III Evaluation of the Present Exchange Rate System

Thus far, the broad characteristics of the present Internal Balance system have been described and the principal changes in the global economic environment have been iden- The first criterion reflects the view that the exchange tified. It is now time to move to an evaluation of the rate system is basically a facilitating mechanism for present exchange rate system. That evaluation is more fundamental domestic economic objectives, such conducted in two distinct steps. The first step is to as price stability, high employment, and sustainable introduce a set of criteria that can be used to evaluate economic growth. That is why, in sharp contrast to not only the present exchange rate system but other some earlier analyses of exchange rate systems (e.g., exchange rate systems as well. The second step is to Williamson (1980)), the degree of exchange rate vari- apply those criteria to the operation of the present ability, for example, is not put forward here as a system over the past ten years. normative criterion. In other words, the assumption The basic reason for separating the discussion of is that exchange rate variability is important only to the criteria themselves from their application to the the extent that it impinges upon (or facilitates) the experience of managed floating is that, even if readers achievement of more ultimate targets of economic disagree about the relative strengths and weaknesses policy. of the present system, the acceptance of a common The channels by which the exchange rate system framework for evaluation at least ensures that disa- might help or hinder macroeconomic policy are many. greements are based on different readings of the In this study, the focus is on the following issues— evidence rather than on different yardsticks. each of which has figured prominently in the ongoing debate on the merits of the present system: (1) Does the system provide the "discipline" necessary for the Criteria for Evaluation imposition of responsible macroeconomic policies, The following four criteria serve in this paper as a particularly for inflation-prone governments? (2) Do exchange rate fluctuations and downward price inflex- basis for evaluating the present exchange rate system: ibility combine to produce an upward ratchet effect Criterion #1—Does the system help or hinder mac- on national and global inflation rates? (3) Does the roeconomic policy in pursuit of fundamental domestic system exacerbate intercountry inflation differentials economic objectives (price stability, sustainable growth, by drawing weaker countries into a "vicious circle" high employment)? of inflation and currency depreciation and stronger Criterion #2—How effective is the system in pro- ones into a "virtuous circle" of price stability and moting external payments adjustment? currency appreciation? (4) Does the system affect Criterion #3—How does the system affect the vol- unemployment rates either by adverse effects on the ume and efficiency of world trade and capital flows efficiency of the price mechanism that increase fric- (and thereby resource allocation in the international tional unemployment, or by generating longer-term economy at large)? exchange rate disequilibria that foster structural un- employment in traded goods industries? (5) Does the Criterion #4—How robust or adaptable is the system nature of the system influence the effectiveness of to significant changes in the global economic environ- monetary policy under conditions of high capital mo- ment? bility? and (6) How well does the system function as Because these criteria figure so prominently in what a shock absorber against different types of disturb- follows, it is useful to discuss first not only their ances? In seeking to answer these questions, both rationale but also what kinds of issues each encom- short-term volatility and longer-term fluctuations in passes. exchange rates will be considered.

13

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

External Balance It should be recognized that this second criterion is meant to encompass not only the question of whether Moving to the second criterion, considerations of the exchange rate system promotes external payments external balance are introduced to supplement the adjustment but also the question of how it does this. internal balance objectives subsumed under the first Specifically, the second criterion leads to consideration criterion. By asserting that a desirable exchange rate of the following external adjustment issues: (1) What system is one that promotes external payments ad- are the respective weights of relative price changes justment, it is meant that the system should set in and income movements as adjustment mechanisms in train an internationally acceptable adjustment mech- that system—and how do these two mechanisms differ anism, either automatic or discretionary, that elimi- in promoting adjustment? (2) What are the implications nates balance of payments disequilibria over a reason- in that system of different speeds of adjustment in able time period. To make such a criterion operational, goods versus asset markets (or in the current account it is necessary to have some definition or concept of versus the capital account)? (3) Does the system balance of payments equilibrium. This continues to be promote symmetry of adjustment between deficit and a thorny problem. For the purposes of this paper, it surplus countries and between reserve centers and is sufficient to think of balance of payments equilibrium nonreserve centers? and (4) Is external adjustment in as a condition under which the current account position that system automatically induced or is it discretion- can be financed by normal capital flows without re- ary?Again, it is sufficient to note that each of these course to undue restrictions on trade, special incentives adjustment issues has been part of the debate on the to inflows or outflows of capital, or wholesale unem- functioning of the present system. ployment.30 This is essentially the definition suggested by Nurkse (1945) almost 40 years ago. It is closely related to the concepts of fundamental disequilibrium Volume and Efficiency of World Trade and and of underlying payments equilibrium employed by Investment 31 Fund staff, and it forms the basis for most definitions The third criterion derives from the proposition, of the equilibrium exchange rate (which is usually given explicit endorsement in the purposes of the defined as the exchange rate that produces this type Fund,33 that global welfare is generally increased by 32 of payments outcome). A finding that a given ex- an expansion of world trade and investment. This is change rate system has produced effective external another area where it is desirable for the exchange adjustment would imply that observed exchange rates rate system to act as a facilitating mechanism for some were, over the medium term, close to equilibrium real more basic economic objective. This criterion deals exchange rates. with the efficiency of trade and investment because, in the real world where international traders sometimes 30 Special incentives to capital flows need to be interpreted broadly. are reacting to temporary relative price signals that They would include not only, say, tax advantages or disadvantages but also an interest rate that was reflecting an unsustainable mix of bear little relation to longer-term changes in compar- monetary and fiscal policies. Also, unemployment should be viewed ative advantage, not all increases in the volume of as a proxy for a whole range of cyclical or temporary factors. trade will be beneficial—that is, it is possible to have Finally, where the country's level of international reserves is unduly low or high, it might be desirable to allow for some change in "false trading," to borrow a phrase from McKinnon reserves. (1976). 31 For example, the 1970 report by Executive Directors of the In attempting to appraise the effects of the present Fund on the role of exchange rates in the adjustment of international payments (IMF (1970, p. 48)) noted: "Thus, the criterion of fun- exchange rate system on the volume and efficiency of damental disequilibrium is wider than the occurrence of a disequi- world trade and investment, it is necessary to examine librium in the actual balance of payments. . . . For example, the two additional questions: namely, (1) How does ex- concept of fundamental disequilibrium could include a balance of payments position that would have shown a deficit but for restrictions change rate uncertainty affect international trade and on trade and payments; or a situation of equilibrium (or surplus) in investment? and (2) Do sizable exchange rate dise- the balance of payments that would turn into a deficit but for an quilibria generate strong and effective pressures for unacceptably low rate of economic activity in the country; or a situation of equilibrium (or deficit) in the balance of payments that protectionism? would turn into a surplus but for exports of capital at a rate that the country concerned did not wish to continue, or but for the Adaptability to Change country's acquiescence in an unacceptably high rate of inflation." A recent Fund study (IMF (1984 b)) implies a very similar The final criterion is different from the others be- definition of equilibrium payments balance. 32 See, for example, Artus (1978) and IMF (1984 b). Similarly, cause it is not associated with any of the familiar Williamson (1983, p. 14) defines the fundamental equilibrium ex- economic objectives—whether foreign or domestic. change rate as that "...which is expected to generate a current account surplus or deficit equal to the underlying capital flow over 33 Article I of the Fund's Articles of Agreement cites as one of the cycle, given that the country is pursuing 'internal balance' as the purposes of the Fund "to facilitate the expansion and balanced best it can and not restricting trade for balance of payments reasons." growth of international trade" (IMF (1978)).

14

©International Monetary Fund. Not for Redistribution Floating Rates and Macroeconomic Policy

The rationale for including it is that, as with political Floating Rates and Inflation constitutions, there are nontrivial costs associated with changing international monetary constitutions, espe- Critics of floating rates have long contended that cially under crisis conditions. Other things being equal, floating rates are inflationary on three principal counts: it is therefore better to have an exchange rate system (1) because they weaken the resolve or discipline to that is relatively robust or adaptable to changes in the fight inflation; (2) because they interact with downward global economic environment. For example, an ex- price inflexibility to ratchet-up both country and global change rate system that will work well only under price levels; and (3) because they trap weaker countries conditions of low international mobility of capital is in a vicious circle of inflation and currency deprecia- undesirable unless it is certain that capital will have tion, thereby exacerbating intercountry inflation dif- low international mobility in the future. Similarly, an ferentials. exchange rate system that relies on all changes in comparative advantage being slow and smooth will Discipline Hypothesis not be as desirable, ceteris paribus, as one that can also accommodate more abrupt changes. The same The discipline hypothesis is based on the assumption logic applies to other environmental factors, ranging that the balance of payments constraint under fixed from the degree of real wage flexibility to the preference exchange rates acts as a check on the pursuit of for a particular reserve currency and even to the inflationary policies. Because a devaluation would be assumed behavior of one particular type of economic regarded by the public as an admission of the failure agent (whether the reserve center country or market of government policies, a high-inflation country will speculators). This is not to say that the durability of sooner or later be obliged to alter its policies in such a system is as important as how well it functions while a way as to bring the inflation rate into line with that it lasts but rather to suggest that it surely counts for of its neighbors. The perception by the public that a something. given parity must be defended prompts the adoption of otherwise unpopular policies of demand restraint. With these four evaluative criteria in mind, the next Surplus countries under fixed rates are said to be step is to use them in a systematic assessment of the 34 subject to a weaker discipline, either because there is past decade's experience with managed floating. In no similar constraint on the accumulation of reserves order to place that experience in perspective, com- or because the revaluation that is necessary to avoid parisons will frequently be made with experience under such accumulation carries no political liability. Under the system of adjustable par values during its last floating rates, this anti-inflationary discipline is absent decade. Also, references will occasionally be made to because (so it is argued) the only consequence of a experience under even earlier exchange rate systems relatively high inflation rate is a depreciating currency. (e.g., the gold standard). The purpose of such com- parisons and references is not to draw conclusions The discipline hypothesis prompts the following about whether managed floating is the best (or worst) observations. of all past exchange rate systems, but rather to guard (1) Whatever the differences in anti-inflationary against holding managed floating to an unduly high or discipline between truly fixed rates and purely floating low absolute standard of performance. rates, these distinctions become blurred in a compar- ison of adjustable par values and managed floating.36 Yet the latter two regimes are the relevant ones for Floating Rates and Macroeconomic Policy35 the observable macroeconomic policy behavior of the postwar period. If floatingrate s do affect inflation, or unemployment, (2) The political cost of devaluation under fixed rates should not be exaggerated. From 1955 to 1971 or the efficiency of domestic monetary policy, or the the longest period without an exchange rate change insulation of the domestic economy from external by any of the 16 OECD countries was five years (1962- shocks, how do they do it? The arguments of both 66). Likewise, the number of exchange rate adjust- the advocates and critics of floatingrate s are examined ments by high-inflation participants since the inception here. of the European Monetary System in 1979, as well as the number of departures by high-inflation countries

