The final afternoon began with a session on the role of IMF surveillance and, more specifically, on the functioning of the international monetary system. Two of the featured speakers had served in the late 1980s as deputies to their country’s finance minister for the international economic cooperation and surveillance activities of the Group of Seven: Canada’s Wendy Dobson and Japan’s Toyoo Gyohten. Ms. Dobson subsequently became Professor of Economics at the University of Toronto and authored a book on international economic cooperation for the Institute for International Economics; Mr. Gyohten became Chairman of the Bank of Tokyo and co-authored a book on international cooperation with Paul Volcker. The third speaker, Jacob Frenkel, participated in the work of the Group of Seven in his capacity as Economic Counsellor and Director of Research at the IMF before leaving that post to become Governor of the Bank of Israel. The session was chaired by Maria Schaumayer, Governor of the Austrian National Bank.


Opening the session, Maria Schaumayer remarked that the issues to be addressed included the role of Fund surveillance in overseeing policy formation, the role of exchange rate policy in promoting sustainable economic growth, and the implications of the liberalization and international expansion of private financial markets. It was clear that it was neither possible nor desirable to return to a global system of fixed exchange rates, however defined. It was also clear that the free flow of private capital put a premium on credible monetary and economic policies. By the same token, under liberalized and globalized conditions, there was a heightened awareness of the need for consistent and flexible economic policies.

In her view, the future surveillance task of the Fund might usefully be oriented toward a “flexible stability strategy”: activating national commitment, providing in-depth research capacity and advice on policy formation, and offering guidance on the execution of formulated policy commitments. The world economy was facing new challenges, and the Fund should—and could—put greater emphasis on its qualitative role, advising members on the best policy mix to achieve or safeguard orderly conditions. Sound policies were not themselves the aim, but experience had shown that they were necessary to promote sustainable growth and prevent financial crises, which increasingly had a propensity to become global.

Wendy Dobson

…the essential purpose of the international monetary system is to provide a framework that facilitates the exchange of goods, services, and capital among countries, and … sustains sound economic growth … a principal objective is the continuing development of the orderly underlying conditions that are necessary for financial and economic stability….

IMF Articles of Agreement, Article IV, Section 1

These words in the IMF Articles of Agreement provide a vision for the IMF’s surveillance role—a role that continues to be appropriate today.1 Yet the vision of a world of efficient markets, sustained growth, and financial stability has not been realized. Fifty years after the Bretton Woods agreement, the institutions exist in a paradoxical and volatile world. It is one in which the growing interdependence of national economies implies the need for effective surveillance. Yet the world is also one in which the significance and credibility of national governments and intergovernmental institutions appear to be declining before a proliferation of interest groups at home and multinational firms and financial institutions pursuing their interests abroad.

The effect of these paradoxes is apparent in three aspects of surveillance on which I will focus. First, the preconditions for successful surveillance and policy coordination are increasingly difficult to meet because of the role that domestic interest groups play, particularly in fiscal policy formation. Second, the perennial debate about exchange rate arrangements needs to be recast in the light of developments in international business and finance. Third, capital flows, freed by regulatory changes and technological innovation, are now so large and swift that markets, rather than international financial institutions, have become a major source of discipline on domestic policies. Yet surveillance still has a key role to play in providing advice on good policy that will sustain economic growth and promote stability; surveillance also still faces a challenge to adapt the process to take advantage of, rather than to be at the mercy of, interest groups.

Let us step back briefly to consider some of the changes in the world economy that affect the transmission of economic policies across national boundaries and, therefore, the functioning of the international trade and financial systems. One of the outstanding characteristics of many of the changes is that they are market driven. As governments have made way for the private sector to be the engine of growth, firms have gained clout as international players.

New industrial competitors have emerged from low-cost industrial bases in middle-income developing countries and the former centrally planned economies. Business strategies of multinational firms based in the industrial countries to meet the competition, to take advantage of production cost differentials, to supply goods and services in these burgeoning markets, and to overcome trade barriers and exchange rate uncertainties have contributed to surges of foreign direct investment that are deepening the links across borders. According to the UN, there are now 37,000 multinational firms with 200,000 foreign affiliates. Sales generated by foreign affiliates totaled nearly $5 trillion in 1991, more than the world’s total exports of goods and services (United Nations Conference on Trade and Development, 1994, pp. 4–7). These flows of direct investment reduce the independence of governments by tying the world’s economies ever closer together. Paradoxically, however, such flows are a manifestation of the increased independence of multinational firms. With business units in several locations, they are better able to insulate themselves from exchange rate volatility than if they existed in only one location.

A second and similar paradox is that the ability of the international monetary system to provide a stable framework is reduced by the size and volatility of private capital flows and the emergence of multiple regional focuses of economic activity and of multiple currencies. At the same time, however, the emergence of sophisticated financial management techniques, such as hedging, is increasing the ability of firms to insulate themselves from financial market volatility.

In this context, there are several lessons to be learned from experience with surveillance in the postwar period.

Surveillance and Policy Coordination in the Group of Seven

Surveillance, or policy consultation, was essential to the functioning of the par value system. It was intended that IMF oversight would ensure that domestic policies conformed with negotiated rules. When the par value system collapsed, the surveillance process was continued in a more informal way; rules were replaced by consultation and peer pressure to monitor performance and to encourage remedial measures judged necessary to adhere to shared values of fiscal responsibility and monetary stability. In this informal system, the IMF required domestic allies to promote desirable remedial action; the large and influential players had to be willing to take the lead in adhering to shared values. If they refused, or reneged, the moral authority of the informal system would be eroded. Subsequently, the large players formed the Group of Five and then the Group of Seven, which, in the second half of the 1980s, attempted closer coordination of policies to achieve greater exchange rate stability than had prevailed between the breakdown of the Bretton Woods system and the 1985 Plaza accord. The IMF was invited to participate in this initiative in a limited way.

Theoretical work on coordination indicated that there were gains to be made, providing certain preconditions existed: that participants agreed on their goals; that they had an agreed institutional framework; and that policymakers could deliver commitments they made to their partners. This early work also indicated that the gains from coordination—even if perfectly implemented—would be small relative to the collective gains from countries individually pursuing good, rather than bad, policies.

Beyond these issues, participants must agree to participate in surveillance of economic performance and policies. Surveillance requires agreed objective benchmarks against which to evaluate performance and policies and to define when corrective action is required and what that action might be. Policy bargains must be identified, consisting of policy changes that provide net benefits to the participants. Finally, sanctions are required for failure to carry out commitments.

Recent evidence indicates that few of these preconditions were met in reality. The goal of Group of Seven policy coordination in the 1980s was to restore sustainable world economic growth by reducing the obstacles to growth posed by large external imbalances. Yet attempts to coordinate were undermined by differing views among participants about how the world economy works. Perhaps more important, these attempts were undermined by differing views among participants in the same country; by erroneous forecasts that undermined the credibility of coordination, such as occurred after the 1978 Bonn summit and following the stock market crisis in 1987; by asymmetry in policy commitments; by such institutional factors as lack of continuity among major players and lack of accountability or effective sanctions; and by political factors, such as breaches of confidentiality that revealed differing interpretations of goals and commitments.2 And, as at this conference, disagreements persisted over how to coordinate. Was the exchange rate to be the point of departure for analyzing the consistency of macroeconomic policies as in the Bretton Woods regime or the European Monetary System (EMS)? Or was the objective to adopt consistent macroeconomic policies that would ensure exchange rate stability?

