The Quest for Exchange Rate Stability: Realistic or Quixotic
Speech given at the Senate House
University of London
© Peterson Institute for International Economics
I recall reading some of the early Stamp Lectures when I was a student of economics a good many years ago. The series had quickly made its mark in promoting the application of rigorous economic thinking to the practical problems of the world. That effort was a distinguishing characteristic of Josiah Stamp in all his enormous range of activities. His long service as Governor and Chairman of the London School of Economics reflected that dedication. I feel singularly honored to deliver his 50th anniversary lecture celebrating a truly distinguished Englishman of the 20th century.
Lord Stamp would not have been surprised one bit to find that the world, here at the end of 1995, is still full of practical problems, the solution of which has eluded our grasp. The particular issue I'd like to address today is the functioning of the international monetary system and more specifically the search for greater stability in exchange rates-matters that have been debated through most of this century.
A couple of years ago, as the International Monetary Fund and the World Bank approached their own fiftieth anniversary and a full generation after the so-called Bretton Woods system broke down, I was asked to convene a group of experienced practitioners and scholars from around the world to reexamine that complex of problems. It is fair to say the group, labeled The Bretton Woods Commission, was to a considerable degree self-selected; interest in serving generally reflected a sense of dissatisfaction with what had been happening. A disproportionate number of members were of my generation, men and women who grew up in a more settled financial world, a world characterized by greater stability and by greater confidence in the ability of governments to manage the instruments of economic policy to achieve agreed goals.
We therefore should not have been surprised that the group reached a high degree of unanimity about the need for the major nations to attach higher priority to achieving exchange rate stability. While I was not the author of the Commission Report, I firmly share that conclusion. But as I observed the drafting process, another thing was equally apparent. The longing for a more stable exchange rate system was not matched by much specificity about how to get there.
There was, indeed, a rather vague call for "flexible exchange rate bands". However, that approach was only to be instituted at some unspecified future time after greater stability and greater convergence had been achieved in the performance of major economies. Even that muted call was received by what could only charitably be described as coolness by Government officials and market participants alike.
We have a serious dichotomy. On the one hand, there is strong sense of dissatisfaction with the performance of the international monetary system-or nonsystem, as many disparagingly term it. On the other, there is little disposition to do much about it.
I'd like to convince you this afternoon that change is not only needed but possible-that the present passivity is not justified. There are useful and intellectually sound approaches toward exchange rate reform that can be introduced now, even if, as a practical matter, change will need to be an incremental and piecemeal process. What is in doubt is whether, collectively, we can marshal and maintain the degree of political commitment essential for meaningful progress.
More than 20 years after the breakdown of the par value system, we can conclude unambiguously that hopes of many economists about the performance of floating exchange rates have not been, and will not be, borne out in practice. Instead of gradual exchange rate adjustments in response to differential inflation rates and orderly market responses to other differences in economic policies or circumstances, currency prices have been prone to abrupt and erratic fluctuations. Contrary to earlier theorizing, there has been no discernible evidence of a spontaneous damping of volatility as time has passed. Apparently, neither learning experience nor the proliferation of new techniques and instruments for operating in exchange markets has led to dominance for the kind of stabilizing speculation so prominent in textbook analyses.
No doubt, the volume of exchange market activity has increased exponentially, much more rapidly than international trade and investment. No doubt, too, broad markets and innovative instruments have greatly expanded hedging possibilities for many businesses seeking protection against short- to medium-term currency risks. But hedging entails costs, sometimes substantial, and it really can't deal effectively with the more extreme swings extending over longer periods. Those swings, up and down, as large as 30 or 40 percent over the course of a year or two and occasionally even greater, have not been uncommon in real as well as nominal terms. That has been the case even among the most widely traded currencies-currencies that are looked to as international media of exchange and stores of value.
Quite obviously, international trade and investment have grown even in the face of that volatility. But surely the persistence of such fluctuations in currency values, with the consequence of enormous uncertainty in relative price levels among countries, is inconsistent with the kind of cool appraisals of relative cost and comparative advantage implied by the theoretical underpinning for open markets and free trade.
