Rethinking the International Monetary System: An Overview

Jane Sneddon Little

When the Federal Reserve Bank of Boston chose "Rethinking the International Monetary System" as the topic for its 43rd Economic Conference, it was clear that the worst international financial crisis in decades had caused tremors within the economics profession and the policymaking establishment. The miracle countries of Asia had suffered sharp currency devaluation and deep economic downturns, the turmoil had spilled over into Russia and Latin America, and a severe liquidity crisis had briefly threatened banking systems in the advanced countries. Thus, economists and policymakers had begun to question some of their most basic beliefs about appropriate international financial arrangements. Of course, as Paul Volcker was to note at the conference, the worst international financial crisis in 50 years occurs about once a decade, but that recurrence conveys a telling message about the efficacy of existing arrangements.

The events of the 1990s--the European currency turmoil early in the decade, the subsequent introduction of the euro, the relatively contained Latin American crises of 1994-95, and the global financial storms of 1997-98--have provoked many proposals for reform. But these proposals reflect differing, even contradictory, views about the underlying problems and their solutions, and they do not always reveal a systemic approach to reform. For instance, while some reformers advocate more flexible exchange rate arrangements, others seek irrevocably fixed regimes, at least for some countries. And while most policymakers remain staunch supporters of free capital markets, some stress that volatile short-term capital flows played a central role in recent crises and argue that capital controls just might be useful. Observers also hold diametrically opposed views about the international lender of last resort. Some believe that the lack of an effective international lender of last resort has contributed to recent crises, while others are convinced that large international rescues have produced moral hazard and more frequent disruptions. As for policy surveillance, while some analysts argue that growing integration requires increased policy coordination, they differ as to whether such cooperation should be achieved through improved transparency and market discipline or strengthened international governance.

In hopes of clarifying some of these issues, the Bank asked conference participants to examine key parts of current international monetary arrangements: the eclectic exchange rate system, international capital markets, the international lender of last resort, and policy coordination. We also asked them to consider how these critical components interact. We hoped that adopting a systemic approach would help to narrow the differences among economic policymakers and identify priorities for reform. Our ultimate goal was to define ways to enhance the benefits of global integration while limiting its costs. This article summarizes the participants' answers to our questions.

A Historical Perspective on International Monetary Arrangements

In his opening address, historian Harold James provided a rather skeptical review of efforts to reform the international monetary system since the revolution in communications and transport of the mid nineteenth century started the process of "globalization." While the current debate on the international architecture has its roots in the recent financial crises of Mexico and East Asia, the underlying problems have been contested in some form or another for a very long time. Over the last 150 years, proposals for large-scale reforms of the international financial system have been numerous, with no shortage of good ideas. Unfortunately, the most common outcome of these past discussions has been the partial realization of grand designs. The establishment of the gold standard in the 1870s, for example, was the byproduct of the more ambitious idea for a world currency union, advocated by Napoleon III in 1867. The idea had the backing of Germany and United States, but eventually failed because Britain would not ag ree to a small change in the weight and value of the pound.

For a reform proposal to have a chance of being successfully realized, James argued, requires a strong will to reform and bilateral, if not unilateral, leadership of the reform negotiations. The key to the success of the Bretton Woods conference in 1944 lay in the unique combination of its timing and the prominent position of the United States in the negotiations. At that time, most analysts saw an urgent need to have an institutional framework in place before the beginning of the postwar period, and, while 44 countries were represented, the negotiations were essentially between Britain and the United States, with the latter clearly predominant.

The history of failed international monetary cooperation in the interwar years provides a stark contrast with the success of the Bretton Woods conference. Insufficient urgency about reform explains why the Bank for International Settlements failed to become an effective institution for international cooperation in crisis prevention. Created in 1930 before the international panic set in, the Bank had already ceased to operate effectively in 1931, paralyzed by inadequate capitalization, conflicting goals, and polarized opinions in France and Britain on the issue of Germany's war reparations. As for the World Economic Conference of 1933, while the financial crises of 1931 and 1932 had made abundantly apparent to all that "the world economy was crippled by monetary chaos ... and trade wars," the lack of a bilateral or unilateral leadership made it impossible to reach any consensus among the 66 participating countries.

All told, James noted, a historical survey of large-scale reform efforts "inclines the observer to rather pessimistic conclusions" about the likely outcomes of new proposals to overhaul the international financial system. While a new bilateral axis between Europe and the United States may be emerging, and while this relationship may succeed in pushing new projects to fruition, it is possible that the outcome could be bad. The risk, according to James, is that the new bilateral relationship will push in the direction of fixed exchange rates, a solution that appeals to many European businesses but that is "exactly the wrong sort of answer to the crises of most emerging markets." In addition, while some changes may be needed to limit a country's vulnerability to sudden capital flow reversals, a substantial buildup of protective measures may in the end harm the international system. This scenario occurred during the Great Depression, when the fear of destabilizing capital flows led to cumulative measures that ul timately caused the collapse of international trade. Not infrequently, James warned, the remedies turn out to be worse than the problems themselves.

Why the Interest in Reforming the International Monetary System?

In the introductory paper, Jane Sneddon Little and Giovanni Olivei provided an overview of recent changes in the economic environment that have sparked interest in reform and of the debates surrounding key aspects of the international financial system. Little and Olivei argued that the interest in reform stems from the perception that the Asian crisis was not simply the outcome of national policy mistakes, but also of shortcomings in current international arrangements. The international financial system may have worsened recent crises in several ways. First, the recent liberalization of international capital markets, widely recommended by the IMF and G-7 countries, severely limits a developing country's ability to pursue an independent monetary policy both under fixed and flexible exchange rate regimes. Further, while previous international rescues may have created perverse incentives, the IMF currently faces handicaps as an international lender of last resort since it cannot lend quickly and reliably. Final ly, while conditions in global capital markets or neighboring countries' policy mistakes create harmful spillovers, opportunities for meaningful policy coordination are currently limited.

Little and Olivei observed that the recent crises have shaken the economics profession's confidence concerning several basic issues, including the ability to prescribe appropriate exchange rate policy. Opinions also differ widely on how to weigh the pros and cons of capital account liberalization. Still, important lessons have already been drawn from recent crises. In particular, it is clear that unilateral pegs pose risks, in that the demise of a peg can precipitate a creditors' panic and the resulting collapse in bank lending can have devastating effects on economic activity. Moreover, countries must be wary of liberalizing their capital accounts without adequate institutions for monitoring their banking sector. And greater transparency, disclosure, and better governance are crucially important to improving supervision and reducing moral hazard.

Beyond these lessons, Little and Olivei posited the need for more fundamental changes. While free markets may promote growth over the long run, capital flows can be highly destabilizing in the short run; thus capital controls may be advisable both as an emergency measure and as a defense against systemic risk when financial supervision is limited, private sector risk-management is inadequate, and financial markets are thin. In addition, they proposed designing an international lender of last resort that could mitigate financial panics by providing timely short-term liquidity to banking systems in need. Greater market-based surveillance could help to limit the scope for international-lender-of-last-resort intervention and might render more effective oversight than multilateral institutions have generally achieved.

Little and Olivei also pointed out that issues of international policy coordination and emergency liquidity are likely to prove irrepressible, and unless some combination of better information, a more reliable international lender of last resort, and more effective surveillance allows governments to achieve greater stability, some emerging economies are likely to seek protection by joining a currency bloc--even if these unions do not represent optimum currency areas. Overall, Little and Olivei emphasized the need to take a systemic view on improving the international monetary system. To date, many proposals for reform have focused on specific aspects of the problem--such as transparency and governance--which may reduce the frequency and severity of future crises, but will not fully resolve the conflicting needs of all countries to participate in integrated markets and to achieve stable economic growth.

In commenting on Little and Olivei's paper, Toyoo Gyohten focused on the recent East Asian experience. Gyohten stressed that many of the problems that have prompted the current debate on "architecture" are region-specific; thus, it is important to devise measures that are tailored to region-specific needs. Many of the crisis-hit economies in East Asia were experiencing macroeconomic imbalances, most notably in the form of increasing external deficits and inflation and interest rate differentials vis-a-vis the United States. In addition, while the countries in the region had a dollar-peg, the currencies were still subject to large swings against the yen that contributed to a worsening of the macroeconomic scenario. Gyohten argued that East Asian countries should be prepared to remain flexible in their exchange rate arrangements when large distortions are present, but that it is equally important that the United States and Japan "cooperate more seriously to achieve greater stability and predictability of the d ollar-yen exchange rate."