34 Other appraisals of experience with managed floating can be 36 Managed floating can still, however, be differentiated from the found in Willett (1977), Black (1977), Artus and Young (1979), adjustable peg system by the greater frequency of exchange rate Goldstein (1980), Dornbusch (1980), Williamson (1983), Solomon changes, by the larger share of the external adjustment burden that (1983), and Shafer and Loopesko (1983). is assigned to the exchange rate, and by the absence of a publicly 35 Many of the arguments in this section draw heavily on Goldstein declared target exchange rate that must be defended except in a (1980). situation of fundamental disequilbrium.

15

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM from the "snake" in an earlier period, suggest that To summarize, the inflation performance of indus- devaluation is not viewed as a political catastrophe.37 trial countries over the past decade has been so poor All of this is also consistent with the emerging literature that it is difficult to reject a call for more discipline. on public choice that relates voting behavior and The near quadrupling (from 1.2 to 4.2 percent) of the government popularity to economic variables. It is ratio of industrial countries' government fiscal deficits found in the literature that only the traditional domestic to gross national product (GNP) between 1963-72 and macroeconomic variables (real income growth, infla- 1973-82 is just one indicator of how times have tion, and unemployment) count and that only recent changed. But it does not follow from this that fixed performance is important (i.e., voters have short mem- exchange rates are either a necessary or sufficient ories) (Fair (1978); Frey and Scheneider (1978)). mechanism for establishing such discipline. The res- (3) In those cases where there have been conflicts toration of anti-inflationary discipline in 1979-82 in the between internal and external balance under floating face of both high unemployment and strong pressure rates, these conflicts have by no means always been for monetary expansion demonstrates that fixed rates resolved in favor of the internal target. Black (1978), are not always "necessary." Likewise, during the past for example, after studying such conflicts over the two decades there are too many examples of exchange 1973-76 period, reports (p. 626): rate targets giving way to employment targets when push came to shove to believe that fixed rates are In most cases, some influence of the external target on monetary or fiscal policy is evident, except for Germany "sufficient." In fact, if it is necessary to generalize in 1973, the United Kingdom in 1974, and Canada in from the experience after World War II, it would be 1974. Furthermore, the influence of external targets ap- more accurate to say that both the degree and inter- pears to have been rising, as the 1976 conflict cases country dispersion of macroeconomic discipline de- (France, Italy, Canada, the United Kingdom, and Sweden) termine the exchange rate regime rather than the have all been resolved in favor of the external target over reverse. the internal target. Such revealed preference for the external target Ratchet Hypothesis suggests that the balance of perceived political costs is more complex than is sometimes alleged. Turning to the ratchet hypothesis, the basic argument (4) The 1979-83 policy experience in industrial coun- is that floating rates have an inflationary bias because, tries is evidence that anti-inflationary discipline can in a world of downward price inflexibility, devaluations be restored without fixed exchange rates. Indeed, the lead to price increases in the devaluing country but to deceleration in growth rates of narrow and broad no (or smaller) offsetting price decreases in the reva- money that took place in most of the major industrial luing country. For example, if Country A devalued its countries in 1979-82 coincided with relatively high exchange rate by 10 percent vis-a-vis Country B and variability of both nominal and real exchange rates.38 then a year later reversed the process by revaluing by This episode of monetary restraint reinforces the point 10 percent—so as to cause no net change in the that, while fixed exchange rates are one way of making exchange rate over the whole period—the ratchet a nonaccommodation strategy more credible, they are hypothesis indicates that domestic prices would be not the only way—and perhaps not even the most higher in both countries and so, too, would the world effective way. Alternative policy instruments that can price level. be used for anti-inflationary discipline include tax- The key to the inflation predictions of the ratchet based incomes policies, preannounced money supply hypothesis is the proposition that prices do not fall in targets, constitutional limits on budget imbalances, the revaluing (appreciating) country (or at least do not stabilization programs of the International Monetary fall by as much as they rise in the depreciating country). Fund, and, more generally, the aversion of public Two explanations have been put forward for this opinion to the continuation of price inflation and its proposition. One is that export prices (expressed in attendant distortions. domestic currency) in the devaluing country will rise by the full extent of the devaluation, so that import prices (again, in domestic currency terms) in the 37 See Ungerer, Evans, and Nyberg (1983) for a listing of these revaluing country will not fall after an exchange rate exchange rate realignments. The "snake" refers to the adjustable change. The second is that import prices will fall in peg system operated by some European countries in 1972-78, under which there were small and frequent adjustments of exchange rates. the revaluing country but, as such a decline will be 38 Whereas the average annual increase in broad money for viewed as temporary and as producers only respond the seven major industrial countries was about 12 percent in both to those cost or demand changes they deem to be 1963-72 and 1973-82, the corresponding figure for 1979-82 was less than 10 percent. If real monetary aggregates were used, the contrast permanent, these declines in import prices will not with the 1979-82 period would be more marked. provoke any decline in domestic prices.

16

©International Monetary Fund. Not for Redistribution Floating Rates and Macroeconomic Policy

While the ratchet hypothesis has a certain intuitive change rates or otherwise) will leave prices unchanged, appeal that derives from the apparent stickiness of while all cost and demand increases are quickly passed money wages and prices in industrial countries, em- forward into price increases. pirical tests of this hypothesis have almost unanimously In sum, while few would disagree that the downward been unkind to it. To begin with, most estimates of inflexibility of finished goods prices and money wages the export price response to exchange rate changes during recessions has been a formidable obstacle to find that export prices do not rise by the full extent of more successful macroeconomic policy in industrial a devaluation except in the smallest, most open, countries over the past two decades, there is little industrial economies (Robinson, Webb, and Townsend evidence that exchange rate fluctuations have com- (1979); Goldstein and Khan (1982)). The same conclu- pounded that problem to any important degree. Fur- sion also applies to export price behavior following ther, the roots of the downward price inflexibility itself tariff changes (e.g., Kreinin (1961)). probably lie in the accommodating macroeconomic Moreover, the implication of the ratchet hypothesis policies followed by most governments over this pe- that declines in the domestic currency price of imports riod. As long as workers and producers believe that should occur very infrequently is not supported by the governments can sustain contractionary demand pol- facts. For example, Pigott, Sweeney, and Willett (1975) icies for only short periods, they will be reluctant to found that such declines occurred in about 35 percent reduce prices and wages during recessions. In this of the quarters from 1957 to 1974, at least for six major respect, one compensation of the 1979 recession is industrial countries taken together.39 Further, the fall that such beliefs are now likely to be held with much in the domestic currency price of imports following less confidence; consistent with that supposition, the specific revaluations or appreciations (e.g., the 1978 trend toward greater downward price inflexibility that appreciations of the yen and the deutsche mark and had apparently been going on since the early 1960s in the 1961 and 1969 revaluations of the deutsche mark, major industrial countries seems to have been halted etc.) also points in the same direction. in the 1979-82 period. Negative changes in import prices also do not appear to have a different effect on domestic prices than Vicious Circle Hypothesis positive changes do. Goldstein (1977), in a series of a The charge that floating rates give rise to "vicious pooled cross-section time series regressions for five and virtuous circles" is heard much less often now large industrial countries, found no evidence of an than it was seven or eight years ago (1975-76) when asymmetry in the domestic price effects of decreases disparities in economic performance between strong versus increases in import prices—and these results industrial countries (e.g., Japan and the Federal Re- held for both the gross domestic product deflator and public of Germany) and weak ones (the United King- the price of manufactures. Similarly, Finger and DeRosa dom and Italy) were particularly large. This may partly (1978) examined the relationship between final product reflect the growing awareness that inflation differentials prices and raw commodity input prices for 20 product are not the overriding determinants of exchange rate groups during the 1950-75 period and found no evi- changes. Even more so, it reflects the change in dence of a ratchet effect. Further, a recent Fund staff fortunes of some of the formerly strong and weak study (IMF (1984 a)) on exchange rate volatility and currencies. world trade could not find any significant independent Nevertheless, some important issues in the vicious effect of exchange rate variability on inflation rates 40 for the seven largest industrial countries over the circle debate warrant an examination. The basic 1958-81 period. proposition is that a depreciation immediately raises the local currency price of imports. These import price Finally, a recent study for the Organization for increases then feed quickly through to domestic and Economic Cooperation and Development (OECD) by export prices, which in turn induce higher money Encaoua, Geroski, and Miller (1983) indicates that the wages, higher domestic prices, more exchange rate trend toward greater downward price inflexibility that depreciation, etc. If the trade balance displays signif- was observed in major industrial countries since the icant J-curve effects and if the exchange rate depre- early 1960s was finally halted during the recession that ciates in response to expectations of future current began in 1979. It is thus dangerous to assume that all account deficits and of higher inflation, the inflation- cost or demand decreases (whether induced by ex- depreciation spiral can be even quicker and more adverse. The virtuous circle is just the opposite, with 39 For some of these countries, the declines were much more frequent. For example, import prices in the Federal Republic of Germany fell in 10 of the 18 years during 1956-73—clearly too often 40 See also Bond (1980) and Bilson (1979) for analyses of the to be regarded as a temporary or unusual event. vicious circle.

