on the quality of supervisors’ systems for credit rating).

As already noted, proposals for controlling capital inflows triggered by the Asian crisis have also covered forms other than bank lending. One such proposal is designed to increase the stability of mutual funds’ investments in securities issued by entities in developing countries by requiring the funds to hold liquid reserves amounting to some proportion of such securities.44 These reserves could then be tapped into in the event of large declines in the securities’ market value and would thus reduce the incentives to dump such securities for the purpose of obtaining the liquidity needed to meet redemptions. Also, the hope is expressed that although such reserve requirements would reduce the speculative returns to mutual funds’ investments in emerging financial markets, the resulting reduction in market risk would none the less increase their attractiveness to long-term investors.

Mandatory requirements for mutual funds to hold liquid reserves of this kind would represent a radical break with existing regulatory practice. Moreover, another feature of this proposal which would entail variations in these liquidity requirements in accordance with the creditworthiness of the countries in which mutual funds made their investments (making the requirements “risk-weighted” in the author’s words) would require the introduction of supervisory procedures for such funds analogous to those for banks (and under this heading an agreed system for rating creditworthiness – a task involving problems that have already been mentioned). Nevertheless, despite the problems it poses, this proposal represents an attempt to confront a source of potential volatility for an increasingly important category of financial flows to emerging financial markets. As such it may serve to stimulate discussion concerning other possible measures for this purpose.

An alternative approach, for example, might build on the exit fees which are a feature of some mutual funds.45 These fees can vary with the holding period of investments and might thus be expected to act as a disincentive to investors seeking short-term returns. If this approach were deemed appropriate, ways could be sought to generalize exit fees for emerging market mutual funds. It has the advantage that it would build on existing market practice. However, it would require agreement among the countries serving as major domiciles of emerging-market funds in order to prevent a flight of such funds to jurisdictions not imposing the exit fees, and an agreement of this kind would not be easy to achieve.

5. Capital controls and other measures for the management of external assets and liabilities

Management of a country’s external assets and liabilities is linked to many other dimensions of economic policy, such as good macroeconomic fundamentals, effective financial regulation and supervision, and even good corporate governance. However, experience shows that these are necessary but not sufficient conditions for the avoidance of financial crises. It also shows that a key role here is played by policies aimed specifically at external assets and liabilities – most importantly capital controls but also certain other measures designed to influence borrowing, lending and asset holding.

Controls on capital flows are imposed both as part of macroeconomic management and in pursuit of long-term policy objectives related to national economic development and autonomy. Controls imposed for macroeconomic reasons are typically closely related to other monetary and fiscal measures, their function being to reinforce such measures or to substitute for them when reliance on other policy instruments is thought likely to be ineffective or to cause undesirably large movements of key variables such as interest and exchange rates. Controls under the second heading have such aims as ensuring that the capital of a country’s residents is invested locally or that certain types of economic activity are reserved partly or wholly for residents.

The transactions which may be subject to capital controls are manifold and from some points of view avoid simple categorization. This applies, for example, to attempting to distinguish between short- and longer-term transactions: certain assets or instruments are clearly associated mostly or exclusively with short-term transactions, but others serve equally for short- or longer-term transactions so long as there exists a liquid secondary market for them.46 Moreover, legal and administrative distinctions embodied in national regimes of capital control do not necessarily correspond neatly to the conceptual classification used by economists (for example, with respect to direct as opposed to portfolio investment).

A by no means exhaustive list of the assets involved in capital transactions might include direct investments, long-term and short-term loans, cross-border holdings of real estate, domestically and internationally issued equity and debt (the latter ranging in maturity from money-market instruments to longer-term notes and bonds), collective investment securities (such as in shares in mutual funds), deposits with banks and other financial firms, guarantees and financial back-up facilities, life insurance contracts, various assets associated with personal capital movements (such as gifts, dowries and inheritances), blocked funds owned by non-residents, and derivative instruments. Because of institutional and regulatory features of financial systems and of effects on incentives, controls imposed on capital inflows or external liabilities may also influence capital outflows and external assets, and vice versa. Such influences can be seen, for example, in controls on portfolio equity and direct investment, where rules concerning the repatriation of capital clearly affect the incentives for inflows. Likewise, rules applying to the portfolios of foreign firms regarding such matters as the freedom to engage in outward as well as inward investment transactions will influence their willingness to establish a commercial presence through direct investment.

More generally, rules concerning the cross-border capital transactions open to financial firms in a country, as well as the matching of the currency denominations of their assets and liabilities, affect their willingness to depend on such inflows and their efforts to attract them. In view of the close connections between capital controls, on the one hand, and certain other instruments of policy, on the other, classification of measures under one or the other heading may be somewhat arbitrary. For example, special reserve requirements concerning banks’ liabilities to non-residents (a policy to which frequent reference is made in the following discussion) can reasonably be classified as either an instrument of monetary policy or a capital control.