I frankly do not see much in the record to justify the hopeful view that those extreme fluctuations have been passing aberrations unlikely to recur. To be sure, the successive oil crises of the 1970s, the difficulties and uncertainties associated with bringing down inflation in the United States in the 1980s, and more recently the monetary pressures growing out of the reunification of Germany brought particularly strong pressures on the system. Those pressures would have required strong responses in economic policies to stabilize exchanges rates even within broad bounds. But another thing is painfully obvious and worrisome. We have had very large fluctuations in the dollar/yen and dollar/DM exchange rates in the past year when inflation in all three of those major economic centers has been well controlled, rates of economic growth have not been extremely divergent, and expectations about future trends have not appeared to be particularly volatile. In sum, exchange rates have been prone to move over a very wide range when underlying economic policies were in question and when they were not.
What is striking, in observing exchange market behavior, is how fragile is any sense of market equilibrium. That is true even during periods when economic conditions have been relatively placid. It is symptomatic that many successful traders even question the relevance of concepts of equilibrium so embedded in conventional economic theorizing about the market. One young economist friend of mine has put it to me that further developments in chaos theory must await better understanding of exchange markets. I am not totally reassured by the thought.
In these circumstances, it seems to me wrong to continue to argue, as typically has been done ever since floating began, that international currency reform and greater stability in exchange rates must await greater domestic stability and convergence among domestic economies. We need to examine at this point, it seems to me, the mirror image of that proposition: specifically, whether attaching greater priority to currency stability might also contribute over time to more consistent economic performance, growth and productivity.
For good reason, small countries heavily dependent on international trade and investment have typically attached high priority to currency stability. Large exchange rate fluctuations are particularly disturbing when both internal price levels and much of the economy are strongly exposed to external influence. Unfortunately, we have had plenty of experience with more extreme circumstances, when individual countries fall into the kind of exchange market crisis that threatens their economic and even political continuity and cohesion. Sooner or later they were forced to respond with painful measures of domestic adjustment. In some cases, truly extraordinary actions have been required to restore a sense of confidence, including strong commitments to defend a fixed value for the currency against a convenient anchor. The recent revival of the concept of currency boards-in effect, providing a stronger institutional framework for restoring a sense of stability and confidence in the exchange rate-is one response to that felt need.
It is, of course, also true that smaller countries, through their own actions, can't deal with the instabilities among more important currencies. And the fact remains that larger countries, with their domestic production and prices less sensitive to exchange market influences, have not felt the same sense of commitment to currency stability. That point is often made with particular force with respect to the United States. But it is not only the United States that perceives that domestic and international objectives may conflict, and is therefore reluctant to commit prestige and policy to exchange rate stability. I need not belabor the point to a British audience.
In defense of that view, for all the turmoil in exchange markets during the past 20 years and more, the world economy has continued to grow. International trade for the most part has continued to expand more rapidly than domestic production. Despite quite a lot of backsliding into nontariff barriers from time to time, there has been on balance a clear reduction in tariffs and other restrictions over the decades. In economically developed countries-and in many emerging markets as well-capital controls are a thing of the past, providing support for international investment and growth. In the developed world inflationary forces have been reined in, if not totally conquered; we are almost back, in that respect, to the relative stability of the 1950s and 1960s. And looking at the broader canvas, we have seen almost everywhere the triumph of the ideas of economic liberalization.
Even a skeptical central banker professionally addicted to worrying might want to render a judgment of "not so bad" about what's been accomplished over the past decade, exchange rate instability notwithstanding. And it is that sense of relative complacency-the sense that, all things considered, the global economy is functioning reasonably well-that dulls any urge for reform.
Nonetheless, a modern Rip Van Winkle, awakening after 20 years, might find that complacency surprising. After all, despite the enormous vitality of emerging Asia, growth and productivity has been much slower than in the first post-World War II decades for most of the developed and developing world. Whatever the uncertainties about some of the data, the liberalization of trade and payment has not yet produced a sense of consistent improvement in living standards. The presumed flexibility for domestic monetary and fiscal policies has not produced lower levels of unemployment or clearly improved cyclical performance.