In purely practical terms, the starting points for the fixed and flexible exchange rate approaches will differ, but the end point should be the same: if there is pressure on exchange rates, eventually there must be agreement to intervene in foreign exchange markets and, if necessary, to change monetary and fiscal policies. Much of the practical problem arises in reaching agreement on the circumstances that require a change in exchange rates, since currency markets are not fully rational and currencies move in response to financial as well as real factors.

Exchange Rate Policies

The fixed but adjustable exchange rate regime was the cornerstone of the Bretton Woods system. Disappointment with flexible exchange rates since the collapse of the Bretton Woods regime has been widespread. Costs of exchange rate variability are considered to be high, as illustrated in the recent report of the Bretton Woods Commission (1994, p. A-4):

The Commission believes that the costs of extreme exchange rate misalignment and volatility are high. When current exchange rates are misaligned, resources are misallocated; when exchange rates are unduly volatile, it creates uncertainty and productive investments are inhibited. Exchange rate misalignment adds to protectionist pressures from vulnerable industries in one major country after another as their international competitiveness waxes and wanes. Exchange rate volatility can be hedged away in today’s sophisticated financial markets, but hedging involves costs, not all risks can be hedged, and not all hedges are perfect.

Since the early 1970s, long-term growth in the major industrial countries has been cut in half, from about 5 percent a year to about 2.5 percent a year. Although many factors contributed to this decline in different countries at different times, low growth has been an international problem, and the loss of exchange rate discipline has played a part.

The underlying concern is that the floating exchange rate regime fails to provide an anchor for expectations of future exchange rates, thereby amplifying volatility and undermining capital formation and economic growth.

Rather than extending these familiar arguments, I have chosen to examine more closely evidence of the costs of exchange rate uncertainty. Evidence of trends in production, trade, and investment suggests that exchange rates may have less impact on the real economy than is commonly assumed. The validity of this assumption is important because in a rapidly changing world economy, the fixed or managed exchange rate alternative imposes costs of its own, since other channels of adjustment must be found.

We begin with evidence at an aggregated macroeconomic level and then move to business decision making. It is true that we have seen misalignments in major currency relationships—both within and outside the EMS system of fixed but adjustable rates—since the collapse of the Bretton Woods par value system: most recently, the U.S. dollar against both the yen and the deutsche mark in the early 1980s. Since 1987, however, a fair degree of stability has existed in relationships between the dollar, yen, and deutsche mark, as nominal relationships in the Chart show. In the past seven years (1987–94), deviations from the nominal means of the yen/dollar and deutsche mark/dollar exchange rates are symmetrical and have been narrowing.3 In addition, growth of world trade and investment has outstripped the growth of world incomes since 1983 (United Nations Conference on Trade and Development, 1991, p. 4).

More specific microeconomic evidence of the real costs of exchange rate variability is inconclusive. An extensive review of available empirical evidence published by Edison and Melvin (1990) failed to find a conclusive impact on trade flows or harm to international investors. Indeed, the authors suggested international investors may actually prefer the distribution of returns under floating rates to that realized under fixed rates. More recently, the search has focused on the relationship between exchange rate uncertainty and investment. New theoretical work by Dixit and Pindyck (1994) on investment decisions under uncertainty shows that the timing of real investment is affected by an increase in uncertainty. In marginal investment decisions, uncertainty raises the required return on an investment. We do not, however, know what uncertainty will do to the average realized return on investment in the long run, since uncertainty can mean an increased probability of higher, as well as lower, returns. Dixit and Pindyck observe that it does not follow that countries with greater economic volatility will have lower ratios of investment to GDP. And indeed, it can be argued that in some cases firms facing an increase in volatility may reduce, not increase, waiting time on investments if the increased volatility raises the expected profitability of those investments.

Another important dimension of the debate about exchange rate arrangements is the impact of exchange rate variability on business decision making. Do firms find nominal exchange rate flexibility a burden in their decision making? Or can they cope? Little systematic evidence beyond the costs and benefits of hedging is available on the net effect of exchange rate arrangements on long- and short-term business plans. As Caves (1989) has pointed out, one of the reasons is that the theory of money neutrality implies that there is no real influence in the long run of a monetary variable such as the exchange rate. Nor are statistics readily available to study this aspect of behavior at the level of the firm.


Nominal Deutsche Mark-Dollar and Yen-Dollar Exchange Rates

(Weekly averages)

Source: Wharton Econometrics Forecasting Associates.

To throw some light on this issue, I sought the views of nearly a dozen chief executive officers, corporate treasurers, and strategic planners in North American, European, and Japanese multinationals on their preferences and experiences with fixed and flexible exchange rate arrangements. Not surprisingly, they expressed a general preference for “flexible stability.” Beyond that lay some interesting insights. Fixed and flexible exchange rate arrangements were viewed in terms of their impacts on short-term business behavior, such as trade, and on long-term direct investment decisions.

Beginning with fixed exchange rates, the perceived advantages included such factors as the absence of costs of currency hedging and the ability to set prices independently from monetary considerations. But the costs were reported to be considerable. Fixed exchange rates induce two illusions: one of the permanency of this key price, and another that it is a correct signal of purchasing power parities between any two economies. These illusions raise the risks of distorted price signals associated with misalignment, risks that are seen to be considerable, as investments made on the basis of false information can be disastrously wrong and difficult to change when currency revaluation eventually occurs. One of the issues here is the expressed lack of confidence in the authorities to make exchange rate corrections on a timely basis. As one chief executive officer put it, “I would prefer fixed exchange rates if I could be confident the authorities would change policies when this becomes necessary. But I do not have this confidence.” The real implication of such distortions is that restructuring to raise productivity is deferred, and opportunities to relocate production in lower-cost locations are missed, with resultant loss of international market share and volume.

With flexible exchange rate arrangements, short-term decisions can be hedged. Hedging is costly, transferring profits to financial institutions in proportion to the magnitude of volatility. But the long-term implications are more favorable because corporations must form their own judgments about the overall economic environment and trends therein. With such a long-term view, they tend to stay more flexible and adaptable (since nothing can be taken for granted) than they do when exchange rates are fixed.

What about the role of exchange rates in longer-term decisions like foreign direct investment? Anecdotal evidence has its limitations, but, in general, exchange rate variability was not seen as a major factor in decisions on foreign direct investment. Rather, as in domestic investment decisions, what matters more are the prospects of future demand and the strategies of competitors. Foreign direct investment flows thus tend to vary with the business cycle. Although exchange rate variability adds to risks, it is only one factor among many considered, along with political and policy risks, such as changes in tax laws.

A further refinement refers to goods and services; exchange variability matters more when one is producing in one area and selling in another, which often happens with production of goods. In service industries, costs and revenues tend to be more closely matched in the host market.

It follows that exchange rate variability matters least to large multinationals. They can reduce many of the costs of hedging by internal matching. They can also locate production on both sides of major exchange rate relationships. Exchange rate variability matters more to small and medium-sized businesses, which have fewer financial options open to them and lack the experience, managerial capacity, and financial resources to anticipate and endure sudden changes in relative prices.