As concerns international capital flows, Gyohten endorsed a strategy for sequencing the liberalization of the capital account, together with a package of emergency capital controls measures. He noted that the East Asian crisis would have been less severe if capital flows had been better managed, and he favored market-friendly measures such as additional reserve requirements or differentiated interest rates to regulate short-term flows. The liquidity shortage associated with the sudden reversal of capital flows in East Asia also points to the need for quick injections of foreign currency liquidity. Yet it is unrealistic to expect that Mexican-style rescues will be readily available in the future. The viable alternative, according to Gyohten, is to establish a regional credit facility financed by countries that have close ties in the region. Such a facility could be "a regional vehicle of the IMF, provided that regional members make majority contributions and hold majority voting rights" to ensure that the fac ility maintains the flexibility necessary for timely interventions.

Finally, the East Asian crisis points to the need for a "forum to conduct dialogues more focused and more relevant to the situation in the region." The lack of an adequate forum for conducting surveillance and applying peer pressure for corrective measures was a serious shortcoming before the East Asian crisis. A regional forum within the IMF would provide a useful step toward better policy coordination in the region.

Ricardo Hausmann argued that the often-mentioned view that moral hazard played a significant role in the recent financial crises is fundamentally wrong. According to the "moral hazard view," the presence of implicit or explicit government guarantees, together with weak supervision and fixed exchange rate arrangements, created a fertile ground for excessive and poorly allocated international capital flows. Hausmann noted that the fundamental flaw in this view is that capital flows across borders are actually surprisingly small. For example, capital-labor ratios in Latin America are less than one-third those in the United States, and at the current rate of net foreign investment into Latin America, this difference will persist for centuries.

Hausmann contended that the true causes lie in two fundamental problems faced by developing countries. First, foreigners are not willing to lend to a developing country in that country's own currency. Second, domestic residents are similarly reluctant to lend long-term in the domestic currency. This reluctance creates currency and maturity mismatches, since in order to finance long-term investment projects, a developing country must either borrow in a foreign currency, or borrow short-term in its own currency. These mismatches can generate currency and financial crises that feed on each other.

The solution, according to Hausmann, is not in greater exchange rate flexibility. Latin American countries with flexible regimes have had difficulties in developing deep financial systems, because floating regimes tend to generate currency appreciations in good times and depreciations in bad times. Thus, domestic residents shy away from instruments denominated in their floating currency because they offer limited opportunities for hedging against income fluctuations.

The solution is for developing countries to abandon weak national currencies and adopt a supranational currency. Such an arrangement signals a serious precommitment not to devalue, thus avoiding currency cum financial crisis scenarios. This arrangement also eliminates the need for an international lender of last resort except at the level of the currency union, where sharing arrangements can be devised. In this way, the scope for IMF interventions would be drastically reduced: Crises would be addressed at a regional level, and the role of the IMF would be redefined to deal only with the exchange rates of the few supranational currencies.

Exchange Rate Choices

Richard Cooper addressed the vexing question of optimal exchange rate arrangements. Economists' lack of fully persuasive answers is especially unfortunate because, "for most countries, all but the largest with the most developed domestic capital markets, the choice of exchange rate policy is probably their single most important macroeconomic policy decision, strongly influencing their freedom of action and effectiveness of other macroeconomic policies." Moreover, the inadequacy of the existing theoretical frameworks for addressing exchange rate choices has translated occasionally into "quite poor advice to decision-makers."

In reviewing the history of thought on exchange rate arrangements, Cooper noted that the interwar experience with flexible exchange rates, admittedly under extremely difficult circumstances, was generally viewed by contemporaries as highly unsatisfactory. In an influential study, Ragnar Nurske strongly argued in 1944 against exchange rate flexibility, on the grounds that floating rates are destabilizing in their behavior and thus are a substantial source of uncertainty for trade and capital formation. This aversion to exchange rate flexibility underlay the postwar economic order negotiated at Bretton Woods in 1944.

Still, the Bretton Woods arrangement of exchange rates fixed beyond a narrow band of permissible variation eventually found itself under severe strain, with monetary authorities holding an unsustainable parity for too long, and then folding to the speculative pressures from resurgent international capital movements. As a result, an increasing number of economists started to favor some form of exchange rate flexibility among major currencies. Harry G. Johnson's essay "The Case for Flexible Exchange Rates, 1969," whose title intentionally drew on Milton Friedman's famous memorandum of 1950, was very influential in shaping economists' views on the topic. According to Cooper, the essay appears "somewhat naive" today, with many of its claims based on "an idealization of the world of financial markets without serious reference to their actual behavior."

The trend toward greater exchange rate flexibility among major currencies that started in 1973 was subsequently reversed in Europe, where flexible rates were considered disruptive to the functioning of the Common Market and were replaced by the European Monetary System. The recent creation of a European common currency and the role played by fixed but adjustable rates in triggering the financial crises in Latin America and East Asia in the 1990s are events that have infused new life into the exchange rate debate.

According to Cooper, several factors have inhibited serious resolution of exchange rate choices, in particular "the continuing use by the economics profession of an extraordinarily primitive theory of money in its theorizing," and the lack of convincing empirical evidence on the influence of exchange rate arrangements on economic performance. Still, the post-Bretton Woods experience with floating has shown that Johnson's prediction that real exchange rate movements would track inflation differentials is largely counterfactual, with real exchange movements dominated in the short and medium run by nominal exchange rate fluctuations. In addition, while floating rates among major currencies have not proved disruptive to the extent envisaged by Nurske, Cooper conjectured that even Johnson would have been surprised by the recent swings in the real exchange rate between United States and Japan, movements whose amplitude "cannot characterize a well-functioning exchange rate regime."

In laying out his view on exchange rate choices, Cooper maintained Johnson's distinction between developing and developed countries. Cooper argued that for developing countries, more is at work than the well-known policy "trilemma," whereby independent monetary policy and fixed exchange rates are incompatible with freedom of capital movements. When financial markets are poorly developed, independent monetary policy and flexible exchange rates are also incompatible with freedom of capital movements. The reason is that a developing country's price level tends to be strongly influenced by the nominal exchange rate vis-a-vis the rest of the world, and with thin financial markets a single large player can move the exchange rate radically. As a result, changes in portfolio sentiment can have large effects on the domestic price level through movements in the exchange rate, and potentially "disrupt [goods and services] markets on which the economic well-being of the majority of residents depends."

Thus, in Cooper's view, the two prescriptions regularly extended to developing countries, to float the exchange rate and to liberalize international capital movements, may be in great tension with each other. For monetary policy to pursue independent goals, some form of restrictions on capital movements may be necessary not only for tightly managed exchange rates, but also for pure floats. The actual exchange rate choice for developing countries is not easy, depending, among other factors, on "how flexible are their wages and rents; on how supple and effective is their management of fiscal and monetary policy; on their administrative capacity to enforce restrictions on capital movements." Given the complexity of the choice, "countries are not obviously foolish for being reluctant to embrace floating exchange rates enthusiastically."

For rich and diversified countries, Cooper noted that exchange rate markets have often moved in ways inconsistent with Johnson's view of farseeing and universally stabilizing behavior. Such a failure is likely to become more apparent in the near future, when financial factors will come to dominate exchange rate determination to an even greater extent than they do today. Thus, volatility among major currencies is likely to increase as exchange rates become "ever more important in determining the profitability of trade and investment." For these reasons, Cooper argued that the recent introduction of a single currency among European countries could be taken a step further, with the creation of a currency union among Europe, Japan, and the United States. While acknowledging that such a proposition is politically unrealistic at this stage, Cooper conjectured that as rich countries become even more diversified, "real shocks among these entities will not be radically asymmetrical." The reduction in asymmetric distu rbances among industrial countries will limit the appeal of flexible exchange rates as a shock absorber. As a result, "the cost-benefit calculation ... will gradually alter the balance against flexibility, even for large countries."

Takatoshi Ito praised Cooper's comprehensive analysis, his care to distinguish between exchange rate choices for developing and developed countries, and his pragmatic approach to capital flows. Ito noted that Cooper did not seem to embrace the two-corner solution view, whereby a developing country should have either a truly fixed exchange rate system (such as a currency board or a currency union), or a freely floating exchange rate. While Cooper was skeptical about the desirability of free floats for developing countries, he did not really answer his own question, "What should developing countries do?" Ito would have preferred Cooper to explore in more detail the "middle ground," and discuss the exchange rate regime's choice in the context of an optimal sequencing of reforms.