I am not about to join the small school that associates the shortfall from earlier standards of performance mainly to the breakdown of fixed exchange rates. Certainly, econometricians haven't yet established a clear relationship. There are other plausible culprits-low savings, high deficits, some would say the rising burdens of social welfare costs. And, the immediate post-World War II decades were, after all, exceptional in the long sweep of economic history.
At the same time, there have been offsetting factors. The last part of the 20th century has been a period of remarkable technological advance. Calculating, communication, and information processing are faster and cheaper by orders of magnitude. Markets, not only in goods and capital but also in services and ideas, are more open. Those should be forces powerfully improving productivity. Yet, growth has slowed.
Classic economic analysis surely suggests that swings in exchange rates that depart radically from equilibrium values-values associated with sustainable trading and investment patterns-must give misleading price signals and add to uncertainty. Those factors must in turn inhibit investment and reduce efficiency. We are told that our large multinational businesses have learned to live with these currency fluctuations by dispersing investment, producing locally for local markets. But measured against the concepts of comparative advantage, that obviously carries real costs.
Whatever the force of those concerns, they are not the sort of thing that move the political process. The damage is insidious. But longer term and subtle influences of that sort do not ring alarm bells and overcome passivity in the face of exchange rate volatility among the key currency countries.
Three factors in my view contribute importantly to that passivity. First, as a matter of economic analysis, it is well known that, in a world of capital mobility, fixing exchange rates implies lack of autonomy in monetary policy. That stricture is moderated as degrees of flexibility are introduced into the exchange rate system. However, at some point a commitment to stabilizing exchange rates will indeed require the active use of monetary policy to that end.
That requirement need not-and I believe typically will not-be at odds with the appropriate posture of monetary policy on more purely domestic grounds, given a reasonable degree of exchange rate flexibility. Moreover, with full use of other instruments of government policy, perceived conflicts with other objectives conceptually could be moderated or even overcome. But therein lies the rub. Few governments have the kind of effective control over fiscal and other policies that flexible manipulation of the economic policy mix would require. To make matters worse, we have learned that Keynesian optimism about the power and predictability of fiscal changes-even if well timed changes were politically to be feasible-are easily exaggerated.
Those difficulties are compounded by a second problem inherent in the nature of exchange rates: they involve more than one nation. In the face of unwanted changes among major currencies, the contentious question inevitably arises as to which country assumes the "burden" of initiating appropriate policy changes. The answer is seldom clear cut, and its resolution entails sensitive political as well as difficult economic judgments. Whatever so-called objective indicators are brought into play, the differing preferences of different countries will lead to delay and weak responses.
The third difficulty is a narrower political question. The traditional strong constituency for stability-the financial community-has effectively reversed its position. That shift no doubt is related to the fact that financial institutions operating internationally have today developed a substantial vested interest in market instability. Their traders are in a strong tactical position to sense market pressures and psychology and to act immediately to take advantage of emerging movements and even to amplify them. To be sure the risks of market makers are substantial and an occasional rogue trader can gravely embarrass-even destroy-a prestigious institution. But those traders have been provided so much freedom precisely because, year in and year out, exchange market and related trading has become an important profit center. The irony is that the same market volatility that provides trading advantages offers further profit opportunities to institutions alert enough and intellectually equipped to design and sell sophisticated hedge products to their commercial customers.
Those that bear the cost in the first instance are the importers and exporters and the international investor. But for them, the currency and hedging costs are part of much larger and presumably profitable transactions. And the ultimate costs of uncertainty and lower productivity are buried among the population at large. Analogous to the politics of protectionism, the gains from instability are concentrated, the losses diffuse.
Discouraging as all that may be to would-be reformers, there are solid grounds for thinking the quest for greater stability is not quixotic. We have before us the evidence that few if any countries are willing to leave their exchange rate entirely to undirected market forces, given the volatility of the response to those forces. As I emphasized earlier, the pattern of smaller countries stabilizing exchange rates with major trading partners is clear. There is also a natural tendency for regional groupings-preeminently the European Union-to seek stability within the group. When confidence in currency values in particular countries is shattered, there is often a strong impulse to seek an exchange rate anchor, even for relative large countries. In all those instances, difficult political and economic choices have been and are being made to preserve and enhance exchange rate stability as a matter of national priority.