Other recent studies of the determinants of foreign direct investment are also instructive. Graham and Krugman (1993) searched for reasons for surges of direct investment such as have occurred periodically since the 1960s. Many observers have suggested that exchange rate changes, particularly the realignment of the yen and the dollar in the mid-1980s, triggered these surges. Graham and Krugman conclude that no single explanation is sufficient. Changes in corporate borrowing capacity and the availability of internally generated funds played a role, but follow-the-leader corporate behavior was also a factor, as was the existence of trade barriers and tax changes.

Other analysts have argued that yen appreciation in the mid-1980s triggered the surge of Japanese foreign direct investment into East Asia. Some argue that Japanese producers were forced offshore, which caused an undesirable “hollowing out” of Japanese manufacturing (Ohno, 1994). Could it be, however, that the surge of outward Japanese investment was inevitable because of Japan’s labor shortage and because of anticipated income and market growth in nearby Asian economies? Could it be that changes in relative prices resulting from exchange rate revaluation accelerated industrial restructuring and contributed to increased global welfare? The transfer offshore of labor-intensive operations has increased the overall efficiency and productivity of the Japanese economy. Flows of foreign direct investment into the Asian economies also have brought about transfers of technology and skills to the host economies and contributed to the diversification of their economic structures, creating higher-paying manufacturing jobs to replace jobs in traditional agricultural and service sectors.

The foregoing is a necessarily abbreviated discussion of the relationship between exchange rate variability and economic growth. The key point is that the evidence that exchange rate variability, per se, undermines economic growth is not conclusive. Indeed, recent IMF studies show how such a conclusion could be wrong (International Monetary Fund, 1994). Should we then be so certain that the world economy would be better off with managed or fixed rates? Perhaps more important is the fact that both growth prospects and exchange rate variability are affected by macroeconomic policies, and it is the challenge of achieving prudent policies on which our attention should focus.

The key concern for business and policy decision makers is that misalignments can occur with either floating or managed arrangements. When exchange rates float, the market can get it wrong. But when exchange rates are managed, the authorities can get it wrong. The difference is that when things are left to the market, firms are better prepared for uncertainty. As important, we have not eliminated uncertainty with managed rates; we have simply replaced the large probability of frequent small movements with a smaller probability of large movements. In addition, when we fix a major price like the exchange rate, we must find flexibility somewhere else in the system to ensure that adjustment occurs. Resorting to flexibility in the real economy through greater adjustment in labor markets or industrial and spatial distribution of economic activity is not easily obtained, as European experience attests; looking for flexibility in fiscal policy presents an even greater challenge, as the recent Group of Seven experience attests.

This skeptical view of the costs of exchange rate variability is not meant, however, to be a defense of the status quo. Rather, the skepticism relates to an appreciation of the costs of abandoning exchange rate flexibility in a world of rapid change in which the ability of governments and central banks to act quickly and decisively is waning. A system of generally floating rates among the major countries can help to cope with divergent cyclical positions and with real shocks. While major exchange rate movements have occurred, there is evidence that most of these movements could have been avoided if individual countries had followed sound fiscal policies. The key challenge is to find ways to encourage pursuit of good policy to anchor expectations of future exchange rates.

Implications of Financial Market Developments

Good policy becomes ever more important because of developments in financial markets. Today, financial markets are increasingly integrated and deregulated; one of the concerns for surveillance and the international monetary system is that although part of these capital flows represents investment in wealth-creating productive capacity, speculative activities have increased enormously, and skepticism of the ability of national authorities to maintain orderly markets has also increased.

Several major developments lie behind the rapid expansion of capital flows since the late 1960s. One is rapid developments in microcomputing and telecommunications technologies. Another is financial deregulation in the major countries and innovations in the supply of financial instruments. Another is the growth of investment demand and related developments in financial markets in China, India, and the middle-income developing countries.

These developments imply several issues for the future. The first issue is the prospect of a global shortage of capital as world growth momentum picks up in the mid-1990s. While industrial countries will grow at annual average rates of 2 percent, the developing economies are growing at average rates that are more than twice as fast. Huge investment requirements in infrastructure and enterprises are forecast; the IMF expects that investment in all developing countries from domestic and foreign sources will rise to 27 percent of GDP in the years ahead (International Monetary Fund, 1993). These countries, because of their current status of capital scarcity, will be attractive hosts for equity and portfolio flows from capital-intensive developed countries. But while some of this capital will be sourced in the industrial countries, many of these countries are struggling with heavy public sector indebtedness. Thus many of these savings requirements will have to be generated locally and channeled through nascent domestic capital markets. Equity and bond markets will have to be developed that are reliable and safe, and these markets, in turn, will have to be integrated into the global system of capital markets.

The second issue, in established capital markets among the industrial countries, is a different one. In these markets, much of the capital flows that influence exchange rate movements is speculative and moves in volumes that far outweigh capital flows that actually allocate the world’s savings. Private speculators operating through the so-called hedge funds, which invest borrowed funds in high-risk, often highly speculative activities, distort exchange rate signals that influence trade, direct investment, and long-term portfolio investment.

These developments imply reduced macroeconomic policy autonomy for national authorities because of the magnitude and speed of capital flows. Central banks and international financial institutions like the IMF collectively lack the resources to offset speculative bets on weak currencies. Some senior observers have called for intervention to regulate and moderate such flows to restore a degree of national autonomy and to lengthen markets’ time horizons. James Tobin’s call for a tax on international currency transactions is one such proposal (for a recent statement, see Tobin, 1994). Henry Kaufman’s call for intervention to lengthen the commitment of mutual fund investors is another (Kaufman, 1994).

But the deeper implication is that those who allocate private capital for speculative purposes are no longer particularly impressed by exchange market signals. The weakness during 1994 of the U.S. dollar against the Japanese yen is a case in point. Some participants believe the alignment of the two currencies is about right. Others point to two expectational factors about underlying policies to explain the persistent weakness of the dollar. One factor is nervousness about the possibility (however unlikely in fact) that the U.S. Administration might depreciate the dollar in reaction to the intractability of U.S.-Japanese trade frictions rather than move to correct the underlying saving-investment imbalance. The other factor is inflation expectations. The heavy public sector indebtedness of a number of industrial countries, including the United States, is a continuing concern. Although more industrial country central banks are gaining independence, financial market participants share a deep suspicion that governments will succumb eventually to the temptation to reduce the real value of their indebtedness by reflating. To anchor expectations, a credible combination of concerted intervention, coordinated interest rate changes, and more action on fiscal consolidation would probably be necessary.

Implications for Surveillance

What are the implications of these observations for surveillance and the international monetary system? I have argued that because market participants have more clout than national authorities, stability of exchange rates must be anchored in prudent and flexible domestic policies and in credible arrangements for international cooperation. I have also emphasized that because domestic interest groups are increasingly influential in democratic decision making, these arrangements should operate in ways that promote coalitions of interest between international financial institutions and domestic allies. The challenge for the future, therefore, is to redesign the IMF’s system of surveillance to increase the credibility and weight of international peer pressure for sound domestic policies. I see six practical attributes of a future system that recognizes these realities and that observes the preconditions for successful cooperation identified at the beginning of this paper: agreed goals, institutional framework, and ability to deliver.