17

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

the appreciating exchange rate lowering import prices, that extend it—need not necessarily be the fault of the which in turn lower domestic prices, etc. Fixed rates policymaking authorities in the country with the are said to break these circles because the relatively depreciating currency. This follows from the widely stable export prices of low-inflation countries restrain accepted premise that the current exchange rate de- the increase in domestic prices in high-inflation coun- pends heavily on expectations about the future ex- tries (and conversely). change rate and from the observation that the list of A fair appraisal of the vicious circle hypothesis factors affecting these expectations is long and varied— would seem to require acknowledging the validity of including not only monetary and fiscal policies in the three aspects of it. home country but also unexpected policy changes in (1) There can be no doubt that a depreciating foreign countries, new political developments abroad, exchange rate does have a significant inflationary effect current account "news" in other countries, changes on the depreciating country's import, domestic, and in intervention practices, etc. In short, anything that export prices. Tables 7 and 8 in the Appendix, which affects the supply of (or the demand for) assets de- are adapted from Goldstein and Khan (1982), give nominated in the depreciating currency or close sub- some representative estimates of those effects for a stitute currencies could initiate an exchange rate change. sample of industrial countries (including the seven Further, even after a decade of floating rates, the largest ones). For an average industrial country, a 10 possibility of "inefficiencies" in the foreign exchange percent depreciation would induce: (a) an 8-10 percent market that magnify and prolong departures of actual increase in import prices within six months; (b) a 1.5- rates from equilibrium rates cannot be ruled out.43 4.0 percent increase in consumer prices within a year These inefficiencies may arise from risk aversion or so; and (c) a 5-8 percent increase in local currency combined with legal and regulatory constraints on export prices within two years. Also, there is a definite open foreign exchange positions (i.e., too little stabi- country pattern to these effects. The smaller, relatively lizing speculation, as suggested by McKinnon (1976) open, and more highly indexed countries obtain less and Artus and Crockett (1978)), or from a mistaken relative price advantage and more domestic inflation appraisal of fundamentals by market participants from a given depreciation than do the larger, less open (Dornbusch (1983)), or from the possibility of changes ones.41 From this perspective, it is perhaps not sur- in regime that overwhelm the authorities' good inten- prising that the countries in the former group have in tions regarding policymaking (the so-called peso prob- the main rejected independent floating in favor of less lem), or from speculative "bubbles" that continue to flexible exchange arrangements. grow larger even in the face of perceived disequilibria (2) A second legitimate point of the vicious circle because nobody knows when the crash will occur hypothesis is that floating rates probably shorten the (Blanchard (1979)). Whatever the cause, the point is time lag between money supply changes and domestic that weak currencies can be subjected to excessive downward pressure relative to longer-term equilibrium price changes because money supply changes are often 44 transmitted rapidly into exchange rate depreciation.42 levels. Tobin (1980, pp. 157-58) has provided a good As acknowledged by Wallich (1977), this problem limits summary of the problem: the scope for antirecessionary action under floating . . . foreign exchange markets are necessarily adrift with- rates, especially in an environment where most indus- out anchors. . . . In these markets, as in other markets trial countries have had to deal simultaneously with for financial instruments, speculation on future prices is high inflation and high unemployment. The problem is the dominating preoccupation of the participants. In the apt to be particularly troublesome when a country with ideal world of rational expectations, the anthropomorphic weak currency and high unemployment has to engineer personified "market" would base its expectations on in- a recovery in the face of an unfavorable foreign interest formed estimates of equilibrium exchange rates. Specu- rate differential. lation would be the engine that moves actual rates to the (3) Yet a third point of merit in the vicious circle equilibrium set. In fact no one has any good basis for hypothesis is the claim that the exchange rate move- estimating the equilibrium dollar-mark parity for 1980 or ments that begin the vicious circle—and even some 1985, to which current rates might be related. That parity depends on a host of incalculables—not just the future paths of the two economies and of the rest of the world, 41 These intercountry differences are reduced but not eliminated but the future portfolio preferences of the world's wealth if one considers deflators of value-added rather than consumer prices (see Table 8 in the Appendix). owners. ... In the absence of any consensus on funda- 42 Consistent with this proposition, Spitaller (1978) found that the response of inflation to its determinants was almost three times as fast in industrial countries from 1973 to 1976 as from 1958 to 1976 43 For a good survey of the theoretical and empirical work on as a whole. Also, Robinson, Webb, and Townsend (1979) concluded efficiency in the foreign exchange markets, see Levich (1984). that the feedback of exchange rate changes to domestic price changes 44 The report of the Working Group on Exchange Market Inter- was larger and quicker in the 1970s than in the 1950s and 1960s. vention (1983) cites a number of such "bandwagon" episodes.

18

©International Monetary Fund. Not for Redistribution Floating Rates and Macroeconomic Policy

mentals, the markets are dominated—like those for gold, economies underestimated the domestic price effects rare paintings, and—yes, often equities—by traders in of depreciation for others, and they oversold the the game of guessing what other traders are going to rationality of the market in correctly valuing weak think. currencies—a rationality that they are increasingly From another point of view, it is not hard to identify questioning now that some of their own currencies are two shortcomings of the vicious circle hypothesis. weak. At the same time, the smaller and more open (1) A first shortcoming is that the vicious circle economies underestimated the role of domestic mon- hypothesis fails to recognize that an excessive rate of etary and fiscal policies in sustaining the vicious circle domestic monetary expansion will often be the driving as well as their own capacity to escape from it by force behind both exchange rate depreciation and high altering these policies. In short, floating rates can domestic inflation. In this sense, the fact that exchange exacerbate intercountry inflation differentials, but they rates typically respond faster to money supply changes need not. than do domestic prices can create the optical illusion that exchange rate movements are causing domestic Floating Rates and Unemployment price increases, when in reality it is domestic monetary policy that is the real culprit. This point becomes As suggested earlier, the behavior of unemployment particularly significant whenever one moves beyond over the past decade has been extremely disappointing. the initial stages of the vicious circle because the Whereas the inflation rate for the seven major industrial domestic price increases induced by depreciation also countries reached a peak in 1980 and has fallen steadily reduce the real value of money balances. As asset since then, the unemployment rate for the industrial holders seek to restore these balances to the desired countries started to turn downward only late in 1983. level, they will spend less on all goods, including From an already high figure of 5.0 percent in 1979, the imports, creating an incipient current (or capital) ac- unemployment rate for the seven largest industrial count surplus and an exchange rate appreciation (and countries rose to 5.7 percent in 1980, 6.4 percent in 1981, and 8.1 percent in 1982; the forecast for 1983 is this will be created even if no relative price advantage 45 is obtained from the depreciation). That is why it is 8.8 percent. For the 1970s as a whole, there appears unusual for a vicious circle to be sustained without to have been a secular increase in the long-duration accommodating money supply behavior. Most studies unemployment rate (Haveman (1978)), an increased of policy behavior prior to vicious circles find evidence mismatch at the margin between job vacancies and of just such monetary accommodation (see Bank for unemployed workers (i.e., structural unemployment) International Settlements (1976) and Gordon (1977)). (Deppler and Regling (1979) and Medoff (1983)), and a It is also the basis for Haberler's (1980, p.31) conclusion perceptible worsening in the short-run trade-off be- that ". . . countries are not by chance on one side or tween inflation and unemployment (Wachter (1976) the other [of the vicious/virtuous circle]." and Sachs (1983)). Critics of floating rates have pointed to two channels (2) The second major deficiency of the vicious circle by which floating rates could increase unemployment: argument is that it puts forward too simplistic a view (1) by inducing labor to shift back and forth between of exchange rate determination. Empirical studies tradable and nontradable goods industries in response show that month-to-month changes in exchange rates to transitory relative price changes attributable to are not well correlated with month-to-month inflation short-run exchange rate variability (i.e., by increasing differentials. Indeed, whenever an exchange rate changes frictional unemployment); and (2) by promoting longer- substantially over a short time span, this change is term but reversible real exchange rate disequilibria almost always accompanied by a significant divergence that leave deindustrialization in their wake (i.e., by from purchasing power parity (Mussa (1983)). Often, 46 increasing structural unemployment). the key variable is the market's evaluation of the prospects for monetary and fiscal policies in the weak country. The turnaround in the U.S. dollar after the 45 See IMF (1983, Table 6)). Using fourth quarter figures rather November 1, 1978 package of measures in the United than annual averages, the unemployment rate for 1983 would be lower than for 1982. States is perhaps the classic case in point, but there 46 It was also sometimes argued that floating rates would reduce are other notable escapes from the vicious circle as unemployment by permitting some countries to maintain higher well (e.g., the pound sterling after the acceptance of rates of inflation than would be possible under fixed rates. This argument has lost its force, however, with the acceptance of the the Fund's stand-by arrangement in late 1976). vertical nature of the long-run Phillips curve (see Santomero and To summarize, with the benefit of hindsight, it is Seater (1978) for a survey of the relevant empirical evidence). In other words, if the natural rate of unemployment is independent of clear that each side in the original vicious circle debate the rate of inflation, then flexible rates cannot buy high-inflation undervalued the arguments of the other. Based on countries more employment, nor can fixed rates cost less than their own experience, the larger and less open industrial employment (see Artus and Young (1979)).