Moreover, since such requirements affect the quality of financial firms’ balance sheets, they can equally be (and often are) classified as a “prudential” measure. Similar alternative classifications might also be attributed to restrictions on banks’ net assets or liabilities in foreign currencies. Traditionally, capital controls focused mainly on cross-border transactions of residents and non-residents. However, owing to deregulation and developments in banking technique making possible the supply of increasingly diversified services, accounts and transactions denominated in foreign currencies are now often available to residents, and they affect macroeconomic conditions, particularly exchange rates, in much the same way as cross-border financial transactions.

Many different measures are available for controlling capital movements, some with a broad incidence and others aimed at more narrowly defined transactions. Controls on inflows of FDI and portfolio equity investment (not always clearly distinguished in the regulations, as mentioned earlier) may take the form of licensing, ceilings on foreign equity participation in domestic firms, official permission for international equity issues, differential regulations applying to domestic and foreign firms regarding establishment and permissible operations, and various kinds of two-tier markets. For example, under a two-tier market investments in a country’s securities by non-residents may be limited to those purchased from other non-residents, and transfers of the country’s currency for such transactions may be limited to those made possible by purchases and sales of such securities by non-residents (a measure designed to reduce the likelihood of falls in securities prices being accompanied by depreciations in the currency or declines in reserves).

Some of these controls can also be imposed on capital inflows associated with debt securities, both bonds and other instruments. Such inflows can thus be subject to special taxes or be limited to transactions carried out through a two-tier market. Ceilings (possibly as low as zero) may apply to non-residents’ holdings of the debt issues of both firms and the government, or approval may have to be sought for the purchase of such issues by foreigners. Moreover, non-residents may be excluded from auctions for government bonds and government paper.

Various other controls are commonly used to restrict external borrowing from banks: the special reserve requirements concerning liabilities to non-residents already mentioned (to raise the costs and reduce the profits associated with on-lending of the capital inflows); forbidding banks to pay interest on the deposits of non-residents or even requiring negative interest rates (“commission”) on such deposits; taxing foreign borrowing to eliminate the arbitrage margin between domestic and foreign interest rates;47 and the imposition on both financial and non-financial firms of cash deposits at the central bank amounting to a certain proportion of their external borrowing (a measure pioneered by Germany in 1972 under the name of “Bardepot”). Controls on outward transactions for direct and portfolio equity investment can apply to residents as well as to non-residents. Restrictions on the latter can be directed at the repatriation of capital, for example, in the form of statutory periods before such repatriation is allowed or regulations providing for phasing of repatriation in accordance with the availability of foreign exchange or the need to maintain an orderly market for the country’s currency. Residents may be restricted as to their holdings of foreign stocks, either directly or through limitations on the permissible portfolios of the country’s investment funds.

Two-tier exchange rates may also be used to restrict residents’ foreign investment by requiring that capital transactions be undertaken through a market in which a less favourable rate than for current transactions generally holds. Some of these techniques are also used for purchases of debt securities issued abroad and for other forms of lending abroad. In the case of bank deposits by residents abroad, their availability can be restricted by law. As already mentioned, the question of controlling “dollarization”, in particular residents’ bank deposits denominated in foreign currencies, as well as banks’ lending to residents in foreign currencies, also falls under the heading of capital controls. Such loans and deposits can increase currency mismatching, which is a potential source of financial instability, and can precipitate and facilitate large shifts between currencies during financial crises, putting pressure on the exchange rate and resulting in widespread insolvencies among debtors.

The Asian crisis has drawn attention to issues deserving separate mention under the heading of measures to manage external assets and liabilities. During that crisis attention focused on the flight to foreign currencies which accompanied depreciations, but it was impossible to distinguish between flight which was due to speculation and flight due to belated attempts to cover foreign exchange exposures. However, as discussed in chapter III, there can be little doubt that mismatches between the currency denominations of the assets and liabilities of non-financial and financial firms made an important contribution, and were facilitated both by the ease of borrowing in foreign currencies and, in some cases, by the issuance to residents of bank deposits denominated in foreign currencies.48 To the extent that such liabilities are matched by assets (including loans) denominated in the same currencies, the foreign exchange risks are shifted to debtors, for whom such risks may be hedged by export earnings but frequently translate into credit risk. To the extent that the liabilities are not so matched, the resulting risks fall directly on the banks. The existence of such assets and liabilities, if sufficiently widespread, may thus pose a threat to the financial system.

Part of the solution to this problem can be found in strict enforcement of prudential rules regarding the matching of the currency denominations of financial firms’ assets and liabilities and measures increasing the costs of foreign borrowing through the imposition of taxes, special reserve requirements or cash deposits at the central bank. But, as already suggested, tighter restrictions might also be applied to “dollarization” itself. These might take the form of limiting bank lending and deposits in foreign currencies.