Surveillance Should Be Formalized Within the Fund

One of the main practical goals of a system of surveillance should be to promote among senior government officials the awareness of, and commitment to, good domestic policy. This important function should be taken out of the Group of Seven economic summit framework and rolled into a restructured IMF.4 Private capital markets are now so prominent that the IMF has a diminishing number of instruments at its disposal to require policy changes, and these instruments tend to be most effective in managing rather than averting crises. Much, therefore, depends on the quality of debates and opportunities for exercising peer pressure. Treasury ministers would still participate in summit meetings, but on an accountability, rather than on a decision-making, basis (exchange rate decisions are almost always excluded from summit discussions anyway). Depoliticization in this way would also strengthen the case for greater central bank and treasury involvement in surveillance discussions. To encourage governments to accord the process the necessary credibility, surveillance discussions—as the Bretton Woods Commission has noted—should be reduced in frequency to ensure that the most senior officials with direct responsibility for policy formation and implementation participate, as well as carry out their responsibilities in capitals.

Exchange Rate Arrangements and Surveillance

As I have argued, the focus of surveillance should be good policy. Time and resources should be spent on understanding fundamentals. This is not to deny that exchange rates do move in ways that do not reflect economic fundamentals. Such circumstances, however, are difficult to recognize, especially during periods of rapid structural change. As Peter Kenen (1994, p. C-15) has pointed out:

We return ineluctably to the quest for a system of stable but adjustable exchange rates—one that will preclude or minimize the risk of large swings in real exchange rates due to interactions between policy mistakes and volatile expectations, but one that will not lead to the ossification of nominal exchange rates characteristic of the Bretton Woods System and of the European Monetary System in the late 1980s.

Kenen’s goal is a sensible one, which takes account of the decision-making constraints that democratic governments face. Several proposals exist for stable but adjustable exchange rates, including Kenen’s pragmatic approach. Another is Bergsten and Williamson’s (1994, pp. C21–29) target zone proposal. As I have argued elsewhere, sophisticated proposals like Bergsten and Williamson’s require levels of technical and operational involvement by national decision makers that it is not yet practical to achieve. In addition, attempts to maintain target zones would require a degree of linkage among monetary policies that does not exist. Indeed, current evidence suggests the opposite: national authorities accord greater weight to domestic than to international objectives in the conduct of monetary policy. Thus, we face a situation where exchange market intervention will continue to be ad hoc; where policy changes to restore order are possible but will also be undertaken on an ad hoc basis. In these circumstances, the quality of analysis and advice will be a critical factor in achieving consensus to act.

Surveillance Should Be Backed by Credible Analysis and Advice

The quality, credibility, and accuracy of policy advice will matter very much. A high priority of a future system of surveillance should therefore be to improve the basics of analytic capability, continuity, and accountability. The IMF has a respectable track record on this score. The credibility of its advice and its ability to act as an “objective referee” will be among the few levers it has to encourage good policy and performance in the large economies. Its analysis should also make greater use of scenarios that demonstrate the consequences of adopting or continuing undesirable domestic policies.

While much emphasis will be placed on policy analysis and discussions, exchange market developments will also be analyzed, but on a confidential basis. Actions to restore order in the markets involving the three major currencies would be even more confidential and directly involve the responsible authorities.

Surveillance Results Should Be Publicized

The Fund currently provides its technical and analytical advice on a confidential basis. Although confidentiality is necessary in exchange rate matters, when it comes to macroeconomic and structural policies, it may be a costly luxury. Informal surveillance requires domestic allies drawn from those groups that accept international responsibilities. When IMF prescriptions are kept confidential, it is difficult to build coalitions supportive of good policies. Surely, the time has come to make public the results of Article IV consultations. It has been argued that publicity can introduce moral hazard into the process because national authorities will hesitate to be candid. But it is possible to envisage that the prospects of publicizing Article IV consultations would have the opposite effect—especially when national authorities recognize that policies that contribute to international balance are also good domestic policies in the long run.

An alternative way to address this issue is for the Managing Director to create an independent global economic panel of wise people acting in their personal capacities to carry out analysis and issue public assessments of, and prescriptions for, desirable policies and performance among the Group of Seven countries. This mechanism would enhance the Managing Director’s clout, help to depoliticize forecasts, and augment peer pressure with public pressure, possibly after a waiting period following confidential consultations.

Participation Should Be Broadened

Delegating surveillance to the IMF, reducing its secrecy, and increasing its professionalism are ways to increase the credibility of its policy advice. But surveillance also suffers at present from insider-outsider tensions. Since the world’s largest economies and most dynamic markets include East Asian economies that are outside the Group of Seven and that are already developing cooperative mechanisms among themselves in the Asia Pacific Economic Cooperation (APEC) forum, the IMF should respond to these developments by creating mechanisms for including them more extensively in surveillance and in peer review. As participants in the Group of Seven have found, informality and frankness are the most valued aspects of the process, yet these attributes tend to decline as the size of the group increases.

The process should have two stages: one, which is decentralized, for consultation among groups of neighboring countries; and a second one at the global level for decision making. The constituency structure should be revised to promote consultations structured around two evolving groups: the Europeans (including the emerging economies of Central and Eastern Europe and Africa) and the APEC group (the United States and Japan and including Canada, Mexico, Latin America, China, and the other East Asian economies). Although the danger is that such structures increase regional awareness at the expense of the world economy, network structures of this kind would allow for more involvement and commitment to global consultation and policy coordination.

Crisis Management

One 30-year veteran has observed that the challenge is to create structures that are small enough to take action but large enough to involve all that wish to be involved. In order to take decisions or to act collectively in a crisis, the surveillance body would probably require an executive committee composed of the largest economies. At present, this body would be the Group of Seven, but provision will have to be made for it to evolve to include China, India, and other large economies that emerge as market-based economies in the next quarter century. If such a formula is not found, informal structures, operating outside the Fund, will inevitably spring up to fill the gap.


I believe I can conclude on a positive note. We are moving in the right direction, not away from it. National authorities have learned important lessons in the past 20 years. Central bankers know the value of price stability and are determined to contain and reduce cyclical inflationary pressures even if they are not ready to link monetary policies. Fiscal authorities are learning that, even though they would like to believe that fiscal policy is one of the last bastions of national sovereignty, persistence with bad domestic policies will eventually be disciplined by financial markets; such discipline provides much rougher justice than would peer pressure and cooperative action applied in preventive ways.

To adapt to the realities of a world in which the significance of national authorities is declining before interest groups and market participants, the political will in the large countries to cooperate is a necessary condition for successful management of the international monetary system. Such cooperation will be possible only when these countries are willing to adopt domestic monetary and fiscal policies that contribute to international balance. Unless they do this credibly, market participants will test their resolve, and the case for a more sophisticated system, in which exchange rates are managed on a systematic basis, will remain an elusive one.

Toyoo Gyohten

Surveillance is not an invention of the 1980s and 1990s. Since the 1950s, it has been the main instrument to ensure the effective functioning of the international monetary system. Under the Bretton Woods regime, surveillance was conducted by the Fund as the key operation to restore international equilibrium and thereby ensure the stable functioning of the adjustable peg system of exchange rates; when a member country ran a serious external deficit and its currency came under pressure, the Fund conducted surveillance and—as it does now—urged the country to adopt the macroeconomic policies necessary to rectify the situation.