19

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

The frictional unemployment argument is not con- etc.)—this same boom-bust cycle can generate an vincing. After some initial experience with exchange increase in structural unemployment. rate volatility, it would be expected that workers and Having said that, it is going too far to attribute the employers would respond only to wage and employ- bulk of the employment troubles in traditional export ment opportunities that they regarded as permanent. industries (so-called deindustralization) to floatingrate s The essence of the problem is that they will not be for at least three reasons. able to make such a distinction between permanent (1) The employment effects of undervaluation and and transitory employment opportunities ex ante with overvaluation of exchange rates are not peculiar to a high accuracy. Nevertheless, given the fixed costs floating exchange rate regime. In this regard, it is associated with changing jobs, workers and employers worth recalling that the latter part of the Bretton are apt to be cautious, especially if they have been Woods era (1969-73) also witnessed real exchange rate adversely affected by previous short-run fluctuations changes for major currencies (the U.S. dollar, the in relative prices. Also, one response to increased deutsche mark, the yen) on the order of 20-30 percent, uncertainty about demand conditions, whether induced and that these real exchange rates were apparently by exchange rates or otherwise, is to hold larger associated with serious distortions in the pattern of inventories of labor (Miller (1971)). Indeed, such in- employment, output, and investment (e.g., Dunn (1973) creased labor hoarding would, other things being equal, and Makin (1974)). lead to a fall in measured unemployment while simul- (2) There were again structural changes during the taneously reducing labor productivity. All in all, short- period of floating rates that had important effects on run exchange rate volatility, by increasing the "noise" both export competitiveness and output and employ- and reducing the "signal" in relative price movements, ment growth in traditional export industries. To take is apt to reduce the efficiency of resource allocation, only two, the real appreciation of sterling from 1976 but probably not to an important degree—and not with to 1981 surely owed something to the discovery and any major implications for unemployment. exploitation of North Sea oil reserves. In this sense, Unfortunately, no such comfort can be taken with at least some of the contraction of the U.K. manufac- respect to the potential unemployment and resource turing industry in the late 1970s represented an equi- allocation effects of longer-term exchange rate dise- librium response to the increase of domestic oil pro- quilibria—that is, disequilibria that last two to three duction and prices.49 In the case of U.S. manufacturing years or longer. Here, because the time frame is longer, industries, a recent study by Lawrence (1983) reaches there is a strong presumption that individuals and firms the following interesting conclusions: (a) the secular will be prepared to overcome the fixedcost s of switch- decline in manufacturing's share of total employment ing resources, especially if the inducements are large. during the 1970s was due mainly to changes in the The real effective exchange rate of the pound sterling domestic composition of output (a revealed preference fell by about 20 percent between 1975 and 1976 and for services) and to the more rapid increase of pro- then rose by close to 75 percent between 1976 and ductivity in manufacturing, and not to foreign trade; 1981.47 Similarly, the real effective exchange rate of (b) in fact, during 1973-80 foreign trade provided a net the U.S. dollar fell by about 10 percent between 1976 addition to output and jobs in U.S. manufacturing; and 1978 and then rose by about 30 percent between and (c) only from 1980 to 1982 did foreign trade 1979 and 1982. To the extent that swings in competi- contribute to the employment decline in manufactur- tiveness of this magnitude exceed movements in real ing—and then it accounted for perhaps a third of the equilibrium exchange rates, there will obviously be an total fall in manufacturing employment. unwarranted cycle in export- and import-competing (3) If a country attaches significant social welfare sectors.48 Furthermore, because resources (especially to the composition of employment within the tradable labor) cannot be reallocated quickly and without cost sector, then the exchange rate is not the proper policy from the tradable to the nontradable sector— or even instrument for ensuring that objective. This is true from some slower-growing tradable industries (e.g., because, whereas the exchange rate can alter the U.K. engineering, and U.S. steel and automobiles) to relative prices of tradables and nontradables, or some- other faster-growing ones (e.g., computers, oil refining, times the relative prices of imports and exports, it

47 The real effective rate used here is the ratio of own to 49 See Forsyth and Kay (1980), who estimate that the equilibrium competitors' normalized unit labor cost adjusted for exchange rate response to the discovery of North Sea oil was a real appreciation changes. of 18 percent. Bond and Knobl (1982) estimate that perhaps half of 48 The same kind of argument can be made about the effect of the real appreciation in sterling between 1977 and 1981 was due to the wrong exchange rate on variations in consumption (i.e., the so- the existence of North Sea oil and to the rise in real price of oil. called welfare cost of disequilibrium exchange rates). See Hause Buiter and Miller (1983) and Williamson (1983) argue for lower (1966). figures.

20

©International Monetary Fund. Not for Redistribution Floating Rates and Macroeconomic Policy cannot alter the relative prices of different classes of exchange rate with a relatively independent monetary exports. Hence, if a country wants to preserve em- policy had become painfully obvious. Nowhere was ployment or slow its decline in some traditional export this more apparent than in the Federal Republic of industries while still permitting overall external ad- Germany and Switzerland, where restrictive monetary justment, it will need to supplement exchange rate measures (taken in large part to avoid imported infla- policy with some more disaggregated scheme of taxes tion) induced capital inflows, official intervention to and/or subsidies. support the U.S. dollar, more restrictive monetary In summary, the historically high unemployment measures, more capital inflows, etc.—creating, in ef- rates that have characterized the period of floating fect, a monetary vicious circle. In February and March rates are best explained by (a) cyclical conditions of 1973 alone, the Deutsche Bundesbank purchased (especially the tight monetary policies existing since $8.5 billion—only to succumb to floating rates the next 1979); (b) changes in the growth of labor supply;50 (c) month. The motive for that decision was clear, as later changes in the demographic, occupational, and indus- confirmed by Emminger (1977, p. 4): trial composition of labor supply and demand; (d) the For countries like Germany and Switzerland, the main— high level of real wages relative to labor productivity; or even only—reason why they went over to floating in and (e) the growth in generosity and coverage of 51 the spring of 1973 was the necessity to regain control unemployment benefits. Real exchange rate move- over their own money supply . . . ." ments surely had an important influence on sectoral employment (i.e., in export industries and import- The message that a country could not simultaneously competing ones), but their contribution over the period maintain a fixed exchange rate, allow freedom for of floatingrate s as a whole to aggregate unemployment international capital movements, and have an inde- appears modest in comparison with other factors.52 pendently determined money supply was not new (e.g., Kouri and Porter (1974); Frenkel and Johnson (1976); and IMF (1977)). It had long been recognized in the Floating Rates and Monetary Policy monetary approach to the balance of payments that, At the time of the move toward greater exchange under fixed rates, decreases (increases) in domestic credit would be offset by increases (decreases) in rate flexibility, there was great optimism about what 53 floating rates would do for the effectiveness of mon- international reserves. Further, for a small country, etary policy. That optimism was essentially based on this offset would be complete so that the authorities two arguments: (1) that floating rates would enable would be able to control the composition of the money countries to regain the control over their own money supply (i.e., the mix between the domestic and foreign supplies that they had lost under fixed rates; and components of the money supply) but not its level; if (2) that floating rates would strengthen the output and the country was large or if there was less than full employment effects of expansionary monetary policy employment, the offset would be only partial, because via the positive effects of the induced exchange rate domestic credit would affect domestic prices and depreciation on the trade balance. Ten years later, output as well as the balance of payments (e.g., Aghevli even the staunchest defenders of floating rates had to and Rodriguez (1979)). concede that much of that optimism was misplaced. Considerable empirical literature now exists on the behavior of international capital flows and on the Control of the Money Supply sterilization attempts of monetary authorities during the period of fixed rates.54 While estimates vary con- Perhaps the main reason why floating rates looked siderably across studies, as a group they suggest that so appealing in the latter years under the Bretton industrial countries found it possible, but at times very Woods system was that the incompatibility of a fixed difficult, to control their money supplies under fixed rates,55 with perhaps the monetary authorities in the Federal Republic of Germany, Switzerland, Belgium, 50 Medoff (1983) notes, for example, that the U.S. civilian labor force grew by 1.1 percent a year in the 1950s, 1.7 percent in the Austria, and France having more trouble than those 1960s, and 2.5 percent in the 1970s; also the forecasts of the U.S. Department of Labor, Bureau of Labor Statistics (1984), for the 1980s and 1990s are 1.4 percent and 0.5 percent, respectively. 51 Consistent with this conclusion, Sachs (1983) finds that he can 53 By a "small" country, it is meant one in which the price level explain the behavior of aggregate unemployment rates in six major and the interest rate are basically determined in the rest of the industrial countries over the 1961-81 period by reference to real world. money balances, the excess of real wages over trend productivity, 54 For recent reviews, see Kreinin and Officer (1978) and Laney lagged unemployment, and a time trend. and Willett (1982). 52 Because of the size of the real appreciation of sterling between 55 Obstfeld (1982), for example, found that capital flows offset 1977 and 1981, the United Kingdom probably stands as an exception about 50-65 percent of the change in the Deutsche Bundesbank's to this conclusion. net domestic assets during the 1960-70 period.