Non-interest-bearing reserve requirements could be imposed on deposits in bank accounts in foreign currencies, thus reducing or eliminating the interest paid on them and diminishing their attractiveness. The Asian crisis has also starkly demonstrated the risks that can result from failure to enforce adequate separation between the onshore and offshore activities of a country’s banks. A number of Asian countries have established off-shore centres, whose activities are subject to lighter regulation and certain tax privileges, with the aim inter alia of facilitating participation by their banks in regional or global banking business. One such centre is the Bangkok International Banking Facility (BIBF), established by Thailand in 1992. As discussed in chapter III, BIBF entities increasingly served as a conduit for interest rate arbitrage between the domestic and international financial markets, much of the financing made available through such arbitrage being used to finance speculation in stocks and property.

There is a contrast between this relatively uncontrolled use of offshore financing in Thailand and the functioning of Singapore’s offshore banking centre established in 1968. 49 Offshore banking in Singapore is conducted through Asian Currency Units (ACUs), which are integral parts of licensed banks. Indeed, except with respect to the segregation of its activities for accounting, fiscal and regulatory purposes, an ACU has no identity distinct from that of the bank in which it is located. The legal framework for ACUs is designed to facilitate their participation in regional banking business, while restricting the use of the Singapore dollar as an international currency and controlling ACUs’ involvement in domestic banking business. ACUs can accept deposits in Singapore dollars only above a certain amount and only from non-residents and from other banks and financial firms, and loans to domestic firms in Singapore dollars are also subject to a ceiling. Since 1983 ACUs have had to obtain official approval for the granting of credit facilities to non-residents of above 5 million Singapore dollars or to residents for use outside the country, a requirement that hinders short selling of the Singapore dollar in currency trading and the use by non-residents of such facilities for portfolio and property investment.

The success of these policies in maintaining the offshore character of Singapore’s ACUs can be illustrated from data on their assets and liabili-ties: 63 per cent of their liabilities in 1996 were from sources outside the country, and 42 per cent of their assets consisted of loans to banks outside the country. By contrast, there are estimates that as much as 95 per cent of the money raised by BIBF entities was lent domestically. The example of ACUs points both to the feasibility of a measure of insulation of offshore banking from the domestic market and to its benefits in terms of the contribution to financial stability. Use of capital controls has been a pervasive feature of the experience of the last few decades. In early post-war years capital controls for macro-economic reasons were generally imposed on outflows as part of policies for dealing with balance-of- payments difficulties and for avoiding, or reducing the size of, devaluations. Moreover, there was widespread use by both developed and developing countries of controls on capital inflows for the longer-term developmental or structural reasons mentioned above.

With the return to freer capital movements from the 1960s onwards, large capital inflows posed problems at various times for the governments of certain industrial countries such as Germany, the Netherlands and Switzerland, which responded with various controls such as those already discussed on purchases by non-residents of domestically issued debt securities and the bank deposits of non-residents. More recently, a number of developing countries experiencing similar macroeconomic problems as a result of large capital inflows have resorted to capital controls as part of their policy response.50

• In Malaysia, initial reliance on sterilization served to widen the difference between domestic and external interest rates, leading to an accelerated surge in short-term capital inflows. In January 1994 the Government responded with the imposition of the following capital controls (gradually removed from 1995 onwards): banks were subjected to a ceiling on their external liabilities not related to trade or investment; residents were prohibited from selling short-term monetary instruments to non-residents; commercial banks were required to deposit at no interest in the central bank monies in ringgit accounts owned by foreign banks; and commercial banks were also restricted in the outright forward and swap transactions they were permitted to engage in with foreigners.51

• In Chile, too, reliance on sterilization in the early 1990s in response to increasing capital inflows led to a rise in interest rates and an acceleration of the inflows. In consequence, the Government had recourse to various policies designed to slow short-term inflows and even to encourage certain categories of outflow, including the imposition of an un-remunerated reserve requirement on foreign borrowing of 20 per cent (subsequently raised to 30 per cent), to be deposited at the central bank for a year.

• In Colombia, sterilization was eventually abandoned as a response to capital inflows in the 1990s, and in its place was established a reserve requirement on loans (other than short-term trade-related credits) with maturities up to five years, which was to be maintained for the loan’s entire duration but whose magnitude was a decreasing function of its maturity.

• In Brazil, sterilization policies adopted to deal with increased capital inflows after the implementation of the country’s currency reform in mid-1994 were supplemented by an increase in the tax paid by Brazilian firms on bonds issued abroad, the imposition of a tax on foreigners’ investment in the stock market, and an increase in the tax on foreign purchases of domestic fixed-income investments.

• In the Czech Republic, a large increase in capital inflows in 1994-1995 led initially to a policy of sterilization, but this was followed by the imposition of a tax at a rate of 0.25 per cent on foreign exchange transactions with banks, as well as by limits on, and a requirement of official approval for, short-term borrowing abroad by banks and other firms.

Evaluation of these controls suggests that in most cases they were effective to varying degrees: inflows significantly declined as a percentage of GDP after their imposition except in Brazil and Colombia, with short-term inflows actually becoming negative for a time in Chile and Malaysia; and in Chile and Colombia there was shift away from short-term in the composition of inflows.