Until the early 1970s, because most countries except the United States were still vulnerable in their balance of payments positions and the financial support of the Fund was critically important, it was natural that member countries felt obliged to comply with the suggestions and the advice given by the Fund. In other words, a country in a state of imbalance agreed to give priority to the stability of the exchange rate over short-term domestic considerations.

Since that time, however, a series of events has taken place, such as the collapse of the Bretton Woods exchange rate regime, the important development of the EMS, and the end of the Cold War. These events brought about fundamental changes that altered the environment in which surveillance was expected to be performed. What were these changes? In my view, there were four

First, the maintenance of global exchange rate stability, namely, the maintenance of the adjustable peg system, ceased to be the main objective of surveillance; we lost the clear target for the exercise.

Second, the international economic balance of power changed dramatically. Particularly, the position of surplus and deficit countries was reversed, with the United States becoming a major deficit country while other industrial countries and some newly industrializing countries became surplus countries. This change brought about a new problem. We now have to redefine how the burden of adjustment should be borne between surplus and deficit countries.

In the heyday of the Bretton Woods regime, with the United States virtually the only surplus country and with its overwhelming economic power, we took it for granted that the main burden of adjustment should be borne by deficit countries. However, what we see today is quite a different situation. Although the United States has become a major deficit country, the general tendency is to ask the surplus countries to bear the burden of adjustment because of the “benign neglect” policy of the United States and the global concern about slower growth and higher unemployment. This asymmetry, in my view, is a major cause of confusion for the surveillance process, particularly among major industrial countries.

Third, the U.S. dollar’s predominance as the key international currency, with guaranteed value, has been eroded. The international monetary system has lost its anchor. Major currencies are floating against each other. Therefore, when we talk about exchange rate stability as a main objective of surveillance, we need to redefine what we mean by exchange rate stability.

Fourth, international financial markets have grown enormously in size, breadth, and fluidity. As a result, the monetary authorities’ power to control the market has been seriously undermined. In that sense, the surveillance exercise among authorities is now faced with a new challenge, because surveillance was intended to achieve its objective through macroeconomic and exchange rate policies conducted by the authorities.

In these new circumstances, our first task is to redefine the purpose of surveillance. In order to argue that surveillance is still necessary in the new environment, we need to make the following three assumptions.

First, all countries want to maintain noninflationary and sustainable growth so that they can ensure a steady improvement of their people’s welfare and contain unemployment. This assumption is a very valid one, and I think we can endorse it without reservation.

The second assumption is that all countries want to maintain the free flow of goods, services, and capital. I am not so sure about this second assumption, as we see some ominous signs of protectionism looming in different corners of the world. But, in a broader context, I think there is agreement on this point as well, at least in principle.

The third assumption is that all countries agree that excessive volatility and prolonged misalignment of exchange rates should be prevented. Again, I am not so sure that this assumption is valid. As we learned from the interesting exchange of views between Fred Bergsten and Stan Fischer yesterday, and as Wendy Dobson has just pointed out, there are different views on this point. But again, I think that there is a broad consensus that this assumption is accepted in principle.

If these three assumptions are endorsed, the purpose of surveillance in the new environment should be, first, to try to establish reasonable balance in each national economy so that serious international imbalances can be avoided and, second, to prevent systemic risk to the international financial market. As was the case in the aftermath of Black Monday in October 1987, there is no doubt that we need some mechanism to conduct surveillance rapidly and agree on prompt action to cope with such crises. Third, surveillance will be needed to cope with the excessive volatility and prolonged misalignment of exchange rates. The mechanism of surveillance for that purpose should be a kind of crisis management. It could well be different from the surveillance mechanism for the first and second purposes I mentioned.

It is a difficult practical problem to decide on the nature and number of mechanisms or forums we should establish to conduct surveillance for those purposes. In any case, I believe the Fund needs to perform the central role as a strong leader and a trusted umpire. In a broad context, we need three different forums: broad multilateral surveillance, within the framework of the Fund in the form of meetings of either Executive Directors or something like the Interim Committee; an improved Group of Seven or summit-type forum that will conduct surveillance for the largest, most important economies; and a forum established specifically for crisis management of exchange rates and financial markets.

In any of these forums, I believe that three basic requirements need to be met for the successful performance of surveillance. First, all participants must acknowledge the benefit of being a member of the surveillance mechanism. This may sound obvious, but it is an important starting point. Second, all participants must be prepared to accept surveillance and give it priority over short-term national policy objectives; in other words, they need to accept the disciplinary feature of surveillance. Third, recommendations produced as a result of the surveillance exercise should be at least intellectually convincing and should be practicable, provided that there is an adequate amount of political will.

Unfortunately, these assumptions and conditions are not fully endorsed at present. Therefore, we cannot take it for granted that, in today’s circumstances, surveillance can function as the main instrument to support a stable international monetary system as it did in the heyday of the Bretton Woods regime.

Having said that, I do not want to sound too cynical or pessimistic, because as long as we agree on the need for closer international policy coordination for the global benefit—and I believe we do—we need to encourage efforts to conduct surveillance in a realistic way as an important and effective means of building mutual confidence and exerting peer pressure. In fact, we are already in that process. What is critically important now is to accumulate successful instances of surveillance. If and when we succeed in proving the usefulness of surveillance, we can create the environment in which it will function with greater authority and effectiveness.

I would like to put forward two practical suggestions to facilitate this process. First, participation of officials in the surveillance exercise should be as senior as possible, so that conclusions reached in the surveillance process can carry greater weight with policymakers and taxpayers in the country concerned. I can endorse fully what Wendy Dobson pointed out in this respect. The second suggestion is to explore ways to make surveillance more an issue of national interest. Unless a critical mass in the country concerned is convinced of the need for, and benefit of, surveillance, we cannot expect a meaningful outcome.

For this purpose, I would argue that some disclosure of the discussions held in the process of surveillance would be useful and necessary. If people at home are informed of what their representatives argued for at the surveillance meetings and how these arguments were received by other participants, they may recognize their country’s position in a global context. The success of surveillance hinges on the crucial condition that voters and taxpayers understand the international implications of domestic policies.

In conclusion, in the increasingly interdependent and ever-evolving world economy, we should not simply dwell on the status quo.

Jacob Frenkel

When the great Princeton University economist Fritz Machlup started addressing a topic, he usually spent the first few minutes analyzing the title and going through a textual analysis of what was on the minds of those who suggested it. The topic of this session is surveillance and the international monetary system. One could speak about surveillance; one could speak about the international monetary system; one could speak about multilateral surveillance; and one could speak about Fund surveillance. I would like to address all of these topics, because each one of them has implications for the subject at hand.

To begin, I think that “surveillance” is a terrible word. Most of us who are used to speaking this language are not aware of all the connotations that this concept and this word carry. Surveillance gives the impression of a policeman chasing criminals. Surveillance means that somebody is looking after somebody, typically in a patronizing way. It means that one needs to coerce somebody who does not implement a commitment of one sort or another. It conveys the notion of dictatorial intervention, intrusion, arbitrariness, and an atmosphere of competition and rivalry between the policeman and those who want to run away. At least, these are the connotations in the English language, and I am told by my linguistic friends that surveillance carries a similar, although perhaps not as strong, connotation in French and other languages.