21

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

in Japan, the United States, the United Kingdom, and Effectiveness of Monetary Policy Italy (Hickman and Schleicher (1978) and Laney (1979)). The second aspect of the case for floating rates as Then, what about the control of the money supply under floating rates? The presumption of greater con- a boon to monetary policy stems from two early trol is derived from the absence of any obligation to theoretical results of Fleming (1962) and Mundell use exchange market intervention to peg the exchange (1968). The first result is that, under conditions of high rate. Thus, exchange market pressures take the form capital mobility, a given increase in the money supply of price changes (exchange rate changes) rather than produces a larger increase in income under floating volume changes (reserve movements) and the foreign rates than under fixed rates. The argument unfolds as component of the monetary base ceases to be a source follows. Under floating rates, the money supply in- of changes in the money supply. crease yields a temporary fall in domestic interest The difficulty, however, is that the authorities must rates relative to foreign rates. This induces an incipient regard the exchange rate exclusively as a policy in- capital outflow, a depreciation of the exchange rate, strument and not as a target. The more they manage an improvement in competitiveness, and an expansion the exchange rate, the more they relinquish the added in net exports. In contrast, under fixed rates, the same degree of freedom. As noted earlier, the heavy amount interest rate differential induces a realized capital of exchange market intervention conducted by indus- outflow that restores the original money supply and trial countries during the past decade stands as testi- the domestic interest rate, thereby preventing any mony that policymaking authorities do not regard the effect on the domestic level of income. The second either/or choice between control of the money supply result is that, under flexible rates and high capital and control of exchange rates as acceptable. Instead, mobility, expansionary monetary policy has a com- their behavior reveals a preference for an intermediate parative advantage in raising the level of domestic solution, where the authorities keep an eye on both income over expansionary fiscal policy. The difference targets.56 This suggests that, under floating rates, is that, whereas the former is accompanied by a fall countries have had more control over the money supply in domestic interest rates, the latter is accompanied than under fixed rates but that the difference is some- by a rise. As a result, the initial income stimulus under what less marked than early supporters of floating fiscal policy is choked off, or at least blunted, by rates had anticipated.57 Two observers sum up recent currency appreciation, but with monetary policy it is experience: reinforced by currency depreciation. Whatever its merits as a representation of how From the experience of the past seven years, it is also monetary policy might have worked under flexible apparent that the behavior of exchange rates influences rates in an environment similar to that of the 1960s, the conduct of monetary policy, but usually only after the Mundell-Fleming model has at least four weak- exchange rates have moved substantially away from what nesses when applied to the 1970s or 1980s. the authorities regard as appropriate or desirable values. (Mussa (1981, p. 24)) (1) The first is the assumption that exchange rate changes translate quickly into changes in competitive- ... in many countries the exchange rate has achieved a ness. This would be true if feedbacks from the exchange comeback in the minds of policy-makers as one of the rate to domestic factor costs and prices were insignif- most important prices in the economy. Not that a fixed icant. As indicated earlier, however, the empirical (or a fixed but adjustable) rate has again become a policy evidence shows that such feedbacks do exist and that goal. . . . Thus, we have seen cases where monetary policy in general, and interest rate policy in particular, they can be sizable, especially in the smaller, more became largely geared to the exchange rate; not only in open, and more highly indexed industrial countries. smaller countries like Belgium or Austria, but also in The greater are these feedbacks, the smaller is the countries like Britain and Germany. (Emminger (1982, competitive price advantage achieved by depreciation p. 2)) and hence the smaller is the expansion in net exports. In fact, when real wages are rigid and unaffected by exchange rate changes, expansionary monetary policy will affect only prices and the exchange rate, but will 56 Argy (1982) provides an account and appraisal of how this have no effect on real output, employment, or the mixed strategy worked during the period of floating rates in the 58 Federal Republic of Germany, Japan, and the United Kingdom. trade balance. This problem is particularly relevant 57 Simple calculations of the standard deviation (or coefficient of variation) of money supply growth across the seven major industrial 58 See Argy and Salop (1979), Branson and Rotenberg (1980), and countries show that the intercountry dispersion has increased Sachs (1983). More generally, the effects of expansionary monetary marginally in moving from 1963-72 to 1973-82. Using more sophis- and fiscal policies on real output also depend on whether workers ticated measures of country synchronization, Swoboda (1983) finds bargain for after-tax real income and whether workers and producers no significant change in monetary interdependence across exchange use different price deflators for calculating real wages (see Argy and rate regimes. Salop (1979) and Dornbusch (1983)).

22

©International Monetary Fund. Not for Redistribution Floating Rates and Macroeconomic Policy

because there are indications that real wage rigidity is of account and medium of exchange functions—is much greater in Europe than in the United States, replaced by stronger currencies. Thus, as recognized with Japan being in the middle (Sachs (1983); Grubb, in the so-called currency substitution literature, high Jackman, and Layard (1982)). This implies that the asset substitutability can limit the scope for expan- United States gets more advantage from exchange rate sionary monetary policy (Calvo and Rodriguez (1977); movements than do the Europeans. Branson (1983, Kareken and Wallace (1978); Brillembourg and Schad- p. 58) summarizes this argument: ler (1979)). (4) Yet a fourth necessary amendment to the Mun- These results suggest a pattern of differences in adjust- dell-Fleming model is the incorporation of exchange ment to exchange rate changes between Europe and the rate expectations into the choice between domestic United States . . . an exchange rate change will move and foreign assets. Specifically, under floating rates, relative prices and the balance on current account in the United States, and also influence output, all in the asset holders will not necessarily select the asset with expected "stabilizing" direction. In Europe, however, a higher nominal interest rate, unless the interest rate the movement in the exchange rate will mainly move the differential exceeds the expected depreciation of that overall price level, with minimal effects on the trade currency relative to the one with a lower interest rate. balance or output. So the exchange rate is reasonably The important implication of this familiar interest-rate- viewed as an effective instrument for stabilizing the parity condition is that monetary policy operates on current account in the United States. In Europe, however, exchange rates not only via its direct effect on interest exchange rate fluctuations are equally reasonably viewed rates but also via its indirect effect on expected future as essentially destabilizing the price level. The result is exchange rates. For example, success in halting a policy conflict based on different implicit assumptions depreciation with restrictive monetary policy is likely about the underlying structure of labor markets and wage to hinge as much on convincing the market that this behavior. policy is relatively permanent (thus affecting the future (2) The second weakness of the Mundell-Fleming exchange rate) as in maneuvering a favorable interest model is the assumption that changes in competitive- rate differential. The shortcoming of looking only at ness will yield rapid improvements in the depreciating nominal interest rates is perhaps best illustrated by country's trade balance. Because in the short run (less noting that the U.S. dollar was depreciating relative than a year) import prices rise more rapidly than export to the deutsche mark from mid-1976 through most of prices in response to depreciation and there has not 1978 despite a rise in U.S. interest rates relative to been enough time for the volume of trade to adjust those abroad (Dornbusch (1979)). A related point of very much, it is quite common for the response of interest is that the movement of the exchange rate trade balance to depreciation to follow a J-curve itself can be a useful indicator of the appropriate stance (Spitaller (1980)). During this short run, the stimulat- of monetary policy under floating rates. Because the ing effects of monetary expansion will thus be reduced, nominal interest rate reflects the sum of the real rate not strengthened, by depreciation (Niehans (1975)). of interest and the expected rate of inflation, a rise in Over time the initial perverse trade balance effects the nominal interest rate could reflect an increase in will be checked and then reversed as export price either of the two unobservable factors, each with increases catch up with import price increases and as different implications for the stance of monetary policy the responses of imports and exports grow large in (Mussa (1981) and Frenkel (1983 c)). The exchange volume. Nevertheless, this means that, if anything, rate may help to disentangle that puzzle. If the ex- the comparative advantage of expansionary monetary change rate is appreciating in the face of a favorable policy lies in the medium to long run, not in the short interest rate differential, it signifies a rise in the real run. rate of interest that may call for easing of domestic monetary policy relative to that abroad; on the other (3) A third limitation relates to the size (and unpre- hand, a joint indication of a currency depreciation and dictability) of the exchange rate change induced by a favorable interest rate differential implies that infla- domestic monetary expansion when domestic and tion expectations are the culprit, thus signaling, ceteris foreign assets (including currencies) are close substi- paribus, monetary restraint. tutes. The point here is that exchange rate changes can go much further than the authorities would like. It is thus clear in retrospect that the case for For example, depreciations that were looked on with monetary policy under floating rates was oversold. favor because they would diminish current account Many of the perceived constraints on monetary policy deficits can become cause for concern, as asset holders during the period of fixed rates turned out not to be switch out of the weak currency, and as the weak constraints imposed by the exchange rate regime but currency's store of value—and perhaps even its unit rather constraints imposed by the openness of national