As concerns Cooper's vision of a common currency for the world, Ito noted that Cooper did not provide a transition scenario toward the single currency. In this respect, Ito envisioned two possible paths. The first is to fix or have a narrow target zone among the three major currencies, and then have developing countries follow the lead. The second path is to have the three major currencies evolve first into regional currencies, and then link among themselves. Ito considered this last scenario as the most plausible, with some probability that the North and South Americas will evolve into a dollar zone, and that Euroland will extend to all Europe and Africa. However, according to Ito, it is very difficult at this stage to discern which currency is likely to become predominant in Asia.

Fred Bergsten agreed with Cooper that fully flexible exchange rates can be extremely costly for most developing economies. He added that the costs were also high for the United States under Beryl Sprinkel during the period 1981 to 1985, when a dollar appreciation prolonged a recession in "much of American manufacturing and agriculture, with some irreversible effects because of induced foreign investment." In addition, given that few countries meet the criteria for the adoption of a truly fixed exchange rate arrangement, the often mentioned two-corner solution is of limited practical relevance. Bergsten thought that Cooper's case for a move toward a currency union among the United States, Japan, and Europe a decade or two into the twenty-first century is correct, but that this leaves open the issue of how a managed float should operate in the next 10 to 20 years.

Bergsten advocated wide band ([plus or minus]10 to 15 percent) target zones among the three major currencies. The purpose of such target zones is to avoid "huge and prolonged misalignments" that are costly both to the countries involved and to the world economy as a whole. According to Bergsten, wide margins would encourage stabilizing speculation, while the target zones could be defended by sterilized intervention and, possibly, by changes in national monetary policies "consistent with the long-term requirements of the domestic economy." The alternative, Bergsten argued, is ad hoc episodic intervention by the G-7 as is actually practiced, with the disadvantage that it is almost always undertaken too late. Target zones would achieve "most of the virtues of both fixity and floating while avoiding the worst features of both," and they would provide a transitional regime that, via a progressive narrowing of the band, would evolve into the currency union envisioned by Cooper.

Catherine Mann questioned Cooper's assertion that the post-Bretton Woods regime with flexible rates has been a disappointment. She noted that growth in employment and incomes has been greater under flexible than under fixed rates, although a causal link is difficult to establish. Mann argued that the evaluation of alternative exchange rate regimes is difficult, since it involves a comparison between second-best alternatives and an assessment of the costs of exchange rate volatility. For example, a developing country's choice of an exchange rate regime is often decided separately from the degree of capital openness. But to the extent that a relatively rigid exchange rate arrangement requires more capital controls than a flexible exchange rate, it is unclear whether the reduction in exchange rate volatility achieved in the rigid regime can more than compensate for the cumulative inefficiencies and distortions caused by the controls on capital flows.

Mann also noted that "policymakers often wish that the exchange rate regime would solve all their policy dilemmas." For example, by establishing an exchange rate target zone, policymakers hope to make private capital a stabilizing force in the exchange rate market. Still, given that the willingness of policymakers to defend the zone is always in doubt, the target zone itself can provide no more discipline to the market than it does to policymakers. In addition, policymakers often wish that the exchange rate regime would force broad and substantive changes in the economy, changes that policymakers themselves cannot accomplish. The euro has been introduced with hopes that it will foster a liberalization of labor laws and promote a more efficient corporate behavior. While it might be too early to judge the euro's effectiveness in achieving these goals, Mann suggested that this is generally "too much to ask of the exchange rate." An exchange rate regime per se cannot force changes in a policy stance, even though movements in the exchange rate may force such changes.

International Capital Flows and the Emerging Markets: Do the Rules of the Game Need Amending?

Sebastian Edwards explored the Chilean experience with controls on capital mobility over the last 20 years. He speculated that while a large-scale reform of the IMF and other major multilateral institutions is improbable, important changes in exchange rate arrangements and in country-specific rules governing capital mobility are likely to occur. The East Asian crisis of 1997-98 has revived a debate on the appropriate sequencing of reforms, and specifically on the appropriate timing for liberalizing the capital account. Analysts have reached a wide consensus on the proposition that major fiscal imbalances have to be tackled first, and that the liberalization of the capital account should only take place thereafter. This sequencing reflects the fact that a real exchange rate appreciation induced by large capital inflows will most likely cause a deterioration in a country's competitiveness, possibly frustrating the reform process.

Recently, several authors, with Edwards figuring prominently among them, have augmented this proposition with the prescription that relaxation of capital controls "should only occur once a modern and efficient bank regulatory and supervisory framework is in place," that is, at a late stage of the reform effort. Poor bank regulation in a newly liberalized environment can result in excessive borrowing from abroad, the more so when implicit or explicit guarantees are in place.

Edwards, however, warned that the efficacy of capital controls in shielding a country from a real exchange rate appreciation, from overborrowing, or from sudden capital flow reversals, should not be overstated. The Chilean experience with capital controls is a case in point because, according to Edwards, it suggests that restrictions on capital inflows are unlikely to reduce a country's vulnerability to financial turbulence. This is especially true when controls encourage "complacent and careless behavior on behalf of policymakers and market participants." In addition, evidence that Chile's controls helped prevent a real exchange rate appreciation or allowed the pursuit of an independent monetary policy is regarded by Edwards as scant, at best.

In examining the effectiveness of Chilean capital controls, Edwards first noted that over the period 1978 to 1982 controls did not prevent a full-blown currency and financial crisis. One of the main reasons for the 1982 collapse was that the restrictions were not accompanied by an adequate effort at supervising the quality of bank portfolios. Despite an environment where short-term inflows had been controlled quite severely, the level of long-term foreign indebtedness in the private banking system surged dramatically in the early 1980s. Such a surge was not seen by regulators as worrisome, on the grounds that foreign indebtedness carried no government guarantee. Still, once the presence of bad bank loans and of an overvalued real exchange rate became apparent, foreign capital inflows came to a sudden stop and domestic investors started a capital flight. The ensuing financial collapse was so widespread that the government had to bail out a large portion of the banking system, at a cost of approximately 18 per cent of pre-crisis GDP. Edwards then proceeded to analyze Chile's experience with capital controls in the 1990s. Controls on short-term inflows were reinstated in 1991 with the two goals of (1) reducing the total volume of flows to delay a real exchange rate appreciation, and (2) of tilting the composition of Chile's foreign liabilities toward longer maturities to mitigate the country's vulnerability to financial instability. In addition, the authorities expected that capital controls would help the country pursue an independent monetary policy.

Overall, Edwards argued, the controls were largely ineffective. Regarding the volume of inflows, Edwards noted that with the exception of a brief decline in 1993, the total volume of flows into the country continued to increase until 1998. The restrictions did affect the composition of inflows, with shorter flows declining steeply relative to long-term flows. As a result, Chile's short-term debt as a proportion of total debt decreased from 19 percent in 1990 to less than 5 percent in 1997. Still, Edwards argued that these figures tend to understate Chile's vulnerability to sudden capital flow reversals. The reason is that the data report the contracted maturity of flows, but do not measure the "residual" maturity, that is, the value of foreign outstanding debt that will mature in less than a year. If one uses this more appropriate measure, the proportion of short-term debt remained above 50 percent in mid 1996, and Chile's position did not appear to differ greatly from that of Argentina, a country with no ca pital controls during the period.

As concerns the effects of capital controls on the real exchange rate, Edwards showed that, contrary to the expectations of Chilean authorities, the introduction of controls did not seem to affect the behavior of the real exchange rate. Comparing a period with no restrictions, 1986 to 1991, with the more recent period with controls suggests that the estimated response of the real exchange rate to an increase in inflows is almost identical.

Finally, the Chilean monetary authority expected that the introduction of controls would help it pursue a relatively tight policy, a stance that had become increasingly difficult to sustain during the late 1980s and early 1990s because higher domestic rates were attracting an increasingly large volume of capital. However, Edwards showed that the equilibrium interest differential (after adjusting for expected depreciation) between Chile and the United States remained remarkably similar during the periods 1986-91 and 1991-96. In addition, despite the presence of controls, over the 1991-96 period deviations from the equilibrium interest rate differential reverted to zero almost as fast as during the previous period, suggesting that the restrictions did not prevent arbitrage and did not allow the monetary authority greater control over domestic interest rates.