The key question is whether among the key countries with globally important currencies that same urge for stability exists to the degree necessary to support a broad reform effort. The answer is clearly "no" if reform is posed in terms of a system of fixed parities à la Bretton Woods. Nor are there practical possibilities of moving to a full-blooded gold standard or a common and stable world currency under the aegis of a world central bank as still more radical reformers suggest. The theoretical and ideological attractions simply fade before the practical and political obstacles to the degree of economic policy coordination that would be involved.
But we must recognize, too, that the other extreme-freely floating rates-has also been found wanting, and in practice abandoned. To be sure, internationally coordinated approaches among the major powers toward exchange market management have been episodic, provoked only by extreme fluctuations that seemed to threaten growth, market access, and the very fabric of international cooperation. Those efforts have not always been fully successful-far from it. But out of all this experience, I think we can draw some conclusions about what is promising and what is not-whether, at the end of the day, some more systematic approach might be found to reconcile needed flexibility with a greater sense of stability.
There are also useful lessons for broader application in what has been accomplished-or not accomplished-so far within Europe. I do not want to enter the debate about the ultimate merits of the European Monetary Union or of the various paths toward that end.
Whatever those judgments, it is plain we do have here in Europe a group of countries dedicated to creating a true single market and fearful that substantial volatility and pronounced misalignments of exchange rates among member countries could be destructive of the Union. The current tensions with respect to the depreciation of the lira add point to those concerns.
At the same time, the member countries obviously differ in economic structure and circumstance, and in the degree to which they are willing to surrender perceived autonomy in economic policy. Despite those differences, a remarkable degree of stability (some would say rigidity) was achieved in most European exchange rates during the 1980s and the first years of the 1990s. That record is all the more remarkable when one considers that during that same period, the remnants of capital controls were removed and European markets, like others, have been exposed to unprecedented volumes of international short-term capital movements.
By the rules of the European Monetary System, short term credit lines have been freely available to support intervention to defend exchange rates. But maintenance of the designated central rates at times of persistent pressure has required more than intervention; monetary policies have been tightened, tightened more than might otherwise have been the case in the face of domestic political and economic circumstances. No doubt, fiscal policies in some member countries have been more disciplined as well. What is not in doubt is that inflation rates have converged dramatically around the DM, the operational anchor of the system, even though significant differentials have remained for Italy, Spain, and the UK as well.
Equally plainly, the record of stability broke down in the face of pressures against sterling and the Italian lira, and the narrow margins of the European Monetary System needed to be substantially relaxed. Nonetheless, it remains true that for the core countries-the countries more committed to the concept of a single currency and with reasonably close convergence in economic performance-intra-EMS exchange rates have remained remarkably steady even in the face of substantial and unsettling fluctuations against the dollar and the yen.
This experience suggests several points of wider application. Intervention alone, however large the supporting lines of credit, is not sufficient in itself to maintain exchange rate relationships. Sovereign nations are simply not prepared to lend or borrow indefinitely, and market speculation will persist in the face of perceived misalignments. It follows that a commitment to stabilize exchange rates requires sustained evidence of political will-the will to give strong emphasis to that objective in the conduct of economic policy generally and of monetary policy in particular. One prerequisite to that commitment is a shared view among participants about the importance of price stability, the sine qua non of exchange rate stability.
The three polar financial centers of the United States, Europe, and Japan have substantially converged toward price stability but the sense of commitment with respect to exchange rate stability has obviously not been present to anything like the same degree. It is also true that even the most reluctant and passive of governments with respect to exchange rates, typically the United States in recent years, have recurrently found it necessary to take a stand, to indicate however vaguely which exchange rate levels are reasonable and appropriate and which are not, and when the market has moved too far to act. Typically, that action has taken the form of intervention. But it is also clear that the response has been most effective when accompanied by signals that the major countries were in agreement, by evidence they were willing to act in unison, and by signs of coordination with respect to monetary policy as well.