With this in mind, and if one looks at the Fund’s history, one can detect that in the past, we have seen situations where the Fund allowed itself to be put in the position of an overzealous policeman, a “bad guy” from the perspective of those being treated like criminals, someone who always wanted to impose something. Therefore, criticizing the Fund became the norm and a legitimate style of intellectual exchange. Authorities negotiate with the Fund and then issue extreme criticisms domestically of this immoral, heartless, senseless, technocratic institution. More than one government has chosen to take the easy way out, telling the population that the Fund has forced it to take measures against its will. These are bad mistakes, economically and politically.

The fact is that economic programs, to be successful, require stamina. Therefore, governments that are not strong enough to sell the economic program as their own undercut their authority in the eyes of their own population and cannot be relied upon to implement the details of the program. This creates a syndrome in which the government cannot be relied upon to implement and therefore somebody must coerce it. This is not a constructive atmosphere for an ongoing effort, unless one believes that the job of the Fund is to offer a quick fix: to come in, make its point, and get out. History suggests that that is the wrong perspective.

I would say that surveillance should give way to concepts of cooperation, partnership, and consultation; of bringing on board the rest of the world’s considerations. With this perspective, many questions that the Managing Director raised in his opening remarks and that have come up in the discussions throughout this conference also can be cast more constructively.

Why should the Fund be so deeply involved in surveillance? I think this is so because it is the organization with institutional memory—an asset of considerable value not only to historians but also to member countries’ authorities, who tend to repeat mistakes and tend to forget the lessons of the past. The Fund is there with its objective capacity to bring about institutional memory. The Fund brings to the process experts, an extensive data base, and the like.

This brings me to the subject of Article IV consultations. First, should one publish consultation reports? I do not see why not. In my view, the Fund should publish reports but give member countries the prerogative to ask the Fund not to publish a specific report. I think we would see an interesting competition, in which most countries would have their Article IV consultation reports published, and outliers would feel embarrassed to ask the Fund not to publish. Much less secretive elements would come into play, and there would be much broader knowledge of what the Fund is really doing.

Those of us who spent our professional lives at university before coming to public service really did not know how the Fund operates, because of its low-key, behind-the-curtains approach. I think that we can create a club of those who are proud to have their consultation reports published, and it will be a universal club in the long run.

Second, the impact of Fund policy advice ultimately depends on its quality, and therefore there should be a sharper distinction drawn between the Fund as a financial resource and the Fund as a resource for knowledge and professional expert advice. I think that, in the long run, the financing aspect of the Fund will need to diminish as the other elements will need to come to the fore, because there is no way that, in serving the financial needs of the future and facing the development of private capital markets, the Fund can have a strong enough impact.

If that is so, there may also be a way to bring the Group of Seven back into the Fund in the fundamental sense of using the Fund’s expert advice in a more operational way. For a long time there has been a feeling that somehow, perhaps because of developments in the 1970s and 1980s, the Fund moved away somewhat from being a major partner, a major player with the industrial countries. A country had to be a user of Fund resources in order to be in an ongoing relationship with the Fund. But if one defines resources not in a financial way but in a more conceptual way, as knowledge and expert advice, then the industrial countries will also become users of Fund resources. This will bring back the importance of Article IV consultations and will move the Fund closer to its original purposes.

One way to elevate the role of the Article IV consultations would be to introduce it more explicitly into the discussions within the Group of Seven. As those countries are both major shareholders in the Fund and major participants in the international monetary system, they should be the first to take an interest in having at their disposal the conclusions of the Article IV consultations with member countries during their own deliberations. At the present time, this information is not introduced as explicitly as it should be.

A second way to strengthen the role of consultations is to promote continuity. Even for a country that is a user of Fund resources, although there is an ongoing consultation, there are on-and-off negotiations, ministers come and go, and the staff comes and goes. For a nonuser of Fund resources, the consultation is an annual event (now that, fortunately, the bicyclic procedure has been dropped).5 To promote greater continuity, I would propose that each Article IV consultation start with a more detailed analysis of the previous year’s consultation. The discussion could perhaps be opened with a scorecard of where the country is, compared with the recommendations of the previous round. Is the economy moving forward? The reality is that ministers change, governors change, but the Fund is always there. Institutional memory is required, and this would be one way to improve it.

The Managing Director asked yesterday morning how growth-generating features of Fund-supported programs can be strengthened. I think we should begin by dispelling four interrelated myths. When one talks about the Fund, it is often suggested that there is a distinction between the short and the long run; between crisis management, which is said to be the role of the Fund, and long-term development, the role of the World Bank; between the Fund’s role in macroeconomic matters and the Bank’s involvement with structural policies; and, similarly, between stabilization and growth.

These are fallacious distinctions. One cannot start a stabilization program that will generate growth unless, from the beginning, one keeps in mind that the goal is not just extinguishing a fire but resuming growth later. To do this, some concerns need to be redirected. The Fund has exerted a lot of effort in assessing budget deficits. Clearly, excessive budget deficits are a serious problem. But a budget deficit means very different things depending on the composition of government spending and of taxes. Government outlays on investment in infrastructure or education have a very different impact on the economy and on generating growth from outlays on consumption and wasteful activities. Similar considerations apply to taxes, such as whether they are levied on the business or the personal sector.

When the Fund started focusing more on the design of programs and on the composition of government spending and taxes, the question naturally arose whether it was too intrusive for the Fund to express a view to governments about those intimate domestic matters. I think that the very way of posing this question—of being too intrusive—belongs to the old regime, under which surveillance meant the policeman pursuing the criminal. If, on the other hand, surveillance means partnership, cooperation, advice, and expert know-how, there is nothing too intrusive when the institution draws on the lessons that it has learned from its own experience and that of its member countries.

Should the Fund encourage further moves in capital market liberalization, the Managing Director asked yesterday. The reality is that most links between countries today operate through the capital account: this is where the massive flows occur and where most of the action takes place. Therefore, I would put the question somewhat differently, taking for granted that countries are linked through the capital account.

I remember the days when it was felt that perhaps countries are linked too closely, and that we should “put sand in the wheels.” That was a mistake then, and I believe it is also a mistake now. As every mechanic knows, if you put sand in the wheels, it is very difficult to take it out. The solution, if traffic is moving too fast and accidents may occur, is not to close the roads but to widen and open them. If we are worried about accidents, then safety belts are called for. So, if we are opening up—and we should be—obviously the regulatory system has to be in place, and the supervisory system has to be in place, but we should not put sand in the wheels.

What then should we add to our policy advice as we take it for granted that openness is going to be the basic focus of the advice? That brings us to the question of protection. A lot of attention is devoted to reducing trade protection: through the General Agreement on Tariffs and Trade (GATT), the North American Free Trade Agreement (NAFTA), and similar efforts. In a world in which exchange rates are flexible (if I may take the extreme case to make the point clear), the capital account is just the mirror image of the current account. Therefore, in principle, if a country is worried about sanctions imposed by the GATT or the World Trade Organization if it enacts protectionist trade measures, it can achieve the same results by putting restrictions on the capital account, because the two are just mirror images. I am not necessarily proposing here a new institution, an analog to GATT, but the concept of capital account protection is clearly the counterpart to normal trade protection.