In sum, Edwards suggested that while controls on capital movements should be lifted carefully and gradually, the true solution to problems caused by volatile capital flows is to be found in pursuing sound macroeconomic policies, avoiding overly rigid exchange rates, and building adequate supervisory and regulatory institutions.

Tom de Swaan agreed with most of Edwards's conclusions. He noted that the optimal sequencing of reforms was neglected by a number of emerging economies, and that "at the time, the multilateral organizations and the international financial community failed to recognize the full significance of this error." De Swaan argued that the Chilean crisis of 1982 raises the question as to whether stricter prudential regulation, for example in the form of ceilings on banks' net foreign debt position and foreign currency exposure, may be preferable to capital controls. In his view, improvements in supervision and regulation can go "a long way toward preventing unjustifiable large capital inflows that might temporarily lead to impressive but unsustainable growth rates."

An effective way to boost the efficiency and improve the transparency and soundness of an emerging country's financial system is, according to de Swaan, to open the financial services market to foreign bank competition. Emerging countries have been traditionally reluctant to do so, because of the competitive threat to local banks. Still, foreign banks' penetration would introduce international standards and facilitate their adoption by local banks.

As concerns the most recent Chilean experience with capital controls, de Swaan argued that the reason why the restrictions were not very effective is that lower short-term inflows were entirely compensated by higher long-term inflows. Exchange rate and independent monetary policy objectives can be achieved only with "very extensive restrictions on short- and long-term in- and outflows of capital." Given the high costs of such extensive restrictions, it is preferable to limit them to emergency situations, provided they are temporary and improvements in domestic macroeconomic policies and financial supervision are also undertaken. Similarly, controls on the capital account should eventually be lifted, but gradually and in conformity with the appropriate sequencing of reforms. In this respect the IMF should monitor a country's progress and set timetables, "as an aid to the introduction of the needed domestic reforms against vested interests at home."

William Cline welcomed Edwards's evaluation of the Chilean experience because Chile is "invariably cited as the example of experience with disincentives to short-term capital inflows." Chine's position was that while capital flows, just like trade flows, contribute to an efficient allocation of resources, Chilean-style disincentives to short-term flows still belong on the menu of policy options, preferably in a temporary form and only in cases of global capital market exuberance and large inflows. While Edwards argued that Chile had a crisis in 1982 despite the presence of severe capital controls, Chine remarked that such a crisis was the result of large current account deficits and an overvalued real exchange rate. In other words, the Chilean crisis in 1982 was not a capital account crisis, as was recently the case in East Asia. Chile still had to finance the ongoing external deficit, even in the absence of short-term debt, and a fixed exchange rate made it easy for domestic residents to shift savings abroa d when confidence started dwindling.

As concerns the controls during the 1991-98 period, Chine was surprised by Edwards's finding that, when considering residual maturities, the share of short-term in total debt in Chile was not dissimilar to that in other Latin American and East Asian countries. Cline argued that this similarity might reflect the fact that Edwards considers bank debt only. Broadening the measure to include nonbank debt shows, according to Chine, that the portion of Chile's total debt maturing in less than a year was quite low, suggesting that controls were not ineffective. Moreover, Edwards's finding that the total amount of long- and short-term external debt was not affected by the presence of controls might be a desirable outcome, since it means that controls can "dollar for dollar, shift your short-term debt to long-term debt."

Chine found Edwards's discussion on the impact of capital controls on the real exchange rate and domestic interest rates to be tangential to the policy debate on architecture, which focuses instead on whether controls can limit a country's vulnerability to sudden capital flow reversals. In this respect, Edwards's case for the limited effectiveness of Chilean-style capital controls was unconvincing, according to Clime. Foreign direct investment to emerging economies has been remarkably stable throughout all recent crises, and in situations where emerging countries are awash with money, it seems prudent to provide tax incentives favoring such long-term inflows.

The Politics of World Monetary Reform

In his address on the politics of the international monetary system, Robert Keohane argued that economic policymaking would be far more effective if it reflected a good understanding of operative political pressures. He presented five major constraints affecting the use of power in the international arena. The most fundamental constraint relates to politicians' self-interest. Because most leaders are chosen domestically, in liberal democracies international agreements will only be made and enforced when they serve the politicians' domestic agenda. Domestic ideas about the impact of international monetary policies on the domestic economy will generally matter a great deal, but, in the short run politicians may simply seek to shift blame, or at least to avoid it (as when the Russian crisis unleashed Congressional approval of IMF funding). Similarly, interest groups, like money center banks, may play a major role in specific cases, especially if the issues involved are not widely discussed.

Second, political/military relationships will influence the development of international institutions. Here, Keohane pointed to the role of these links in the creation of the Marshall Plan and the European Monetary Union. Third, credibility is likely to be the key source of power in the Information Age; thus, states where monetary institutions are not fully trusted will seek links with nations/institutions of unquestioned credibility, for example, the Federal Reserve or the European Central Bank. Fourth, in multilateral organizations, precedents matter, and players must consider the impact of today's bargains on tomorrow's rules. Thus, Keohane speculated that international decision-making in institutions governed by precedent may strengthen the pull of the long-term collective interest.

Fifth, referring to Albert Hirschman, Keohane noted that the exit options available to mobile factors of production amplify their political voice, possibly disproportionately. By contrast, democratic institutions can empower immobile factors by providing a governing framework within which market forces must operate. In this context, Keohane speculated that democratic publics will want to expand the scope of issues governed by multilateral institutions. In summarizing these points, Keohane concluded that global institutions, which are hard to create and hard to change, matter because they help define incentives and capacities for, as well as constraints on, the use of power.

Arguing that politics involves social purpose as well as imperatives, Keohane then proposed that three important values--autonomy, equity, and accountability--be added to the efficiency/political feasibility trade-off which economists often use. Regarding the efficiency/autonomy trade-off, he suggested that the value of local control depends on the quality of domestic decision-making; autonomy is valuable in democracies, less so in autocracies and kleptocracies. Because all political systems, including democracy, give little weight to equity, Keohane also suggested that the designers and managers of the international monetary system should seek to promote that value at the margin. Finally, Keohane concluded that the new century will need a Madisonian moment in which we learn to create and simultaneously constrain power in the world economy. Quoting Madison's advice that "... you must first enable the government to control the governed and in the next place advise it to control itself," Keohane called for the creation of international institutions that can deal effectively with world financial issues and remain accountable in the long run to democratic publics. Because democracies will only support global financial institutions viewed as serving the public interest, the policy process must be transparent, and over time policy outcomes must reflect the public will. But since policy effectiveness can create legitimacy, this requirement need not imply direct electoral control, and short-term policy measures can be insulated from short-term political pressures.

International Lender of Last Resort: What Are the Alternatives?

In addressing the question, "Does the world need an international lender of last resort?" Jeffrey Sachs answered "yes," as long as the emphasis is on last, as opposed to first. Although various measures can and should reduce the demands on an international lender of last resort (ILLR), in the end, we are going to need such a facility.

According to Sachs, the need for an ILLR arises during liquidity crises, which, despite skepticism on the part of some theorists, occur when borrowers cannot obtain short-term funds even when the rate of return on the investment in question would exceed the market cost of capital. Identifying three types of liquidity crises, he pointed first to financial panics which result when short-term debt looms large relative to short-term liquidity, some trigger spooks investors into calling loans, and the borrower cannot refinance. Sachs sees the Mexican and East Asian crises as prime examples of financial panics. The second type of liquidity crisis, a debt overhang, involves a bankrupt debtor in need of working capital. In this country, Chapter 11, section 364 of the U.S. bankruptcy code allows the court to facilitate an otherwise unavailable flow of short-term capital to bankrupt entities. The third type of liquidity crisis stems from the collapse of the public sector, as in a revolution, when the state cannot coll ect taxes or deliver basic public goods. The market will not provide funds during such a collapse, even absent a debt overhang or bank panic.