23

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM economies. As succinctly put by Frenkel (1983 c, world of high capital mobility in the 1970s and 1980s. p. 49): Indeed, one of the key messages of the asset market . . .These constraints are reflected in either a reduced view of exchange rates is that anything that affects ability to influence the instruments of monetary policy asset supply or asset demand can alter exchange rates (like the nominal money supply under fixed exchange and thus affect real variables (i.e., real output and rates), or in a reduced ability to influence the targets of employment) in both the home and the foreign country. monetary policy (like the level of real output), or in an In this sense, the relevant question is not whether increased prudence in the use of monetary policy because floating rates can transmit foreign disturbances but of the potentially undesirable effects on expectations. rather how they do so vis-a-vis fixed rates. Two rather well-known results from the theoretical In view of that, those who counsel that monetary 61 policy be directed more toward stabilization of the literature are worth repeating: (1) a foreign monetary exchange rate need to consider what policy instruments disturbance will have opposite effects on foreign and will then be directed toward domestic objectives.59 In domestic output under floating rates but will move this respect, the past record of fiscal policy hardly output in the same direction under fixed rates; and makes it an attractive, especially in view of the (2) a foreign expenditure disturbance, whether induced seemingly structural nature of some present-day budget by fiscal policy or otherwise, will be transmitted to deficits. Thus, while the comparative advantage of domestic output with greater strength under floating monetary policy under flexible rates is undoubtedly rates than under fixed rates. In brief, these results are smaller than originally thought, it is still evident, based on the assumption that foreign monetary expan- especially in the case of countries that conduct mon- sion lowers the foreign interest rate while a foreign etary policy in a stable and responsible way. expenditure disturbance raises it; hence, the two types of disturbances produce opposite exchange rate move- Floating Rates and Insulation Against ments which, in turn, imply opposite net trade balance Shocks and real output effects for the home country. Perhaps the key policy implication of these theoret- This is another area where some initial expectations ical results is that countries cannot rely on floating about the potential of floating rates have been disap- rates to protect them from foreign policy changes. pointed. The expectation was that floating rates would Instead, if they want such insulation, either they or provide effective insulation against a wide variety of the foreign country must take some countervailing foreign shocks or disturbances, thereby permitting action. For example, if the home country wants to macroeconomic policy to concentrate on combating prevent foreign monetary expansion from reducing disturbances of domestic origin. In fact, it has been home output or foreign fiscal expansion from increasing rather forcefully demonstrated that, while floatingrate s home inflation, it will have to prevent an interest rate alter the nature of the transmission process for foreign differential from appearing so as to stabilize the ex- disturbances from that under fixed rates, they by no change rate and thereby choke off the main channel means eliminate such transmission effects. Also, while of transmission. Similarly, if an autonomous shift in floating rates provide better insulation against certain asset demand is not to impinge upon domestic policy, types of foreign disturbances than fixed rates do, either the home country or the foreign country will insulation is worse against other types. While many have to alter the relative supply of assets of home and factors are relevant for assessing the insulation prop- foreign currency so as to offset these demand shifts. erties of alternative exchange rate regimes, two that In principle such countervailing actions can always be deserve special attention because of their prominence devised, but there may be formidable constraints on in the 1970s are the degree of international capital their practical application (e.g., fiscal policy may not mobility and the distinction between real and monetary be flexible enough, especially in the direction of re- shocks.60 straint, to make rapid changes in the policy mix; or the substitutability of domestic and foreign assets may Capital Mobility and Insulation be so high as to preclude small changes in relative asset supplies from having much of an effect on The notion that floating rates can insulate a country exchange rates; or the home country might not want from foreign disturbances is not a bad working as- to follow the foreign country's policy lead because its sumption if the international mobility of capital is low, domestic unemployment/inflation picture is different). but it is serious misrepresentation when applied to the In any case, the essential point is that, with high 59 Solomon (1983) expands upon this argument. international mobility of capital, countries will be 60 Other significant factors are the country's access to international capital markets and the degrees of openness, of factor mobility, of export diversification, and of wage-price flexibility. 61 See Mussa (1979) for a demonstration of these results.

24

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment

obliged to work hard to obtain a reasonable degree of tries has typically been somewhat higher under floating insulation from foreign disturbances and, even then, rates.63 Such evidence, however, is consistent not only many constraints will preclude complete insulation. with greater transmission of disturbances under floating rates but with other hypotheses as well, including the Monetary Versus Real Shocks greater incidence of common external shocks and the usual policy responses to them in the period of floating A second reason why the insulating properties of rates. This is, for example, the explanation favored floating rates may have been overestimated is that by Artus (1983) in interpreting Swoboda's (1983, floating rates seem to have a comparative advantage p. 103) evidence of increased synchronization: against monetary or overall price level shocks—the types of shocks that probably predominated in the In my view it is largely because of these common 1950s and 1960s. For example, a floating rate provides tendencies [the increase in the size of social transfers, good potential for insulation against a rise in the world the policy of monetary accommodation of wage and price price level because an appreciation of the domestic increases, the growth of rigidities in labor markets, and currency proportionate to the increase in foreign prices the decrease in the share of income going to capital, mainly in Europe], and because of the two waves of oil prevents wealth or relative price effects from taking price increases, that industrial countries have all jointly place. Floating rates cannot, however, provide effec- moved into a period of stagflation since 1973. All being tive insulation against relative price changes among in the same situation, it is not surprising either that they different classes of traded goods (e.g., between oil or have tended to adopt fairly similar policies. food and other tradables) because they cannot alter relative prices at that level of disaggregation. In fact, To summarize, the past ten years make it clear that floating rates have a comparative disadvantage in the old view of floating rates as premier insulators protecting against real shocks because, unlike fixed against a whole range of foreign disturbances is inac- rates, they effectively prevent the balance of payments curate. Instead, floating rates should be considered as from serving as a cushioning device to smooth domestic having a comparative advantage against some types consumption (Frenkel and Aizenman (1982); Flood of disturbance and a comparative disadvantage against and Marion (1982)). These distinctions are not aca- others. Also, because floating rates cannot provide demic because, as is well known, the most important complete insulation against the representative bundle shocks of the 1970s were real shocks involving large of foreign disturbances, the case for policy activism changes in the relative price of tradable commodities.62 to combat such disturbances (including in some in- In this connection, it is worth reporting that theoretical stances greater exchange market intervention) is thereby models that simulate the effects of a relative price strengthened. By the same token, the case for coor- increase for an important intermediate input (e.g., dination of policies is also strengthened so as to energy products) in an economy with floating rates minimize conflicts between countries' policy actions. and sticky real wages typically find that the authorities are powerless to protect the economy from some Floating Rates and External Adjustment increase in unemployment, except under some restric- tive conditions (Buiter (1978); Argy and Salop (1979)). Attention is now shifted from domestic economic In the real world, countries will be faced with both objectives toward the role of the exchange rate system monetary and real shocks, and they will not know in in securing external payments adjustment. The focus advance which types of shock will predominate. In is on two broad questions: (1) Has the extent of this situation, an intermediate degree of exchange rate external payments adjustment during the period of flexibility can be optimal, but exchange rate policy floating rates been different than that during the last should not be expected to produce complete insulation decade of adjustable par values? and (2) How does from either real or monetary shocks. the process of external adjustment itself operate under From an empirical viewpoint, there is no evidence managed floating? to suggest that the period of floating rates has been characterized by a weaker international transmission The Degree of External Adjustment of disturbances than was the period of adjustable par As argued earlier, probably the best way to identify values. Studies by Ripley (1979), Hickman and and to measure the degree of external payments Schleicher (1978), and Swoboda (1983) all find that synchronization of movements in real economic activ- 63 The only macroeconomic variable that seems to show markedly less synchronization in the period of floating rates is the rate of ity and of monetary variables among industrial coun- inflation (see Swoboda (1983)). Even here, however, the data may also be reflecting the tendency for the variability of inflation, both 62 To the extent that fiscal imbalances affect real interest rates, within and across countries, to increase with the mean rate of errant fiscal policies can also be considered as real disturbances. inflation (see, for example, Logue and Willett (1976)).