Moving to the next topic, I will make just two brief remarks about the concept of the international monetary system. Incidentally, although some people refer to it as a nonsystem, there is a fundamental contradiction in the concept of a nonsystem. A system merely links parties or components into something that may be more or less synchronized, but it is nonetheless a system.

The international monetary system, or the exchange rate system, has n currencies and, as Robert Mundell told us long ago, n minus 1 exchange rates. There is always one degree of freedom in a system that links n currencies. Only n minus 1 countries can choose what exchange rate they want to pursue, and the nth country is the extra. The Bretton Woods system was very explicit about how the world would use this degree of freedom: all countries would peg to the U.S. dollar, and the system would use its degree of freedom by having the dollar pegged to gold. So the system had consistent features. I would suggest that any system, whether based on target zones or some other key, must be explicit about how to use the degree of freedom that is left. If we are not explicit about it, something is missing in the characteristics of the system.

How do we judge whether a system is good or bad? There are basically two perspectives. Either we could say that a system is good if it normally works acceptably well, or we could presume that any system normally works and conclude that the proper test is whether a system can handle crises and other abnormal situations. This second perspective will give us a very different criterion. The analogy here is the story of the architect who, having noticed that under normal conditions, all buildings stand very solid and rigid, concluded—with disastrous results—that in order to withstand storms he must build very rigid buildings. The criterion for judging the merit of a system is how it operates in a crisis; what is the safety valve that will absorb severe disturbances? I am also reminded in this context of U.S. President Truman’s dictum: “If it ain’t broke, don’t fix it.” There is a real question of whether to fix the present system and of how to know whether the system works or not.

I will conclude with a quote from John Maynard Keynes. Fifty years ago, at the closing ceremony of the Bretton Woods conference, he said, “I am greatly encouraged, I confess, by the critical, skeptical, and even carping spirit in which our proceedings have been watched and welcomed in the outside world. How much better that our projects should begin in disillusion than that they should end in it!” That was 50 years ago, and I think that we are ready for the next round.

General Discussion

Jacques Polak remarked that Frenkel’s reference to Fritz Machlup’s methodology raised an interesting point about the Fund and surveillance. The original Articles of the Fund did not mention surveillance. There was a very good reason for that: the original Articles embodied a system, the par value system, that countries had been expected to obey; thus, the system had included punishments for nonobedience, as France had discovered in 1948. It was only after the par value system had disappeared and the Fund had, to a large extent, lost its surveillance power over its members that it had introduced into Article IV the dictum that “the Fund shall exercise firm surveillance over the exchange rate policies of members ….” Since then, it had been trying to find a way of exercising that firm surveillance.

The bravest attack on Fund surveillance was the period during which it had appeared as if the Group of Seven was exercising surveillance over its own members. He was grateful to Wendy Dobson, who had written a book on that issue, for her recognition of the fact that the Group of Seven countries had not been able to exercise firm surveillance over themselves, and for her conclusion that the surveillance function should be taken back by—not given back to—the Fund.

It was an illusion to believe that the Group of Seven would conduct its meetings by reviewing carefully the policies of the members of its group and, through peer pressure, exercise surveillance. At its most effective, the Group of Seven had succeeded only in an occasional case of arm-twisting, not surveillance as such. The Fund’s acceptance of its role of exercising surveillance over all its members did not mean that there was no role for the Group of Seven; as Dobson had correctly said, there was an important role for the Group in pursuing common purposes with respect to the system, and he was confident that it would continue to do just that.

With that in mind, it was regrettable that Dobson had retreated somewhat from her earlier position, namely, the primacy of Fund surveillance over all its members. In her presentation, she seemed to suggest that surveillance should be decentralized, with surveillance over Europe by one group and surveillance over Japan and the United States by another group. There was a strong case for arguing that if the Fund were to have responsibility for surveillance, it should be conducted by the Fund, not by a subcommittee.

Dobson replied that shortage of time had forced her to be a bit cryptic in her presentation. She had not retreated from her earlier position on surveillance. Rather, she had suggested a two-stage process, the first of which might be based on regional consultation, given that learning and sharing were valued elements of the surveillance process. She did not see surveillance decisions themselves, evaluative or otherwise, being taken in regional groups. Formal surveillance would still be conducted at a senior level in what she would expect to be a revamped Fund, by a body analogous to the current Executive Board.

Peter Kenen observed that the focus of the discussion had been on Article IV consultations, essentially a bilateral process of consultation between the Fund and the individual member and a subsequent discussion of that consultation in the Executive Board. Another issue was involved, however, which Dobson might have had in mind: the exchange rate relationships among the large industrial countries must be discussed with them jointly.

It was one thing to discuss with a small country its exchange rate vis-à-vis its largest trading partner, which was, in a sense, a one-sided story. It was not meaningful, however, to talk about “the” dollar rate: there was a dollar/yen rate, a dollar/deutsche mark rate, and so on. In every case, a collectivity of exchange rates among a group of major countries was involved. The same was true within the EMS. Thus, it was not enough to contemplate an intensification of the bilateral process; it was necessary to consider a strengthening of the process of consultation between the Fund and those countries, whose exchange rate relationships, together, were key to the functioning of the system.

Frenkel considered that the perspective that Kenen had provided was correct. To assess exchange rates among the major industrial countries, it was necessary to include all parties in the discussion. In a sense, that structure was approximated by allowing for the participation of the Managing Director of the Fund as the representative of the rest of the world.

Implicit in the present discussion was an idea that Fred Bergsten had touched on earlier. In the early days, the discussions on exchange rates among major industrial countries had focused on the management of exchange rate intervention—issues largely of a technical nature. As time passed, it came to be recognized that it was not sufficient to focus on exchange rate intervention without discussing the basic economic policies underlying exchange rate behavior. That realization had brought about the discussion of economic policy coordination.

The semantic debate about cooperation versus coordination, illustrated most vividly by contrasting the German and French views, was also very telling. The view of some officials was that they could not deliver on commitments that were made in such a forum, or that they did not believe that such a forum was appropriate for discussing the details of economic policymaking. In their view, the aim was not coordination but rather cooperation—exchanging views, data, assessments, and objectives.

That debate was still open, in the sense that some of the countries participating in those discussions believed that there was more to coordination than just cooperation. Some, however, believed that all that could be hoped for was cooperation. Since the strength of the chain depended on the weakest link, the fact that one of the major countries believed that the coordination mechanism meant cooperation ensured that that view would prevail, which was the situation today.

Toyoo Gyohten commented that Kenen was right in pointing out that there were several aspects of surveillance. It was probably not possible to cover all its aspects in one forum, which was why he had put forward the idea of a three-level surveillance mechanism. In dealing with exchange rates in the context of a financial market crisis, for example, the forum would need to be more compact, efficient, and flexible than other, somewhat larger forums.

Salvatore Zecchini, referring to Frenkel’s comments on the evolution of the Fund toward being a provider of expert advice rather than financing, wondered how effective such advice could be if it were detached from financing.