In the face of such liquidity crises, the ILLR has four functions, Sachs argued. First, the ILLR is meant to forestall panic by its very existence. Then it is supposed to lend into panic, debt overhangs, and public collapse. Nevertheless, Sachs suggested, alternatives to ILLR loans are (or could be) available and may be preferable. For example, controls on short-term capital inflows, prudential limits on bank liabilities, and flexible exchange rates could have prevented recent crises by keeping the ratio of debt to foreign currency reserves from rising to panic-stirring levels. Another alternative to ILLR lending into panic is suspension of payment--the natural solution in the absence of official rescues, Sachs argued. Indeed, as Sachs sees it, banks would roll over their credits to emerging market borrowers more readily if the IMF did not provide the funds that allow them to extract all their assets. He pointed out that, despite the IMF loan package announced in early December, the Korean crisis did not rea lly end until late that month "when the Federal Reserve engineered a roll-over of Korea's short-term debts."

In the case of a debt overhang, the alternative to ILLR lending is a standstill on debt repayments, with a new legal regime to facilitate the flow of working capital to the bankrupt state. A simple statement that the next $100 million tranche gets priority should suffice to start the flow of debtor-in-possession finance, Sachs suggested. Currently, the international system pursues clearly bankrupt debtors for years, until a Brady Plan or HIPC program eventually cancels parts of the debt in a very clumsy fashion. In the third kind of crisis, a public sector collapse, Sachs sees no alternative to ILLR loans.

Sachs went on to argue that the current system handles international liquidity crises very badly. He pointed in particular to excessive use of exchange rate pegs, reckless capital account liberalization, and the way in which the risk-weights used to calculate BIS capital requirements encourage short-term inter-bank lending. He also claimed that the (insufficiently) "big bailouts" used to pay off creditors did not restore confidence, in part because LLR activities require a sensitivity that the IMF seems to lack. Referring to the negative reaction to IMF-inspired bank closures in Indonesia, Sachs suggested that IMF actions often worsened Asia's panic. By contrast, Sachs pointed approvingly to the Fed's role as a crisis manager, rather than a crisis lender, in the rescue of the hedge fund, Long Term Capital Management. In such cases, according to Sachs, the manager calls in the private lenders and says, "Smile, we're rolling over"--to widespread relief that a crisis has been "avoided." Thus, Sachs would encour age finding ways to bail-in the private sector. He also reiterated the need for a regime, akin to Chapter 9 for U.S. municipalities, that would recognize the fiscal insolvency of sovereign governments. He applauded the IMF's newly created facility for lending to countries in conflict or immediate post-conflict circumstances, such as civil war.

In sum, Sachs concluded that the world does need an ILLR, but it does not need a multilateral institution trying to manage 70 countries, as now exists. This unfortunate situation has evolved because the current system does not allow nations to discharge debt, and because recent policies on exchange rate pegs and capital account liberalization have allowed volatile short-term capital flows to "rule the system."

In commenting on Jeffrey Sachs's remarks, Henry Kaufman began by pointing out that LLR issues are much more complex in a global than in a domestic setting, where the beneficiaries are usually commercial banks. He also noted that the new multilateral European Central Bank has very restricted LLR responsibilities and that the IMF was not designed as an LLR--given its limited resources and cumbersome decision-making process.

Thus, starting from a different premise than Sachs, Kaufman also focused on the need to reduce crises through improved supervision and regulation although, he noted, the emphasis on banks may be misplaced in an era of widespread securitization. Kaufman also sought to highlight the need for the major industrial countries to recognize their own responsibilities. He pointed out that the major countries sometimes flood the world with liquidity that seeks a "decent" return in the emerging markets, where borrowers lack the finesse to say no. After the ensuing crisis, investors from the industrial nations offer to "help out," by buying the emerging country's banks and other businesses--at prices far below their levels of four or five years earlier. Noting aspects of unpalatable financial imperialism in this chain of events, Kaufman feared that vested interests might thwart the needed improvements in supervision and regulation in the industrial countries.

Kaufman also flagged financial trends that may have aggravated recent crises. For instance, while increased securitization may give the illusion of improved liquidity (reducing risk aversion), marketability is not the same thing as liquidity. Moreover, the trend toward marking all asset prices to market may have strengthened the harshness of market discipline since marking to market is not a science; the last quote is just an indicator and may not prevail. Ironically, market participants now want to quantify and model risk just as the supervisory authorities want to shift to a more judgmental approach--rightly, according to Kaufman, since risk models are based on historical patterns and cannot cope with developments beyond historic bounds.

All in all, Kaufman saw no need for an ILLR but advocated improved supervision and regulation, starting in the industrial countries. Financial crises affecting lenders in the major industrial countries are more likely to create truly systemic problems than are crises within the emerging markets, he contended.

In his discussion, Jeffrey Frankel began by agreeing with Kaufman that the IMF cannot serve as a traditional LLR because it cannot print dollars or lend freely against good collateral. While the IMF can create SDRs, it can never do so at short notice, and true collateral rarely exists. Although the new Contingent Credit Line is a limited step in the right direction, the world is not ready for a big expansion of Fund resources.

Still, Frankel sees today's IMF as raising many of the issues that surround a more traditional LLR. Citing his Theorem on the Legion Criticisms of the IMF ("For every critique, there exists an equal and opposite critique"), Frankel responded to several pairs of countervailing views and found that, on balance, the IMF/international community reacted appropriately to recent crises. Grouping many criticisms under the umbrella arguments, "The IMF is too generous and creates moral hazard" versus "The IMF is too severe and creates needless recessions," Frankel contended that moral hazard was not the fundamental market failure since current capital/labor ratios suggest the need for larger, not smaller, capital flows to the developing countries. On the other hand, to those who see the IMF as too strict, he pointed out that many crises were largely homegrown.

More specifically, under "markets work best without interference" versus "financial markets work badly," Frankel first acknowledged the existence of contagion. But to those arguing that financial liberalization is dangerous for developing countries, he replied that free capital markets are on balance helpful; like superhighways, they smooth the way but require a cautious approach and, possibly, airbags. Concerning the conditions attached to IMF loan programs (too weak/too strong), Frankel pointed out that critics disagree about what mistakes the crisis countries made; for example, except for the baht, it is not clear that Asian currencies were actually overvalued. Moreover, the downturns that followed the crises largely resulted from devaluation and loss of confidence, not from the IMF response. Finally, Frankel argued that the IMF's evolving emphasis on micro reforms (supervision, governance) as a condition for lending was fully justified. Frankel views conditional IMF lending as preferable to the likely al ternative, bilateral aid with less objective or constructive conditions.

Frankel ended by discussing the uncertain availability of international rescues. He argued that investors could not possibly have counted on a Mexican-style bailout since the U.S. Congress had said "never again," and had ruled out using the Exchange Stabilization Fund in Thailand, while the Senate was refusing to authorize increased IMF funding. But Frankel also claimed that such uncertainty is useful since, like the ambiguity surrounding "too big to fail," it limits moral hazard. In the global context, moreover, the ambiguity is real since no one knows beforehand what the political process will produce.

Policymaking in an Integrated World: From Surveillance to ... ?

Barry Eichengreen began by noting that while recent crises have underscored the need to adjust domestic policies to account for cross-border spillovers, they have also sparked doubts about the efficacy of multilateral surveillance. Moreover, calls for reform show little consensus on how to strengthen this process and avoid and manage crises. In presenting his roadmap, Eichengreen argued that international standards must form the basis for future multilateral surveillance. He also offered suggestions for making IMF oversight and crisis management more effective.

To build his case for standards, Eichengreen posited that the world community has an interest in seeing that all countries participating in global markets adopt minimally acceptable policies on transparency and supervision. In Eichengreen's view, these standards would define agreed principles but allow countries to meet the criteria in ways that reflect structural and cultural differences. Thus, they might mute recent criticism that IMF surveillance has become too micro, invasive, and ill suited to local conditions.

Because the IMF and the G-7's Financial Stability Forum lack adequate expertise in all relevant areas, Eichengreen urged that the private sector be closely involved in developing the standards. But he expects the IMF to play a crucial role in gaining compliance. While self-regulatory groups or private credit rating agencies could possibly issue the timely, realistic reviews of national policies required for market discipline, if they fail to provide them

(as in 1997 Asia), the IMF may have to publish its evaluations. Further, because market participants can become overoptimistic, the IMF may have to use program conditions and differential interest rates on its loans to foster compliance.

Will the IMF be willing to criticize its members? Eichengreen has doubts. Indeed, he believes that IMF policies often serve the political agendas of its dominant members; thus, he recommended giving the IMF more independence by prohibiting its Executive Directors from taking instruction from national governments and by giving them an explicit mandate to foster policies that "maximize stability, prosperity, and growth." He also urged requiring that IMF policymaking be more transparent.