This recent recourse to capital controls, sometimes for significant periods of time, has taken place in a context marked by international initiatives aimed at restricting countries’ freedom to deploy such measures. A major target of these initiatives is the only global regime applying to such movements, that of the IMF.52 The primary original aim of this regime with respect to such movements was the promotion of world trade and economic activity through the elimination of restrictions on current transactions. Freedom of capital movements was not a principle of the IMF’s original Articles of Agreement. Indeed, under article VI, section 3, members are explicitly accorded the right to regulate international capital movements so long as the controls do not also restrict current transactions, and under article VI, section 1(a), resources from the Fund’s General Resources Account are not to be used to finance a large or sustained outflow of capital.53

However, gradual relaxation of initial limitations on the IMF’s involvement in the liberalization of capital transactions has been evident in a number of decisions and other changes since the late 1970s. In the amended version of the Articles of Agreement which took effect in 1978, article IV contains the statement that an essential purpose of the international monetary system is to provide a framework that facilitates the exchange of capital among countries. In April 1995 the list of developments that may trigger discussions between the Fund and a member country under IMF surveillance of exchange rate policies was extended to include “unsustainable flows of private capital”.54

In December 1997 approval was given to the establishment of the Supplemental Reserve Facility, under which financial assistance is extended to a country experiencing balance-of-payments difficulties due to sudden, disruptive losses of market confidence which are reflected in pressures on its capital account and reserves. Lastly, under a current initiative approved by the Interim Committee the Fund’s articles are to be amended to include the liberalization of capital movements amongst the organization’s purposes and to provide a formal extension of its jurisdiction to such movements.55

If such an amendment is adopted, it is capable of having knock-on effects on other internationally agreed rules. For example, under the WTO agreements regulating trade in goods and services, when a country has recourse to restrictions to safeguard its balance of payments, these restrictions are to be consistent with its obligations under the IMF’s Articles of Agreement, and the Fund has the role of assessing the restrictions’ justification on the basis of the country’s balance-of-payments and reserves position.

The proposed extension of the Fund’s formal jurisdiction to capital transactions might thus result in a reduction of countries’ existing autonomy regarding control of capital transactions under the WTO regime.56 But recent financial crises and the frequent recourse by countries to controls to contain the effects of swings in capital flows point to the case for continuing to accord governments such autonomy. The discussion in this chapter does not suggest that ways have yet been found at a global level to eliminate cross-border transmission of financial shocks associated with greater global financial integration or other pressures connected with capital movements which are capable of triggering financial crises.

These weaknesses in the existing armoury of policy measures raise serious questions as to the appropriateness of current steps to promote the liberalization of capital movements as an international policy objective. Indeed, for the foreseeable future, flexibility rather than additional constraints or obligations would appear to be necessary.

6. Exchange rate policies

Questions about connections between exchange rate regimes and financial crises have been raised by the contribution of stable exchange rates (with their accompaniment of excessive short-term external borrowing and increased currency risk) to the build-up of financial fragility in East Asia, and by the role of the subsequent movements in exchange rates in triggering the crisis and spreading it throughout the region. The conclusion of the discussion which follows is that no regime is likely to provide foolproof protection against such crises. However, managed exchange rates, in combination with controls on capital transactions, can do much to prevent large swings in capital flows, thus making an important contribution to macro-economic stability.

As described in box 2, at the outbreak of the crisis the East Asian economies most seriously affected, except Hong Kong, China, operated foreign exchange regimes under which the central bank intervened to stabilize the spot rate according to generally understood guidelines, while Hong Kong, China, had a currency-board arrangement. During 1990-1996 many East and South Asian countries were recipients of large capital inflows. Faced with such inflows, monetary authorities can either let the currency appreciate or intervene to prevent an appreciation, and most of the East Asian recipients chose the latter course. The question has been posed whether freely floating exchange rates would have been preferable to the managed rates in force in several East and South Asian countries before the crisis. Floating rates in the early 1990s would probably have led to sharp appreciations in comparison with the levels actually observed, which would almost certainly have provoked stiff political resistance owing to their effects on exports. If freely floating exchange rates had brought about greater instability in relative rates, they might have discouraged arbitrage flows but also ultimately have threatened the pattern of relatively stable exchange rates which underpinned economic development in the region, and might have risked causing tensions in trading relations.