25

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

adjustment is to compare a country's actual balance remembering that the external adjustment that was of payments with an estimate of the equilibrium balance achieved under managed floating took place at a much of payments. The equilibrium payments balance can, lower level of capacity utilization and at a much higher in turn, be defined as one where the current account rate of unemployment than during the preceding decade equals normal net capital flows, after adjustment for (see Table 1). While both of these factors would be the effects of (a) temporary factors (such as dock incorporated into a proper measure of the equilibrium strikes or bad harvests), (b) abnormal capacity utili- payments imbalance, they are either excluded or only zation or unemployment, (c) permanent exogenous partially accounted for in the crude measures. changes in the terms of trade, and (d) undue restrictions or incentives on trade and capital movements. If the actual payments balance is close to the equilibrium Current Account Imbalances Alone payments balance, then the presumption is that exter- As suggested earlier, serious problems are associated nal payments adjustment has been satisfactory; if not, with measuring external payments adjustment by ref- it implies that there has been a lack of adjustment. erence to the current account alone. Nevertheless, in Similarly, by extending the same methodology to a view of (a) the elusive nature of normal capital flows, group of countries, some judgment can be reached (b) the prominence given to the current account as a about the effectiveness of external adjustment for the barometer of the need for adjustment in some earlier system as a whole. periods,66 and (c) the availability of relatively long time series data, there is merit in reviewing the historical Unfortunately, two formidable practical problems record. limit the analysis of external payments adjustment. Table 2 presents three characteristics of current The first one is that a consistent and reasonably long account imbalances (in relation to gross national prod- time series on equilibrium payments balances simply uct) for seven larger and eight smaller industrial coun- does not exist, even for the large industrial countries— tries during the period of floating rates (1973-82) certainly not one long enough to compare the period and during the last decade of adjustable par values of floating rates with the period of adjustable par (1963-72).67 The mean ratio of the current account to 64 values. Faced with this situation, a second-best gross national product is employed as a rough indicator procedure has been adopted. Two crude measures of of the average degree of current account adjustment equilibrium payments balances that could be extended over the period,68 while the standard deviation and back to 1963 or 1965 were constructed, so that a first-order serial correlation statistics proxy the vari- comparison could be made across exchange rate re- ation and persistence of these current account imbal- gimes. Next, the available estimates, based on more ances over the period. Ceteris paribus, external ad- comprehensive definitions of equilibrium payments justment is assumed to be less satisfactory, the larger balances, were examined for whatever light they could are these three summary statistics.69 shed on the degree of external adjustment within the The story told by Table 2 is straightforward. For period of floating rates. the larger industrial countries, there is no indication that current account adjustment has been less satis- The second problem is the same one encountered factory under floating rates than under the Bretton earlier in analyzing the role of the exchange rate system Woods system. In fact, the figures in Table 2 suggest in facilitating the pursuit of domestic economic objec- that average current account imbalances for the larger tives—namely, how to hold other (nonexchange rate countries have been noticeably smaller and less per- system) things equal so that the effect of the exchange sistent (although with larger variation) under floating rate system is not confused with the period of managed floating. In this respect, the external disturbances faced 66 Salop and Spitaller (1980) provide an interesting sample of by industrial countries during the 1973-82 period un- official pronouncements on the current account during the 1970-79 period. doubtedly made external adjustment more difficult in 67 Expressing current account imbalances as a ratio to GNP not that period than in the preceding decade. For example, only serves as a scale adjustment for cross-country comparisons the average annual percentage change in the terms of but also makes some adjustment for current account changes achieved via abnormally high or low growth rates of real income. trade for industrial countries was 0.3 for 1963-72 but 68 65 In calculating the mean ratio of the current account to GNP for -1.6 for 1973-82. On the other hand, it is worth each individual country, account is taken of the sign of the current account. In contrast, in computing the group average, absolute values of the ratio are employed. 64 As detailed later, there are few estimates of equilibrium 69 While the assumption that the optimal current account is zero payments balance for industrial countries and they typically begin is difficult to defend in principle, it is interesting to note that Penati only in 1975. and Dooley (1983) found that, for 19 industrial countries as a group, 65 The analogous figures for the seven major industrial countries there was no long-run tendency for the ratio of the current account were 0.4 for 1963-72 and -2.1 percent for 1973-82. to GNP to depart from zero over the 1949-81 period.

26

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment

Table 2. Current Account Imbalances as a Percentage of GNP: Selected Industrial Countries, 1963-72 and 1973-82 1 Mean2 Standard Deviation Serial Correlation3 Country 1963-72 1973-82 1963-72 1973-82 1963-72 1973-82 Larger countries United States 0.30 0.05 0.46 0.57 0.86* 0.19 Canada -0.85 -1.31 1.17 1.16 0.60* 0.74* Japan 0.79 0.20 1.15 1.02 0.83* 0.22 Germany, Fed. Rep. 0.53 0.20 1.07 1.63 0.61* 0.74* Italy 1.90 -0.96 1.47 2.19 0.81* 0.34 United Kingdom 0.21 -0.19 1.11 1.85 0.46 0.62* France -0.184 -0.25 0.574 1.18 Unweighted average 0.68 0.46 1.00 1.37 0.705 0.485 Smaller countries Australia -2.62 -3.04 1.71 1.92 0.83* 1.01* Austria -0.48 -1.65 0.71 1.33 0.45 0.68* Denmark -1.95 -3.27 1.15 1.13 0.68* 0.96* Finland -1.44 -2.35 1.24 2.78 0.66* 0.73* Netherlands 0.04 1.37 1.14 2.07 -0.12* 0.75* Norway -1.88 -4.19 1.78 5.95 0.03 0.84* Spain -0.50 -1.90 1.39 2.05 0.63* 0.73* Sweden -0.24 -1.64 0.85 2.01 0.61 0.80* Unweighted average 1.14 2.42 1.26 2.39 0.50 0.81 1 Current account includes goods, services, and all current transfers, both private and official. 2 Country means take into account the sign of current account imbalances. In contrast, group means are based on absolute values of country means. 3 Statistic reported is the estimated coefficient on the lagged dependent variable in the first-order autoregressive equation; * indicates statistical significance at the 95 percent level. 4 1967-72 only. Prior to 1967 balance of payments data for France included data for overseas territories. 5 Excludes France. rates than during the 1963-72 period. At the same or if the partial data for France are excluded from the time, the current account performance of the smaller calculations. industrial countries does seem to have deteriorated Chart 3 provides some complementary informa- during the period of floating rates, with larger average tion on current account adjustment in 1973-82 versus current account imbalances, greater amplitude in cur- 1963-72 by showing the cumulative current account rent account swings, and more year-to-year persistence imbalance for each of the sample countries. This time, in current account imbalances.70 In interpreting Table the imbalances are not scaled by gross national pro- 2, it is also worth reporting that the main qualitative duct (GNP) and the calculation starts fresh in each conclusions for both country groups are unaffected if period. Most interesting, and not previously hinted at the 1953-62 period is substituted for 1963-72,71 or if in Table 2, is how long it takes for initial current official transfers are excluded from the current account, account imbalances to be reversed—under both float- or if a rough adjustment for the global current account ing rates and adjustable par values. For whatever asymmetry is made for the larger industrial countries,72 reason, when such a complete reversal does take place, it typically requires from three to seven years—hardly 70 In interpreting the results for the smaller industrial countries, a rapid adjustment. it is useful to recall that most of these countries adopted some form of limited flexibility or pegged exchange arrangements during the past decade. Thygesen (1979) provides an analysis of exchange rate policy for these countries over the period of floating rates. Current Account Positions and Normal 71 For example, the mean ratios of the current account to GNP Capital Flows during 1953-62 were 1.17 and 0.69 for the larger and smaller industrial countries, respectively. Also, Dornbusch (1980) reached the same One important drawback of measuring external ad- conclusion for the four largest industrial countries in a comparison of the 1960-73 and 1973-79 periods. justment by considering only the current account is 72 In making this adjustment for the statistical asymmetry, it was that it ignores the possibility of continuing intercountry assumed: (1) that there was no asymmetry in the 1963-72 period, differences in savings behavior and in real rates of and (2) that for each year during the 1973-82 period, the share of each large industrial country in the global asymmetry could be return on investment. Such intercountry differences approximated by its share of world exports during 1973-82. These make it possible for a country with a relatively low figures for asymmetry were then added to the current account for domestic savings rate but with relatively attractive that year to derive a new ratio of current account to gross national product for each large industrial country. The results were very domestic investment opportunities to run a persistent similar to those reported in Table 2. current account deficit by drawing on foreign savings.

27

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

Chart 3. Cumulative Current Account Balances for Selected Industrial Countries, 1963-72 and 1973-82 1 (In billions of domestic currency)

1 Current account includes goods, services, and all current transfers, both private and official. 2 In trillions of yen.

28

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment

Chart 3 (continued). Cumulative Current Account Balances for Selected Industrial Countries, 1963-72 and 1973-82 1 (In billions of domestic currency)

29

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

Chart 3 (concluded). Cumulative Current Account Balances for Selected Industrial Countries, 1963-72 and 1973-82 1 (In billions of domestic currency) 40

30

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment

Table 3. Current Account Imbalances Plus Normal Private Net Capital Flows as a Percentage of GNP: Selected Industrial Countries, 1965-72 and 1973-811 Mean2 Standard Deviation Serial Correlation3 Country 1965-72 1973-81 1965-72 1973-81 1966-72 1973-81 Larger countries United States -0.62 -0.60 0.60 0.66 1.51* 0.57* Canada 0.25 -0.26 1.15 0.75 0.17 -0.29 Japan 0.94 0.04 0.93 1.00 0.86* 0.19* Germany, Fed. Rep. 0.75 0.35 0.97 1.31 0.85* 0.54* Italy 0.64 0.00 1.00 2.51 0.29 0.27 United Kingdom 0.07 -0.13 1.57 1.51 0.62* 0.25 France 0.294 0.18 1.104 0.90 ...... Unweighted average 0.51 0.22 1.05 1.23 0.72 0.35 Smaller countries Denmark 0.21 0.31 1.03 1.16 0.19 -0.11 Netherlands 0.76 0.41 1.11 1.53 0.33 0.42 Norway 0.95 1.64 1.73 3.11 0.02 0.69 Sweden -0.01 0.18 0.72 1.16 0.47 -0.04 Unweighted average 0.48 0.64 1.16 1.74 0.25 0.31 1 Current account includes goods, services, and all current transfers, both private and official. Private net capital flows include net errors and omissions. 2 Country means take into account the sign of imbalances. In contrast, group means are based on absolute values of country means. 3 Statistic reported is the estimated coefficient on the lagged dependent variable in the first-order autoregressive equation; * indicates statistical significance at the 95 percent level. 4 1967-72 only. Prior to 1967 balance of payments data for France included data for overseas territories.