Turning to crisis management, Eichengreen saw two sets of problems requiring IMP loans: country problems and systemic problems. To reduce moral hazard, loans triggered by country problems should be limited to an amount allowing a government to perform its core functions but not to pay off all existing creditors. By contrast, in systemic crises, when adverse external events threaten to destabilize countries with strong policies, the Fund must be able to provide large loans on an emergency basis. He noted that the new Contingent Credit Line is intended to meet this need.

To make Fund efforts to limit its loans and reduce moral hazard credible, Eichengreen argued that the world community must make orderly workouts easier by including collective representation, majority voting, and sharing clauses in debt contracts. But since a first mover problem has emerged, he suggested that the G-7 make collective action clauses a condition for issuing bonds in their markets and that the IMF lend at lower rates to countries adopting them.

Ideally, banks should internalize currency and maturity risks by hedging. But in practice, Eichengreen believes that regulators in emerging markets may have to resort to measures like taxes on capital inflows, in addition to improving bank regulation and opening their capital accounts with caution. Because flexible exchange rates encourage hedging and discourage excessive use of foreign currency credit, Eichengreen also argued that developing countries should eschew explicit exchange rate targets and limit intervention. While he observed that some countries may want to dollarize to gain better access to global capital markets, he concluded that for most of them, currency unions can only be a vision for tomorrow; exchange rate flexibility is the reality of today.

Eichengreen sees little promise in regional surveillance or regional funds to supplement the IMF. While proponents argue that neighbors have the greatest stakes, can exert the strongest pressure, and offer the most relevant policy advice, Eichengreen pointed out that most regions lack Europe's historical taste for integration. In addition, competitive markets for advice may not work as well as competitive markets for good ideas, for the most palatable advice is not necessarily the best. Finally, as the failure of the EMS in 1992-93 demonstrated, strong-currency countries are rarely willing to extend unlimited support to weak-currency neighbors, even in Europe.

In summing up, Eichengreen proposed four central pillars for the new financial architecture: international standards; prudential taxes on capital inflows; greater exchange rate flexibility for most countries; and the inclusion of collective action clauses in loan contracts to create a viable alternative to ever bigger IMF bailouts. He argued that the four are a package. If, for example, the international community does not facilitate loan restructuring, then the IMF cannot credibly refuse to rescue crisis-stricken countries; and if the IMF cannot plausibly limit its loans, emerging market countries will have little incentive to adopt more flexible exchange rates. Thus, he gave priority to gaming G-7 support for all four pillars. An independent IMF is a task for the future.

In commenting, Ralph Bryant noted that the world's political structure has become increasingly complex and confrontational as the number of governing units has grown. Nevertheless, economic interdependence is surging, even while the market--government mix differs greatly across states. Thus, the world's political economy remains at an untidy, intermediate stage of evolution.

Inevitably, Bryant believes, international collective-action problems will grow in importance, forcing nations to shift a wider range of functions to global institutions. For example, to offset market failures, most societies have established domestic institutions of collective governance, with agreed accounting standards, prudential regulation, and limited facilities for crisis lending. If ensuring stable financial markets at the national level requires collective governance, logic suggests that a similar approach may be needed at the global level as well. But, he maintained, the political preconditions for expanding global governance do not yet exist. While the world community is beginning to agree on standard accounting practices, for instance, it does not yet agree on how to monitor their use. Similarly, global institutions with oversight responsibilities remain embryonic, and G-7 coordination is largely undeveloped even though the cumulative impact of their policies is critically important to world welf are. Labeling his preferred approach "pragmatic incrementalism," Bryant encouraged reformers to stretch multilateral cooperation and strengthen global institutions but warned that they must not demand too much too soon.

Characterizing Eichengreen as a fellow pragmatist, Bryant's criticisms concerned Eichengreen's sins of omission, not of commission. First, Bryant suggested, Eichengreen fails to give adequate attention to the role of the advanced nations in the problems that need fixing. Recalling the U.S. savings and loan crisis, and Japan's ongoing problems with weak financial institutions, Bryant suggested that if reformers want to encourage further collective action, they should be careful about focusing blame for inadequate accounting, bankruptcy and oversight procedures on the emerging markets alone.

Bryant suggested that Eichengreen's second sin of omission lies in his restricted view of surveillance, his focus on financial standards and financial supervision. Bryant believes that surveillance must apply to all types of policies, particularly macro policies, particularly in the major countries. For him, encouraging stable, predictable, and mutually consistent macro policies should be the central aim of surveillance. Such a goal requires a far better understanding of the interactions among national economies than currently exists.

Turning to Eichengreen's proposal for making the IMF more independent, Bryant asked why, given the current state of the world's political organization, international institutions should be less, rather than more, accountable to member governments. He also lamented that most reform discussions, even Eichengreen's, fail to note that IMF loans are intended to ease a variety of balance of payments problems, not just those linked to financial crisis, and that not all IMF loans result in moral hazard. Still, he finds Eichengreen's views on the moral hazard in crisis lending persuasive and his suggestions for changes in bond contracts sensible. He joined in calling on the U.S. Treasury to introduce collective action clauses in its own bond contracts or explain why not. As for exchange rate regimes, Bryant expressed strong dissatisfaction with the new convention that most nations have just two options: a free float or an irrevocably fixed peg. Bryant finds the search for the optimal regime misguided, since exchange rate policy is context-dependent. No regime is best for all times and circumstances, even for a single nation.

In his discussion, Vitor Gaspar referred to the extraordinarily large number of reform proposals now afloat and suggested that the ratio of architects to builders has grown too large since, in the world economy, management and enforcement are at least as important as grand designs. Turning to Eichengreen's four pillars, Gaspar agreed with Eichengreen's arguments concerning international standards and collective action clauses. He did, however, have concerns about Eichengreen's comments on exchange rates and Chilean-style taxes on capital inflows, as well as on his proposals for an independent and accountable IMF. While favoring independent and accountable institutions, Gaspar found Eichengreen's suggestions wholly unrealistic for the foreseeable future since IMF independence requires a degree of financial autonomy unlikely to be forthcoming. Moreover, accountability requires a clear mandate, and Gaspar thought Eichengreen's reference to facilitating policies that maximize stability, prosperity, and growth fa r too vague to allow meaningful delegation.

Gaspar agreed that we are likely to see greater exchange rate flexibility, since it is now quite clear that pegs can lead to spectacular crises, but he does not endorse the corner (fully free/fully fixed) solution. No exchange rate regime is intrinsically superior, he argued, and many intermediate arrangements are viable--witness Denmark's experience with an interim solution. Crucially important is the consistency between a country's exchange rate regime, macro policies, and micro structure.

Finally, while Gaspar acknowledged that capital controls, like Chilean-style taxes, can be useful when prudential regulation is inadequate, they can also postpone important reforms. Thus, he reiterated the well-known benefits of financial integration for risk diversification, consumption smoothing, and the efficient allocation of investment funds and urged all countries to strengthen their supervisory capabilities to the point where they can participate fully in global financial markets.

In response, Eichengreen added that he did not intend to exclude the industrial countries from responsibility for improving reporting requirements and dealing with the problems linked to derivatives. He also assumed that the need for surveillance to cover macro policies was entirely uncontroversial-- witness the IMF's new codes on monetary and fiscal policy. As for an independent IMF, he noted that most central banks have a broader mandate than that given the ECB and that the idea was no more quixotic than a world currency or a true lender of last resort.

Priorities in Reforming the International Monetary System

A panel of distinguished policymakers from industrial countries and emerging markets brought the conference to an end. Pedro Pou led off with a discussion of Argentina's interest in dollarization. In trying to identify the root cause of recent crises, he discarded globalization and increased capital flows, observing that, by some measures, the world was more "global" in the late 1800s than now, and capital/labor ratios are far lower in developing than in industrialized countries. Instead, he pointed to the increased use of fiat monies in the 30 years since the end of Bretton Woods.

But, Pou asked, why should every country produce its own money when no one suggests it should make every good? Why not buy money from an efficient producer? Although money is a potent symbol of sovereignty and most nations want an independent monetary policy, most small countries also have a low capital/labor ratio and need an open capital account. But, under that condition, a small economy cannot have an independent monetary policy with either fixed or flexible exchange rates. Moreover, many countries have not yet developed institutions capable of producing a money that is an effective store of value and medium of exchange. Because investors regard these unstable monies with great suspicion, emerging-market borrowers can only borrow short-term in their own currency or long-term in other currencies. Faced with the dangerous mismatches already discussed, "sustainable" conditions can turn unsustainable very fast.