At the other extreme, the suggestion has been made that crises like the Asian one might be avoided by the establishment of currency-board systems involving exchange rates rigidly pegged to an anchor currency. Under such a system there is an unequivocal commitment to supply or redeem monetary liabilities of the monetary authority at a fixed rate. Moreover, these are the only terms on which such liabilities are exchanged. Two particularly well known systems of this kind are those of Argentina and Hong Kong, China, but a number of other countries, including some transition economies, also operate them. In their purest form currency boards cannot extend credit to the government, the banking system or other borrowers, and interest rates are market-determined, the monetary base being rigidly linked to the country’s foreign exchange reserves. These conditions do not hold strictly in so far as the currency board has external reserves in excess of the economy’s monetary base (which has recently been the case, for example, in Hong Kong, China) or in so far as the legal framework permits some of the reserves backing the monetary base to be held in forms other than foreign currency (as in Argentina). The benefits attributed by its advocates to a currency board include the credibility conferred by such a regime on the monetary authority and the elimination of the problems of external debt management which result from mismatches between the currency denomination of borrowings and that of revenues generated by the activities they finance.

However, as recent experience demonstrates, such regimes do not insulate economies from instability of external origin since the impact of capital inflows and outflows is transmitted, via their effects on the monetary base, to levels of economic activity and to goods and asset prices. Moreover, in the absence of a lender of last resort, the contraction of deposits which typically follows capital outflows under such a system can threaten banking stability.57

A currency-board system may serve an important purpose in certain circumstances, such as to halt hyperinflation. But the consequent reduction in policy autonomy means that such systems will remain acceptable and appropriate only for a small minority of countries. There is thus no reason to condemn managed exchange rate regimes on the basis of recent experience (though their restoration in East and South Asian countries is impossible in the absence of a return to more orderly conditions in their currency markets).

The alternatives of floating or rigidly fixed exchange rates can also impose costs which can outweigh their benefits. However, recent experience has also shown that managed exchange rate regimes are vulnerable to large accumulations of short-term external bank debt and of other potentially volatile external investment. Occasionally, introducing greater flexibility by widening the band of intervention could help to eliminate one-way bets and discourage arbitrage flows. But such regimes are likely to be sustainable only if accompanied by active management of external liabilities, which may often entail recourse to capital controls.

7. Regional consultation and collaboration

Regional economic arrangements often include modalities for mutual consultation and collaboration covering a broad range of subjects of economic policy. Among the aims of these modalities is frequently prevention of financial crises in member countries which might have unfavourable cross-border effects and thus prejudice achievement of the objectives of the arrangement concerned. There is a comprehensive set of procedures of this kind for the European Union (EU), with the objective of ensuring that the functioning of the common market is not adversely affected by macroeconomic or financial developments in member States. The Treaty of Rome establishing the European Economic Community (Part Three – Title II, chapters 1-3) covered cooperation and consultation among members regarding monetary and other conjunctural policies as well as the balance of payments. Article 107, for example, enjoined members to treat exchange rate policies as a matter of common concern.58

Furthermore, consultation and surveillance have been part of EU procedures for the provision to members of financial support for intervention in currency markets and for helping to solve balance-of-pay-ments difficulties. Collaboration and consultation at the regional level have also been proposed elsewhere for the purpose of helping to prevent financial crises, much of the impetus behind such initiatives coming from the objective of avoiding contagion effects.

Some of the ideas put forward in this context involve mutual surveillance intended to help ensure that policies for economic and financial stability designed by the IMF are properly applied. But particular policies are not an integral part of regional collaboration and consultation in pursuit of financial stability. An ongoing initiative in ASEAN, for example, involves a mechanism for monitoring aspects of members’ economic positions and policies in accordance with guidelines mutually agreed for this purpose. Awareness among ASEAN members of the need for cooperation to prevent financial crises was already evident during the period before and leading up to the crisis. For example, at a meeting in March 1997 the ASEAN finance ministers acknowledged that a regional surveillance mechanism might be established for this purpose. Two months later, pressure on the Thai baht prompted coordinated intervention by a number of Asian central banks in support of the currency.

However, no such intervention took place at the time of the widespread abandonment of managed exchange rate regimes in July 1997, a fact which suggests a decision by governments that an attempt to defend exchange rates at that stage would have been too costly. Nevertheless, consultations within ASEAN on mutual surveillance continued and at a meeting in December 1997 the ASEAN finance ministers recommended implementation of the proposal for the establishment of a regional surveillance mechanism, which has subsequently come to be known as the ASEAN Monitoring Mechanism.

The function of the Monitoring Mechanism would be to help ASEAN governments to prevent future financial crises inter alia by serving as an early warning system which would enable corrective actions to be taken by individual countries or collectively and by protecting the region’s interests during the process of global financial integration via examination of financial and monetary issues raised in international forums. This initiative responds to an urgently felt need among ASEAN members but might also eventually lead to the extension of such cooperation to other Asian countries. If in the future a decision were to be taken by ASEAN and possibly other East and South Asian countries to establish more formal arrangements for the provision of mutual external financial support than those deployed in May 1997 in defence of the baht (which might be similar to some of the EU facilities mentioned above), it would be possible to envisage a link between these arrangements and whatever surveillance mechanism is then in place.