On the issue of publicity, a policy of openness should be phrased in terms of providing either full and adequate information or no information at all, since partial information might be misleading, particularly to the markets. Markets did not always have full information, of course, which occasionally led to market-induced misalignments. In the case of the exchange rate mechanism of the EMS, for example, the exchange rate relationship between the French franc and the deutsche mark had been put under pressure on occasion, owing perhaps to a lack of adequate information.

With respect to Gyohten’s point about the goal of the international monetary system, namely, to prevent imbalances vis-à-vis other countries, it was useful to bear in mind that imbalances could be good or bad. Gyohten’s point hinged on the concept of the sustainability of imbalances and, in particular, the extent to which the exchange rate mechanism was part of a process aimed at making good imbalances sustainable.

Klaus Engelen recalled that, among the themes to emerge from the May 1994 conference organized by the Institute for International Economics, two specific points stood out in the context of the present discussion. The first, which might come as a surprise to some, was that Germany’s relationship with the Fund in the early days was not without problems. The second point was that some German officials had felt that a careful reading of the original Articles of Agreement of the Fund suggested that the Fund would be in existence for only a limited time. Perhaps that perception changed with the establishment of the General Arrangements to Borrow in 1962.

In looking at the more recent history of the Fund, one was struck by the extent to which the Fund had not been able to put pressure on the major European or Group of Seven countries on such key issues as exchange rate misalignments, whether in the context of the recent EMS crises or the earlier overvaluation of the U.S. dollar. With that experience in mind, he wondered whether the Fund would be able to find a way, as de Larosière had proposed, to convince the major countries to accept an international role for the Fund. In a sense, the Group of Seven was acting in much the same way as the United States had done under the Bretton Woods regime. Although the Group of Seven might well be reduced to a Group of Three, key countries, and not the Fund, would probably continue to make the rules for the next 50 years.

Gyohten, responding to Zecchini’s point about the link between the exchange rate and sustainable internal balance, observed that while exchange rates could produce changes in economic performance, they also reflected economic performance. Thus, there was a two-way relationship. In urging that countries aim to improve the internal balance, he had in mind that each country should try to maintain a reasonable balance between domestic savings and investment at the highest possible level. If each country were successful, the exchange rate would reflect the fundamentals, namely, the inflation rate and the productivity of the respective countries.

Dobson remarked that the persistent differences of view on surveillance among participants in the present conference were informative. In her view, the way forward was very much through what Frenkel had usefully relabeled a cooperative process, facilitated by the Fund. The process would remain informal until the major countries were willing to accept a role for an objective referee.

Frenkel said that he would not be surprised if participants in a conference commemorating the one hundredth anniversary of Bretton Woods were to have a discussion along the lines of the present debate. The difficulty in reconciling the two sides of the debate had less to do with the inability of academics and other observers to design elegant solutions than with the realization that, in the end, international monetary relations involved an element of judgment. The standard debate about fixed versus flexible exchange rates, for example, would probably always be characterized by proponents of one system pointing to the weakness of the other. Thus, among the chief executive officers expressing a preference for “flexible stability” to whom Dobson had referred in her presentation, no doubt there were those who really preferred “stable flexibility.” Therefore, it would be naive to believe that further, even exhaustive, discussion of the issues would point to a single system; the solution to the dilemma had to be earned, so to speak, through good economic policies.


  • Bergsten, C. Fred, John Williamson, “Is the Time Right for Target Zones or the Blueprint?” in Bretton Woods: Looking to the Future, Vol. 1, Commission Report, Staff Review, and Background Papers (Washington: Bretton Woods Committee, July 1994).

    • Search Google Scholar
    • Export Citation
  • Bretton Woods Commission, Bretton Woods: Looking to the Future, 2 vols. (Washington: Bretton Woods Committee, July 1994).

  • Caves, Richard, “Exchange-Rate Movements and Foreign Direct Investment in the United States,” in The Internationalization of U.S. Markets, ed. by David B. Audretsch and Michael P. Claudon (New York: New York University Press, 1989).

    • Search Google Scholar
    • Export Citation
  • Dixit, Avinash, and Robert S. Pindyck, Investment Under Uncertainty (Princeton, New Jersey: Princeton University Press, 1994).

  • Dobson, Wendy, Economic Policy Coordination: Requiem or Prologue? (Washington: Institute for International Economics, 1991).

  • Edison, Hali, Michael Melvin, “The Determinants and Implications of the Choice of an Exchange Rate System,” in Monetary Policy for a Volatile Global Economy, ed. by William S. Haraf and Thomas D. Willett (Washington: AEI Press, 1990).

    • Search Google Scholar
    • Export Citation
  • Graham, Edward M., and Paul Krugman, “The Surge in Foreign Direct Investment in the 1980s,” in Foreign Direct Investment, ed. by Kenneth A. Froot (Chicago: University of Chicago Press, 1993).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, World Economic Outlook (Washington: Internatinoal Monetary Fund, 1993).

  • International Monetary Fund, World Economic Outlook (Washington: International Monetary Fund, 1994).

  • Kaufman, Henry, “Structural Changes in the Financial Markets: Economic and Policy Significance,” Federal Reserve Bank of Kansas City Economic Review, Vol. 79 (No. 2, 1994), pp. 515.

    • Search Google Scholar
    • Export Citation
  • Kenen, Peter, “Ways to Reform Exchange-Rate Arrangements,” in Bretton Woods: Looking to the Future, Vol. 1, Commission Report, Staff Review, and Background Papers (Washington: Bretton Woods Committee, July 1994).

    • Search Google Scholar
    • Export Citation
  • Ohno, Kenichi, “The Case for a New System,” in Bretton Woods: Looking to the Future, Vol. 1, Commission Report, Staff Review, and Background Papers (Washington: Bretton Woods Committee, July 1994).

    • Search Google Scholar
    • Export Citation
  • Tobin, James, “A Tax on International Currency Transactions,” in Human Development Report 1994 (New York: Oxford University Press for the United Nations, 1994).

    • Search Google Scholar
    • Export Citation
  • United Nations Conference and Trade and Development, World Investment Report 1991: The Triad in Foreign Direct Investment (New York: United Nations, 1991).

    • Search Google Scholar
    • Export Citation
  • United Nations Conference on Trade and Development, World Investment Report 1994: Transnational Corporations, Employment, and the Workplace (New York: United Nations, 1994).

    • Search Google Scholar
    • Export Citation

Valued research assistance by Walid Hejazi is gratefully acknowledged.


For more analytical detail on these observations, see Dobson (1991).


Deviation of the nominal yen/dollar exchange rate from its mean in 1987–94 has been between 22 percent and 24 percent; deviation of the deutsche mark/dollar exchange rate has been about 18 percent in the same period (author’s calculations).


The OECD would continue to focus on its review of industrial countries’ overall economic performance, recognizing its heavy reliance on peer pressure to achieve changes in policies.


The bicyclic procedure, which was introduced in 1987 for a limited number of countries, provided for a discussion by the Executive Board only every other year, with a simplified consultation in the intervening year. It was dropped in 1993 in response to concerns that it might have weakened the continuity and evenhandedness of Fund surveillance.

The Future of the IMF and the World Bank: Proceedings of a Conference held in Madrid, Spain, September 29-30, 1994