How can these vulnerabilities be reduced? The possibilities, Pou suggested, include capital controls, an improved ILLR, or dollarization. But, for countries short on capital, controls are counterproductive and ineffective, and they promote corruption. Moreover, while an "improved" ILLR might have reduced the risk of contagion and, thus, Argentina's borrowing costs, and while the new Contingent Credit Line shows promise, Pou finds the issues surrounding the ILLR to be so "difficult" as to preclude much near-term progress on that front. That leaves dollarization as the most viable route to stability.

To explain the benefits of dollarization, Pou pointed out that Argentina has had a successful economic program for eight years. Beyond establishing a currency board, it has undertaken substantial reform and now experiences lower inflation and faster productivity growth than the United States. Yet, Argentina still faces large and variable country-risk spreads that reduce investment. At the current pace, it will take another eight to 16 years to eliminate this spread, and, even then, as a small, open economy Argentina could not have an independent monetary policy.

By eliminating currency risk, dollarization would lower interest rates, foster deeper domestic capital markets, boost investment and growth, and reduce government debt service. But, Pou pointed out, dollarization is not a substitute for good policies and is not for everyone. He ended by listing the preconditions for dollarizing, which include the following: a period of exchange rate stability with the dollar; inflation and productivity growth similar to that in the United States; a strong fiscal position and strong financial system with no hidden public liabilities; price and wage flexibility; and as large a ratio of international reserves to currency in circulation as possible, since these reserves will provide the collateral enabling a domestic LLR.

Noting that her comments reflected Indonesia's experiences, Miranda Goeltom emphasized the need to prevent future crises by building stronger public and private financial institutions. She endorsed international standards for both the public and private sectors but cautioned against excessive reliance on market discipline.

Acknowledging the benefits of free capital markets for development, Goeltom noted that capital account liberalization did not lead to an efficient utilization of resources in Indonesia. Indonesia did not follow the (now) accepted order in its liberalization sequence. It freed the capital account before the current account and before developing a strong regulatory system. With hindsight, less haste might have been preferable since inadequate supervision, a tradition of implicit government guarantees, and pervasive weak governance contributed to a serious misallocation of credit and inflated asset prices. Moreover, Indonesia's managed float was actually a fixed-rate regime that encouraged highly leveraged corporations to accumulate large unhedged liabilities in foreign currencies.

Because an open capital account is now the only viable option, Indonesia is taking steps to strengthen its financial system by improving disclosure, transparency, and prudential supervision. In addition to improved risk management within the banks, direct regulation will also play a role, via such measures as limits on open positions and maturity mismatches. Moreover, while capital controls are not a long-run solution, they belong on the menu of policy options. Goeltom also stressed enforcement, including closure of insolvent lenders.

Capital account liberalization requires consistent macroeconomic and exchange rate policies. Indonesia's new central bank law will contribute by making Bank Indonesia independent and narrowing its focus to promoting price stability. Moreover, the newly flexible exchange rate (the rupiah has been floating since August 1997) will underscore the need to hedge, slowing the growth in foreign liabilities.

Goeltom concluded with observations on the international lender of last resort. Indonesians expected that the multilateral rescue would help them regain access to international markets. But the conditions imposed, which included removal of popular subsidies and the closure of 16 banks in a system without deposit insurance, sparked huge public protests that undermined official support and aggravated market doubts about Indonesia's ability to deal with the crisis. Moreover, sizable IMF packages produced modest initial disbursements. Thus, Goeltom suggested that the restructuring of Indonesia's interbank debt did more to restore confidence than the IMF rescue. She remains skeptical about the ability of an ILLR to play a useful role.

Explaining why more significant reform is under way than most observers perceive, Jack Boorman chose IMF surveillance, the new Contingent Credit Line and private sector involvement to illustrate. First, Boorman emphasized the greatly increased scope and strength of IMF surveillance reflected in the new transparency reports, the data dissemination standards, and the codes of good practice on fiscal policy and on monetary and financial policy. While the Fund is far from having a rating system for member performance on these standards, Boorman suggested that the Fund is moving cautiously in that direction. He stressed that future surveillance will be more public than in the past. The first transparency reports (for the United Kingdom, Australia, and Argentina) are already on the IMF's website, and IMF members have agreed to experimental release of the Fund's annual surveillance reports. This increased flow of information--with judgments as well as facts--should go a long way toward meeting private sector demand s for information that would help prevent future crises. But, Boorman noted, whether or not differences in compliance show up in interest rate spreads will be crucial to the effectiveness of these big changes in surveillance policy.

Boorman characterized the new Contingent Credit Line as another step toward the new financial architecture. The eligibility criteria for the CCL include many elements--transparency, standards, private sector involvement--associated with the new architecture. Moreover, the debates on the new financial architecture made the CCL possible, since these discussions helped clarify the conditions under which IMF members are willing to precommit IMF resources. Although the CCL goes only a short way toward addressing the call for an ILLR focused on systemic risk, Boorman believes that the new facility holds promise of dramatic change in how the Fund responds to incipient crises.

Finally, Boorman noted that much discussion at the conference and beyond has focused on the need to involve the private sector in resolving future financial crises. To date, however, few of the proposals for bailing the private sector into crisis resolution have garnered wide support. So far, the IMF Board has only endorsed lending into arrears, but many of the ideas raised by Eichengreen and others are under discussion. While some reformers hope to develop a set of rules for the private sector, Boorman prefers a degree of ambiguity that, with the help of some official financing and some debtor country adjustment, can lead to spontaneous recovery of market confidence.

While Edwards and others argue that all the talk of reform will bring little real change, Boorman is not convinced. At the moment, he stressed, private lenders have the ball. They called for it, saying, "Give us the problem cases, and we will help these countries through." They now have the problem cases--in Pakistan, Ecuador, Ukraine, Romania. If they do not help, sentiment for more substantive reform is likely to grow.

In summarizing the impressions he would take from the conference, E. Gerald Corrigan pointed to the need for humility on the part of policymakers and practitioners. He then set out a list of reminders for reformers to keep in mind. First, despite all the talk about Chilean-style capital controls, in most crises domestic capital is the first to flee. Second, most emerging market countries are very small and very open. Third, a triangle not widely discussed at the conference is the incompatibility of large current account deficits, fixed exchange rates, and weak banking systems; with luck, a country may get away with one or two of those conditions, but it will never get away with all three. And when the current account is part of the problem, the resolution requires domestic contraction; there is no painless alternative. Fifth, while weak banks and a buildup of short-term foreign-currency liabilities are usually at the core of a crisis, the quality of supervision is not the only issue-- consider the U.S. finan cial problems in the 1980s. Still, while Kaufman is right that conditions in the industrial countries contributed to the crises in emerging markets, the crises were largely homegrown. And while bailing in the private sector may be a good idea, like climbing Mount Everest in sneakers, it will be difficult to execute. Finally, since future shocks are inevitable, reform requires a sense of urgency.

Turning to the next steps, Corrigan embraced the form and substance of Bryant's "pragmatic incrementalism." The issue is not architecture, he said, but plumbing and engineering. Designing new institutions would take too long; we must find ways to make existing institutions work better. Citing the lessons of Edwards's paper, he also warned that capital controls represent a slippery slope, postponing needed reform and providing incentives for circumvention. But being a pragmatist, he added, "Never say never." As a case in point, like Jeffrey Sachs, Corrigan endorsed the use of prudential limits on short-term, foreign-currency loans to banks in emerging markets. While the difference between controls and improved supervision is substantive as well as semantic, the crucially important improvements in banking supervision will take time. Finally, noting the importance of fundamentals in determining the viability of any exchange rate regime, he advocated flexibility as the more prudent choice for most emerging marke t countries, at least for now. As for the preferred regime for the industrial countries, Corrigan chose to punt to his good friend, Paul Volcker.