Notes

1 This idea actually goes back to the Committee of Twenty. It was revived by the IMF in 1994, and elaborated in a paper by the management, “A short-term financing facility” (Washington, D.C.: IMF, September 1994). For discussions of the issues raised see E.V.K. Fitzgerald, “Intervention versus regulation: The role of the IMF in crisis prevention and management”, UNCTAD Review, 1996; and J. Williamson, “A new facility for the IMF”, in UNCTAD, International Monetary and Financial Issues for the 1990s, Vol. VI (United Nations publication, Sales No. E.96.II.D.7), New York and Geneva, 1995.

2 M. Feldstein, “Overdoing it in East Asia”, Foreign Affairs, March/April 1998, p. 26.

3 IMF Survey, Supplement on the IMF, September 1996, p. 15.

4 IMF Survey, 12 January 1998, p. 7

5 IMF Survey, 20 March 1995, p. 86.

6 See, for example, A.J. Schwartz, “The world’s Central Bank: Time to terminate the ESF and IMF” (Washington, D.C.: National Bureau of Economic Research, 1998), mimeo.

7 TDR 1986, annex to chapter VI.

8 See also J. Sachs, “External debt, structural adjustment and economic growth”, in UNCTAD, International Monetary and Financial Issues for the 1990s, Vol. IX (United Nations publication, Sales No. E.98.II.D.3), New York and Geneva, 1998; and UNCTAD, World Investment Report 1992, box XI.1.

9 For a comparison between the United States, the United Kingdom and German bankruptcy codes see B. Eichengreen and R. Portes, Crisis? What Crisis? Orderly Workouts for Sovereign Debtors (London: Centre for Economic Policy Research, 1995); and J. Franks, “Some issues in sovereign debt and distressed reorganizations”, ibid., annex 2.

10 For such issues see D.M. Sassoon and D.D. Bradlow (eds.), Judicial Enforcement of International Debt Obligations (Washington, D.C.: International Law Institute, 1987).

11 This was also the case in Mexico during 1994-1995 when the Government was solvent, in the sense that it could repay all the outstanding holders of dollar-linked tesebonos in pesos, but the contract bank did not have sufficient reserves to allow these holders to convert the pesos into dollars.

12 See L. Nurick, “The International Monetary Fund Articles of Agreement”, in Sassoon and Bradlow (eds.), op. cit.; and Eichengreen and Portes, op. cit.

13 For a number of rulings based on such an interpretation see Nurick, op. cit., pp. 111-113.

14 International Capital markets. Developments, Prospects, and Policy Issues (Washington, D.C.: IMF, 1995), p. 11.

15 Ibid., p. 11.

16 This suggestion has been put forward by K. Raffer. See his “Applying chapter 9 insolvency to international debts: An economically efficient solution with a human face”, World Development, Vol. 18, No. 2, 1990.

17 Sachs, op. cit., p. 52.

18 Eichengreen and Portes, op. cit., pp. 49-50.

19 For a discussion of these objections see Eichengreen and Portes, op. cit., pp. 43-44; and Raffer, op. cit.

20 One of the best-known cases is the Chilean debt crisis of the early 1980s, resulting from the so-called Southern Cone experiment with liberalization. A very large part of Chile’s external debt during the late 1970s and early 1980s was accumulated by private banks and corporations and without government guarantees. “Those private debts have been included in debt rescheduling being negotiated between the Chilean state and the foreign bank advisory committee for Chile. Apparently the Chilean government caved in under pressure from the bank advisory committee ... To make their viewpoint absolutely clear, foreign banks apparently tightened up their granting of very short-term commercial credits to Chile during the first quarter of 1983, a technique reportedly used with some success 10 years earlier vis-à-vis the same country. The International Monetary Fund, also active in the debt rescheduling exercise, has not publicly objected to this threat” (C. Diaz Alejandro, “Good-bye financial repression, hello financial crash”, Journal of Development Economics, Vol. 19, No. 1/2, September-October 1985, p. 12).

21 From the ruling of the New York Court of Appeals in the Costa Rican case of 1984.

22 See TDR 1986.

23 G.A. Calvo, “The management of capital flows: Domestic policy and international cooperation”, in UNCTAD, International Monetary and Financial Issues for the 1990s, Vol. IV (UNCTAD/GID/G24/ 4), New York, 1994.

24 Executive Board Decision No. 5392-(77/63) of 29 April 1977.

25 Group of Ten, “The functioning of the international monetary system”, a report to the Ministers and Governors by the Group of Deputies (Washington, D.C., June 1985), para. 40. For further discussion see J. Williamson and M. Gavin, “International monetary issues in 1985”, and Y. Akyüz and S. Dell, “Issues in international monetary reform”, both in UNCTAD, International Monetary and Financial Issues for the Developing Countries (United Nations publication, Sales No. E.87.II.D.3), New York and Geneva, 1987.