I know there have been efforts to replay the past with complicated econometric models about how the world would have differed had economic policies been more strongly directed to exchange rate stability. I am told the evidence is inconclusive at best, and that according to the model criteria, an exchange rate signal for monetary policy would often be perverse. However, in looking back over the last quarter century or so, what strikes me in terms of United States experience is that at really critical junctures the exchange rate was given too little emphasis, not too much. Sure, with hindsight, the United States would have been better off with more restrictive monetary policies in the early 1970's as the fixed exchange system broke up and the dollar began its long depreciation. That was when inflation first reached levels well above earlier US experience. Surely, when the United States in 1978 did mount a well coordinated intervention effort in response to another sinking spell for the dollar-an effort clearly supported by monetary tightening-it was late in the day. Both exchange rate and inflationary expectations were already deeply embedded. Those expectations could later be rooted out only with large transitional cost.
Conversely, the Super Dollar of the mid-1980's-partly a by-product of the subsequent war on inflation-conveyed a clear message about the excessive budget deficit-a message largely ignored. Later in 1987, the ambitious effort at the Louvre to set out guidelines for an acceptable range of exchange rate fluctuation was, to my regret, not adequately supported by policy action either in Japan or in the United States. More recently, the disconcerting fluctuations in the value of the yen might well have been moderated by more flexibility in Japanese fiscal and monetary policies.
What is at issue now is whether we are ready to build constructively upon all this experience. I do not envisage a leap into a new Bretton Woods-a highly structured system involving scores of countries. Something much more modest is both more realistic and more desirable. What we need is an approach that will moderate and reverse exchange rate fluctuations among the key currencies before they become extreme, rather than being forced to respond defensively, after substantial risk to the world economy is already evident. At least initially, the main steps should be taken by the G-7 (or even the G-3), working closely with the IMF and respecting both its experienced advice and its perspective as the voice of the larger world community.
This approach would require agreement in several broad and interrelated areas.
- First, the major countries, in consultation with the IMF, would need to reach a consensus on broadly appropriate equilibrium values for their currencies. The choices would necessarily imply a fairly wide range of values. They would have to be expressed as ranges, say of plus or minus 10 percent around a central point. (Starting with even broader ranges, à la the present European Exchange Rate Arrangements, could be a transitional device.)
- Second, the participating countries would need to be prepared to jointly defend the indicated range with intervention, on a substantial scale if necessary. Such intervention would be required at or near the margins, but could be tactically desirable well within those limits.
- Third, that intervention will be reliably effective only to the extent that, in the event of persistent pressure, the participating countries are prepared to modify their monetary policies in support of the exchange rate objective. Reasonably prompt adjustments could well forestall the need for stronger action in the face of strong speculative pressures.
- To clarify choices, promote consensus, and help influence expectations, the IMF management should be involved in the discussions when the need for policy action arises. The IMF management should be prepared at its own initiative and confidentially to propose an appropriate course of action.
The extent to which countries are prepared to announce publicly the "equilibrium ranges", and the frequency with which they might be modified are sensitive points. What seems clear is that the range should be relatively wide-a pragmatic compromise between the desire, on the one side, for stability and on the other, to avoid unrealistically rigid commitments. What seems critical to success is that defense of the outer edges of the range be taken seriously by governments and markets alike, something that can only be demonstrated by experience.
Governments will not easily-not now anyway-give up their ability to change the kind of exchange rate ranges proposed. The width of the range does, however, suggest that such changes could be made in a way that minimizes the possibility of easy "one way bets" by market operators. Specifically those changes should be made by amounts substantially less than the width of the range (and not necessarily only when the market rate is already at the margin). In that way, the market exchange rate need not move much, or necessarily at all, when such adjustments are made.
One source of hesitancy about defending exchange rates, even with wide margins, is the common talk these days of the enormous resources available to market operators, and how those resources can overwhelm Treasuries and Central Banks. That is, of course, true if exchange rates are defended without conviction-without a willingness to modify, if necessary, the more basic instruments of economic policy. But I see no evidence that markets are not responsive to monetary policy; quite the contrary, both domestic and international markets are constantly preoccupied in their trading with anticipating or adjusting to moves by the Central Bank. A world in which strong and persistent inflationary forces have been turned back with little support from fiscal policy is hardly a world of central bank impotence.