In taking up the challenge, Paul Volcker confessed to a little skepticism as he listened. He was reminded, he said, of George Willis, who was in charge of international finance at the U.S. Treasury when Volcker arrived. In 1971 the world was facing the biggest international financial crisis in 50 years (such a crisis occurs once every 10 years, Volcker noted), and a major international meeting was convened. Whenever the delegates turned to George Willis for his reaction to a proposal, he would growl, "It won't work." Pressed to say what would work, he would respond, "Nothing." Volcker argued that Willis's response remains relevant because financial crises are built into the human genome. When the whole genome is mapped, we will find the genes for greed, fear, and hubris, guaranteeing future financial crises.

Explaining his skepticism about many reform proposals, Volcker suggested that most, including those advocating more responsible economic policies, were more akin to interior decoration than to architecture. The conference outline had noted that "misguided national policies produce harmful spillovers," but, Volcker argued, it could have read "policies produce harmful spillovers," because good policies create harmful spillovers as well. Indeed, the better a country's policies, the more capital it is likely to attract--and the more likely a bubble, eventual collapse, and the conclusion that the country actually had bad policies. He noted that IMF and World Bank documents published in mid 1997 would have made any red-blooded U.S. investor want to invest in Indonesia and Thailand, although the materials did mention structural weaknesses if you read far enough.

If good macro policies will not solve all problems, neither will strong bank supervision. To illustrate, Volcker referred to the United States, the home of strict supervision, and to our experiences with the Texas banks in the eighties, with the largest bank in the country in the nineties, and, in 1998, with LTCM, which required an officially sponsored bailout to avoid the possibility of bringing down the well-supervised, open American banking system.

Having disposed of supervision, transparency, and accounting standards, Volcker then stated as his basic theme that we are facing a broad systemic problem that people have not been willing to recognize. All the arguments about the IMF suggest a certain amount of myopia or self-delusion, he suggested. While IMF actions must be judged on a case by case basis, the real roots of the crisis in global capitalism lie in technological change and national asymmetries. The rapidity with which money can move around the world is widely recognized; less widely recognized are the problems caused by differences in size. Many U.S. banks are larger than the entire banking system in the smallest countries, he noted. And no matter how good their macro policies and how strong their banking systems, these small countries are liable to be inundated by huge capital inflows. Indeed, the more attractive the country, the greater its likely vulnerability.

In self-defense, countries and institutions seek to gain economic weight the only way they can--by joining a bigger, more diversified entity, Volcker pointed out. All of Argentina's big banks are now foreign owned, and foreigners will soon buy up Asian institutions in sales these countries would have resisted just a few years ago. While this process will spur recovery, Volcker agreed with Kaufman that these nations are likely to feel they have been forced to sell their birthright within five years of joining the global financial system. We had better make sure that their troubles are temporary, he urged.

That need brought him to exchange rates. The idea of a small country floating is unworkable, he said; the practice does not occur in the real world, where the instinct to fix is strong. He also noted that an exchange rate is a multilateral phenomenon; you cannot float when others fix or fix when others float. Thus, the exchange rate system requires a coherence that is currently lacking. While the solution for Mexico may be obvious, given its links to this country, the answer for the Asian countries that trade almost equally with Japan, Europe, and the United States is not so clear. Indeed, Volcker concluded, exchange rate arrangements present systemic problems that require a coordinated response.

Summing Up: The Conference Themes

While the conference did not lead to broad proposals for reforming the international monetary system, or even to widespread consensus on many topics, the open and engaged atmosphere suggested an intense effort to understand the problems that, from time to time, confront all countries participating in world markets. The discussion revealed a measure of humility among the participants and a general recognition that the questions raised were more difficult and the answers less obvious than most of us had thought not too long ago. Nevertheless, while systemic solutions remained elusive, the conference participants approached agreement or clarified their differences on several important issues. This summary section outlines the major themes developed at the conference and, by defining areas of disagreement, suggests where economists and policymakers may want to focus their future efforts.

Exchange Rates

Although a few participants claimed that, with good fundamentals, any exchange rate regime becomes feasible, most seemed to believe, with Cooper, that the choice of an exchange rate regime is crucially important and not at all simple. Indeed, Ralph Bryant may have put it best when he said that no exchange rate regime is right under all circumstances and at all times, even for a single country. Still, some arrangements are clearly worse than others, and most attendees accepted that fixed but adjustable rates are a recipe for disaster. Yet many also recognized that the volatility of freely floating rates can create serious problems for small, capital-scarce countries that really need an open capital account. Given that these small, open economies can find it difficult to maintain an independent monetary policy under any exchange rate arrangement, a subset of the participants expressed sympathy for the idea that such countries might want to join a currency union or adopt a dollarization scheme. But since few co untries are currently candidates for such a step, dollarization was generally viewed as a future possibility.

As for the large countries, a few participants argued that even the G-3 may eventually develop a taste for target zones or a single world currency. In making this argument, Cooper cited the currency markets' periodic instability, the probability that financial developments will increasingly determine exchange rates, and the likelihood that national differences will shrink in importance. Finally, as Paul Volcker noted, exchange rates are by definition multilateral, and the lack of coherence in current arrangements represents a systemic problem that will ultimately require a coordinated response.

Capital Controls

Without exception, everyone at the conference acknowledged that open capital markets are crucially important for the efficient allocation of resources and optimal risk-sharing. Thus, as a general rule, and in the long run, free capital markets are to be encouraged. But, given the obvious volatility of short-term capital flows and their role in recent financial crises, most participants now view the issue of capital controls in the context of the optimal sequencing of reforms. As several speakers, including Edwards, Chine, and de Swaan, emphasized, we now recognize, belatedly, that current account liberalization, labor market flexibility, and fiscal prudence are all prerequisites for successful capital account liberalization. Moreover, open capital markets are likely to lead to disaster in countries that lack strong financial supervision and transparent accounting standards. In a world in which the social risk associated with large capital flows to small emerging markets greatly exceeds the private risk, many participants have also come to accept, with varying degrees of reluctance, that temporary capital controls should remain on the menu of policy options. Several individuals suggested that the imposition and removal of such controls must be subject to IMF surveillance.

Surveillance

Most participants embraced the use of internationally agreed standards as the basis for multilateral surveillance, as recommended by Eichengreen. They noted that the development of these codes of good conduct will require input from expert practitioners but that compliance will likely require IMF assessment as well as private sector use of these assessments. A few individuals also stressed that surveillance must cover macro as well as micro issues in large and small countries. The plight of small countries that have sound policies and attract huge capital inflows when world liquidity is ample highlights this need. While many participants were skeptical that the extensive talk of reform would result in anything more than tinkering, a few, including Boorman and Eichengreen, argued that the scope and bite of surveillance has already increased significantly. They emphasized the role of transparency in enhancing the credibility and influence of multilateral institutions.

International Lender of Last Resort

Although a few participants dismiss the need for an ILLR, preferring to rely on the private markets to allocate liquidity, most agreed that, in the end, the world must have an international lender of last resort. Nevertheless, noting the constraints facing the IMF as ILLR and the hard issues surrounding its rescue programs, most attendees would clearly prefer to limit the need for ILLR activities. As helpful in that regard, they suggested improving supervision, transparency, and good governance; ending the use of exchange rate pegs; and imposing capital controls on an emergency basis. In addition, many joined Sachs in advocating that alternatives to IMF loan packages be explored. They would give the private sector more responsibility for crisis resolution, for instance, and develop the IMF's role as crisis manager, rather than as crisis lender. Still, several individuals were concerned that "bail-in" efforts could be destabilizing, and few advocated increased use of standstills on debt repayment.

While the majority emphasized the international community's obvious reluctance to give the IMF, or any international organization, the resources and powers it would need to function as a true ILLR, a few noted that the new Contingent Credit Line shows promise of representing a significant step in that direction. Finally, while most participants endorsed Bryant's "pragmatic incrementalism" and resisted pushing too hard for reforms the world is not yet ready to accept, a few saw a growing need to create institutions of collective governance to offset market failures in the international arena. Despite his warning, Bryant encouraged efforts to stretch and strengthen multilateral cooperation, while Keohane predicted that democracies will want to expand international government to balance the influence of the mobile and immobile factors of production.

In the end, thus, the participants left with a full agenda for design and implementation work in each of the key areas covered. Overall, developing the political will to proceed appears the main challenge.

Jane Sneddon Little and Giovanni P. Olivei Assistant Vice President and Economist, and Economist, respectively, Federal Reserve Bank of Boston.

COPYRIGHT 1999 Federal Reserve Bank of Boston
COPYRIGHT 2004 Gale Group