26 During relatively short periods a country’s real effective exchange rate can vary by amounts which are large in percentage terms in comparison with its average tariff level, and the resulting changes, even away from the equilibrium level of the country’s exchange rate, may persist for some time. Thus the economic impact of movements in exchange rates can substantially exceed that of multilaterally agreed tariff changes, even though the estimated elasticities of trade flows with respect to tariff changes are considerably higher than those with respect to variations in exchange rates; see for example C.F. Bergsten and J. Williamson, “Exchange rates and trade policy”, in W. Cline (ed.), Trade Policy in the 1980s (Wash-ington, D.C.: Institute for International Economics, 1983).

27 Interim Committee Communiqué of 16 April 1998.

28 There has been much recent econometric analysis of the determinants of currency and banking crises as part of attempts to develop leading indicators of them. While this work has served to clarify the issues involved in the development of such indicators, substantial reliance on them for forecasting financial crises under global surveillance seems unlikely (as indeed is their replacement of existing quantitative and qualitative, if more ad hoc, indicators currently used by banking supervisors and financial analysts in investment banks). Efforts so far have produced indicators which significantly overpredict banking crises. For a brief discussion of these issues see M. Goldstein, “Early warning indicators of currency and banking crises in emerging economies”, in Financial Crises and Asia, CEPR Conference Report No. 6 (London: Centre for Economic Policy Research, 1998).

29 Credit risk results from the possibility that a bank’s counterparty will default on its obligations, and market risk is that of loss due to changes in the market value of a bank’s asset before it can be liquidated or offset in some way.

30 For example, William McDonough, President of the Federal Reserve Bank of New York, has made this point as follows: “... formerly, you could look at the balance sheet of a financial institution and quickly get a sense of exposure and risks. Today balance sheet information is clearly inadequate for this purpose ... the fast pace of activity in today’s market renders financial statements stale almost before they can be prepared”. See J.A. Leach, W.J. McDonough, D.W. Mullins and B. Quinn, “Global derivatives: Public sector responses”, Occasional Paper No. 44 (Washington, D.C.: Group of Thirty, 1993), pp. 15-16.

31 R. Dale, The Regulation of International Banking (Cambridge: Woodhead-Faulkner, 1984), p. 83. The author’s argument concerns crises in international bank lending but could easily be extended to financial crises more generally.

32 See, for example, M. Goldstein, “The case for an international banking standard”, in Policy Analyses in International Economics, No. 47 (Washington, D.C.: Institute for International Economics, 1997), p. 33.

33 See M. Goldstein and P. Turner, “Banking crises in emerging economies: Origins and policy options”, BIS Economic Paper No. 46 (Basle: BIS, 1996), pp. 9-14.

34 The nature of these guarantees has been well described in a recent book on the Asian financial crisis as follows: “In every economy, public authorities stand behind the viability of their domestic financial system ... This cannot be interpreted as a market distortion; it is a feature of a capitalist economy, in which markets for financial assets are an indispensable feature. These markets, however, cannot be liberalized in the same manner as one would a goods market ... This creates at least an implicit, if not explicit, guarantee that monetary authorities stand behind the foreign liabilities of, as a minimum, the explicitly supervised part of their financial system”. M.F. Montes, The Currency Crisis in Southeast Asia, updated edition (Singapore: Institute of Southeast Asian Studies, 1998), p. 26.

35 For a survey of proposals generated by that crisis see D.F. Lomax, The Developing Country Debt Crisis (London: Macmillan, 1986), pp. 255-280.

36 H. Kaufman, “Ten reasons to reform”, Euromoney, November 1992, p. 57. Kaufman has returned to this proposal since the outbreak of the Asian crisis, for example in his speaking notes for the Extraordinary Ministerial Meeting of the Group of 24 in Caracas in February 1998.

37 G. Soros, “Avoiding a breakdown”, Financial Times, 31 December 1997 and 1 January 1998.

38 The Multilateral Investment Guarantee Agency (MIGA), an affiliate of the World Bank, provides insurance against certain risks (such as transfer risks) associated with foreign investment and selected other international transactions, including loans linked to insured investments. There is some overlap between its insurance facilities and those of national export credit agencies, but the coverage of risks connected with international lending by the latter is generally more extensive and thus more suitable as a model for an ICIC.

39 See, for example, S. Irvine, “Rating agencies: Caught with their pants down”, Euromoney, January 1998.

40 It should be noted, however, that contagion effects of the kind witnessed during recent crises in emerging financial markets, which from the point of view of insurance bear some resemblance to natural catastrophes, would complicate the task of setting the premiums which would be charged by the ICIC. The majority of the ECAs of OECD countries experienced long series of cash-flow deficits on their operations after the developing-country crisis of the 1980s.

41 It has been suggested by M. Mayer that “the international community needs some sort of registry that would call attention to any bank’s or national banking system’s continuing increase in short-term borrowings from financial firms”. For most firms which are significant participants in international financial markets such information exists at national level (though for a market such as that for inter-bank transactions, which operates on a continuous basis, the choice of the time at which inter-bank positions must be disclosed may not be easy). See M. Mayer, “The Asian disease: Plausible diagnoses, possible rem- edies”, Jerome Levy Economics Institute Working Paper No. 232 (Annandale-on-Hudson, New York, April 1998), pp. 31-32.