That illustration does point up a real difficulty greatly complicating the objective of stabilizing exchange rates. Much too commonly, large and persistent government deficits in the main countries have had the effect of limiting the flexibility of the monetary authorities that must be sensitive to their overriding responsibility as guardians of price stability. The Keynesian vision of prompt and effective adjustments of fiscal policy may be gone and unlamented. But it would be of decided assistance if structural budget positions could be brought into line with the needs of better balanced economic policy-for instance, the United States, as a chronically low savings country, should aim not just for structural budgetary balance but for surplus. Moreover, without a sound structural budget position, the possibilities of fiscal stimulus to help deal with cyclical weakness are limited.
There is another point relevant to the politics and psychology of governments. Relatively wide and potentially movable exchange rate ranges are in a sense a compromise between the logical extremes of fixed and floating rates. The idea, for all its analytic appeal, does not lend itself to slogans or sound bites, nor to instinctive political or public support. The question will be asked, when defense of the range is required, if 10 percent is all right, what about 11 or 12 or more? Is it really worth spending money in the exchange markets, modifying monetary policy, and taking care to balance the budget just to save another percentage point or two?
The answer must be yes. What is at issue is not that last percent but whether governments will succeed in inducing the market itself to stabilize exchange rates. The success or failure in that effort is plainly dependent on the credibility of official intentions. But when that credibility is established, markets will work with governments, not against them, to maintain a sense of equilibrium.
What is lacking today is a sense of motivation and urgency. There is a reluctance to make a sufficiently strong commitment to stability for fear the effort could fail, at political and economic cost. What is not adequately weighed in the balance, is the disintegrating force of present exchange rate arrangements, with its inherent uncertainties and false pricing signals.
The irony for me is to observe the enormous energy and political capital dedicated in recent years to reducing already low tariffs to minimal levels, only to see the potential gains in efficiency and trade overwhelmed by the volatility of exchange markets. In the same vein, in all our discussions of the problems of development and our praise for the liberalization of emerging economies, we don't give much weight to their stake in more stable exchange markets.
The most important prerequisite for exchange rate stability-the rock upon which all else must rest-is a strong sense of commitment of the major powers to reasonable price stability. That is most of all true for the United States itself, as the issuer of the key reserve currency.
Happily, enormous effort has been made by the leading powers generally to restore price stability, and as part of that process to bring government deficits under control. That effort, in terms of political and economic effort, far exceeds anything I am talking about with respect to exchange rate stability. It also seems clear the goals are basically mutually supportive. And today, precisely because so much has been achieved in that respect, the time to deal with the exchange rate question is now.
Let me conclude by suggesting a practical if limited test of these ideas. Japan and the United States, presumably with support in Europe, seem sensibly to have concluded a dollar/yen rate of about 100 is appropriate and broadly consistent with a sustainable equilibrium. They have together acted constructively in the markets in support of that view, and to some degree the objective of stabilizing the market has influenced Japanese (perhaps even American) monetary policy.
Why not acknowledge that state of affairs more explicitly, suggesting substantial deviations in either direction would be resisted by both countries? A change in the desired range would be made, if at all, only in small steps over time, congruent with changes in underlying circumstances.
Then let's see whether those ranges could be maintained. That will require keeping inflation under control, a desirable and attainable objective in both countries. It will also require that Japanese growth be stimulated by an appropriate mix of monetary and fiscal policies. The slow process of market opening in Japan should be reinforced. The United States will need to implement those good intentions with respect to ending its budget deficits. Those are all desirable goals in their own right. If further encouraged by a commitment to exchange rate stability, all the better.
I have no way to guarantee success. What we do know, on the past record, is that exchange markets won't settle down by themselves-there is no real hope of redemption by faith alone.
My sense is that, at the end of the day, we will find success easier than feared by so many-that the market will more often than not respond constructively to a firm and intelligent lead by governments and exchange rate stability will reinforce prospects for growth. One thing is for sure: without trying, we will never know.