42 As reported in an editorial in the Financial Times, 11 May 1998.

43 These tendencies have been especially evident in banks’ Eurocurrency operations. Traditionally, such operations were defined as those in currencies other than the currency of the country of domicile of the participating bank (originally mainly United States dollars but subsequently also those of other major OECD countries). But with the establishment of international banking facilities (IBF) in the United States and Japan the definition has been extended to specified transactions of banks in their domestic currencies, mainly with non-residents or in connection with international activities, subject to a regulatory regime similar to that for traditional Eurocurrency operations. These regimes are generally characterized by lighter regulation than those for domestic banking (though recent deregulation has led to a reduction in these divergences), and they have often also benefited from tax advantages.

44 See S. Griffith-Jones, “Regulatory challenges for source countries of surges in capital flows”, in J.J. Teunissen (ed.), The Policy Challenges of Global Financial Integration (The Hague: Fondad, 1998). As set out there, the proposal is characterized by the somewhat unfortunate term “prudential capital charge” for the liquid reserve requirements of mutual funds, institutions whose liabilities consist of shareholder capital.

45 This idea was broached in A. Cornford and J. Kregel, “Globalization, capital flows and international regulation”, Jerome Levy Economics Institute Working Paper No. 161 (Annandale-on-Hudson, New York, May 1996), p. 29. For more detail on the exit fees of mutual funds see J.C. Bogle, Bogle on Mutual Funds (Burr Ridge, Illinois: Irwin, 1994), pp. 193- 194.

46 The situation is further complicated by derivatives. Recent innovations have increasingly made possible the engineering of “synthetic” financial instruments and portfolios with cash flows through time that match those of more traditional assets. As a result, if a government wishes to target certain traditional assets in controlling capital movements, it may need to expand the scope of its action to “synthetic” instruments or portfolios.

47 This tax is sometimes referred to as an “interest equalization tax”. The original tax so designated was imposed by the United States on foreign lending, initially in 1964 on foreign securities with a maturity of more than three years and subsequently, in 1965, extended to bank loans.

48 Accounts denominated in foreign currencies are available in the East Asian countries most affected by the crisis, in most cases subject to only limited restrictions. But in most OECD countries such availability is a relatively recent development associated with the more general liberalization of capital transactions. As late as the mid-1980s such accounts were still not permitted in some instances.

49 Concerning the origins of offshore banking in Singapore see M. Ishihara and H.C. Kim, “Financial system of Singapore”, in R.C. Effros (ed.), Emerging Financial Centres (Washington, D.C.: IMF, 1982). For more recent developments the discussion in the text relies on an unpublished paper by M.F. Montes and T.K. Giap.

50 The account which follows of capital controls in selected developing countries relies heavily on an un-published paper by C.M. Reinhart and R.T. Smith, “Temporary capital controls”, August 1997 (mimeo); and on V.G. Le Fort and C.L. Budnevich, “Capital-account regulations and macroeconomic policy: Two Latin American experiences”, in UNCTAD, International Monetary and Financial Issues for the 1990s, Vol. VII (United Nations publication, Sales No. E.97.II.D.5), New York and Geneva, 1997.

51 An outright forward exchange transaction involves an agreement between two parties to exchange currencies after a period of more than two days hence, while a foreign exchange swap has two separate legs, one consisting of the sale or purchase of a foreign currency and the other of a repurchase or resale of the currency at a subsequent date (thus reversing the first leg).

52 Developed countries are also subject to obligations in this area under the OECD Code of Liberalization of Capital Movements and (for members of the EU) under the EEC Council’s 1988 Directive on capital movements and the Maastricht Treaty. Some developing countries have undertaken such obligations as part of treaties of friendship, commerce and navigation or of regional agreements such as the North American Free Trade Agreement (NAFTA).

53 This restriction does not necessarily apply to the IMF’s special facilities, which use borrowed resources.

54 IMF Executive Board Decision No. 10950-(95/37) of 10 April 1995 (amending Decision No. 5392-[77/ 63] of 29 April 1977).

55 The financial rescue package presented by the IMF to the Government of the Republic of Korea contained conditions relating to the liberalization of capital transactions.

56 For further discussion see A. Cornford and J. Brandon, “The WTO agreement on financial services: Problems of financial globalization in practice”, section E, in UNCTAD, International Monetary and Financial Issues for the 1990s, Vol. X (forthcoming).

57 With the objective of avoiding another liquidity squeeze like that experienced owing to capital outflows in the aftermath of the Mexican crisis in 1995, Argentina has arranged a stand-by financing facility from private international banks for use in the event of the resumption of such outflows.

58 These articles were amended by the Maastricht Treaty, which provided for the establishment of Economic and Monetary Union.

end of Chapter IV

Back to Top                   Back to |Chapter III            Back to Globalization