Further, so long as the host country invests those to GNP.75 Given such a measure, it is then possible foreign savings wisely (i.e., obtains a rate of return in to use the sum of the current account and normal excess of the cost of borrowing), there is no reason capital flows as an approximate indicator of the extent why it cannot sustain a current account deficit for a of external disequilibrium—that is, the closer this sum prolonged period and, just as important, there is no is to zero, the more satisfactory external adjustment presumption that such a continuing current account is presumed to be. imbalance would be suboptimal from a global welfare Table 3 presents the relevant comparison between viewpoint.73 the periods of floating rates and the adjustable peg For these reasons, the equilibrium balance of pay- for this second crude indicator of external adjustment.76 ments is usually defined not in terms of the current Because of considerations of data availability for account alone but rather in terms of a current account capital flows, the number of smaller industrial countries position that can be financed by sustainable or in the sample falls from eight to four. Also, the use of "normal" capital flows. As is well-known, there are a four-year moving average for normal capital flows a host of serious problems in defining and measuring involves the loss of two observations from the begin- normal capital flows.74 For the purposes of the present ning of the data period and one from the end; hence, study, normal capital flows were approximated by a the period of adjustable pegs is now 1965-72 and the four-year moving average of the ratio of actual net period of floating rates is 1973-81. private capital flows (inclusive of errors and omissions) Two conclusions stand out in Table 3. The first important conclusion, is that, as in Table 2, external 73 As noted earlier, one has to be careful here to consider also whether the real rate of return on domestic assets is not artificially 75 high or low because of the presence of special incentives for, or Specifically, normal capital flows (NCF ) are defined as NCFt restrictions on, international capital flows. If, for example, the real = [CFt+1 + CFt + CFt - 1 + CFt - 2]/4.0, where CFt is the ratio of interest rate is artificially high because of an unsustainably large net private capital flows to GNP in year t . The allocation of errors budget deficit, then the structure of that country's balance of and omissions to the capital account is rather arbitrary, but does payments is not likely to be either sustainable or internationally not seem to affect the qualitative nature of the conclusions reached. optimal. More generally, a proper evaluation of the equilibrium A more serious problem with any such mechanical formula for balance of payments should consider the sustainability and optimality normal capital flows is that it cannot totally filter out short-term of the public sector's tax and expenditure structure. disequilibria. The experience of Japan in 1973/74 and again in 74 These problems include adjusting observed capital flows for 1979/80, when unusually large capital flows through the domestic large changes in fiscal positions, for structural economic changes banking system were coincidental with short-term oil deficits, serves (e.g., natural resource discoveries), and for changes in restrictions as a good case in point. on capital flows. Also, one has to decide which (if any) official 76 Because the calculations in Table 3 do not account for any capital flows to include. All these problems are discussed at some changes in desired reserves over time, the possibility cannot be length in the recent Fund staff study (IMF (1984 b)) on issues in dismissed that the figures reflect some combination of adjustment the assessment of exchange rates of industrial countries. efforts and of changes in desired reserve holdings.

31

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

adjustment for the larger industrial countries shows worse" (Dornbusch (1980, p. 173)). These periods of up as considerably better during the period of floating strong disequilibrium and of great pressure on ex- rates than during the last decade of adjustable par change rates also provide one explanation for the values. Both the average size of the imbalance and its heavy official intervention that took place then, as persistence (i.e., the serial correlation) are noticeably authorities "leaned against the wind" to dampen smaller in 1973-81 than in 1965-72. On the minus side, exchange rate movements. the average dispersion of these imbalances is larger in As regards the smaller industrial countries, there is 1973-81 than in 1965-72. Perhaps deserving of special a more pronounced trend in net private capital flows notice is the dramatic improvement in external ad- associated either with the financing of large-scale justment recorded in 1973-81 by the two countries energy projects (e.g., Norway) or with a substantial (Japan and the Federal Republic of Germany), with increase in official or quasi-official borrowing (e.g., the largest estimated average external imbalance in Denmark and Sweden).80 Note, however, that for these 1965-72.77 As before, however, it is apparent that the smaller industrial countries the negative covariance smaller industrial countries have not shared in this between the current account and private net capital improved external adjustment. When account is taken flows still holds and that, as a result, external imbalance of normal capital flows, external adjustment is still inclusive of capital flows is typically much smaller worse (in terms of average size, amplitude, and per- than without them. sistence) in the period of floating rates, although the period differences are much reduced vis-a-vis the The Equilibrium Balance of Payments current account measures shown in Table 2.78 Even though estimates of a more comprehensive The second important conclusion emanating from construct of the equilibrium balance of payments do Table 3 is that the size of external imbalances is much not extend back beyond 1975 or 1976, these estimates reduced during both periods when external adjustment are still worth examining for what they imply about is defined to include the capital account as well as the the adequacy of external payments adjustment during current account. In other words, on average, private the last seven or eight years of managed floating. Two net capital flows have acted as an offset to current such estimates or studies merit separate mention. account imbalances—although certainly not a perfect The first is that recently done by Williamson (1983). offset. This point is brought out more clearly in Although he estimates discrepancies between actual Chart 4, which shows the annual ratios to GNP of the and so-called fundamental equilibrium exchange rates current account and of net private capital flows for rather than differences between actual and equilibrium each of the subject countries over the whole 1963-83 payments imbalances, it is possible to infer the latter period. For the larger industrial countries, the typical from the former because Williamson (1983, p. 14) pattern is for both the current account and the private defines his fundamental equilibrium exchange rate as capital account to fluctuate around their long-term essentially the exchange rate that would make the trend values (of zero to roughly 1 percent of GNP), actual balance of payments equal to the equilibrium and for positive (negative) deviations in the current one. Also, the types of adjustment that are made to account to be paired with negative (positive) deviations actual current account and capital account balances in the private capital account.79 Thus, the rule still (e.g., adjustments for cyclical demand effects, per- holds that private capital flows help to finance current manent exogenous changes in the terms of trade, account imbalances rather than add to them. This does continuing intercountry differences in labor productiv- not mean that there have not been episodes when ity, and the switch in the status of the United Kingdom capital flows exacerbated current account problems. since 1977 from a major oil importer to a significant Indeed, Chart 4 shows that the last few years of oil exporter) are at least close relatives of the types of the Bretton Woods system (1969-72), as well as the adjustment suggested here. For these reasons, it is 1977-79 period, fit the characterization that "when the probably legitimate to use Williamson's estimates of current account gets bad the capital account gets real exchange rate disequilibria (i.e., differences be- tween actual and fundamental real equilibrium ex- 77 Also worth noting is the cessation of the explosive trend in the change rates) as a rough index of external payments serial correlation of the U.S. payments indicator that was evident maladjustments. in the 1966-72 period. 78 This statement is also true when the calculations in Table 2 are redone for the 1965-72 and 1973-81 periods used in Table 3. 80 The financing of energy projects is also relevant for U.K. capital 79 For the six larger industrial countries (excluding France), the flows in the 1970s. Similarly, the increase in quasi-official borrowing only case where the mean current account or mean capital account also accounts for much of the rise in Italian net capital inflows since ratio gets much above 1 percent is Canada, where both ratios stand the mid-1970s. Finally, the relaxation of capital controls is relevant at about 2 percent for the 1963-72 period. Even in this case, for explaining net capital flows in Japan, the United Kingdom, and however, the mean ratios carry opposite signs. the United States during part of the 1970s (see IMF (1984 b)).

32

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment

Chart 4. Ratio of the Current Account and Private Net Capital Flows to GNP in Selected Industrial Countries, 1963-82 1

Sources: IMF, International Financial Statistics and the data file of the External Adjustment Division of the Fund's Research Department. 1 Current account includes goods, services, and all current transfers, both private and official. Net private capital flows include net errors and omissions.

Williamson's (1983) estimates cover the five major of floating rates, this mean conceals substantial mal- currency countries for the 1976-83 period. Two of his adjustments, often in the opposite direction, during conclusions are relevant for this paper. First, there is individual one-year or two-year periods (e.g., the no indication that the size of external adjustment United Kingdom in 1976 versus 1980-83, Japan in 1978 imbalances has been declining over time during the versus 1982-83, and France in 1980 versus 1982-83). past seven years. On the contrary, for three of the five The second study, or series of studies, on equilibrium major industrial countries (each of which has adopted payments balances was done by the Fund staff over an exchange rate arrangement of independently float- the 1971-83 period. A brief history of the development ing), the maladjustments in the 1981-83 period have of that methodology in the Fund (usually referred to been as large or larger than at any other time during as the underlying payments balance method) is con- the past seven years. Second, whatever the mean tained in a recent Fund staff study (IMF (1984 b)) degree of external payments adjustment for the period dealing with issues in the assessment of exchange rates

33

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM of industrial countries.81 While the approach taken by or changes in real exchange rates, the outlook for the staff to estimate equilibrium payments positions effective external payments adjustment was not en- in those exercises falls within the general framework couraging from the vantage point of November 1983. outlined here, there is one important difference. Whereas To summarize, the record of external payments the earlier calculations in this paper have been static adjustment during the period of floating rates has been or backward looking, the exercises in the Fund are a mixed one. From the available summary indicators, explicitly forward looking. Specifically, they seek to there is little evidence that external adjustment has estimate the likely paths of the current account (ad- been less complete or slower under floating rates than justed for temporary factors) and of normal capital under the adjustable peg system. In fact, the perform- flows during the next two or three years, given the ance of the larger industrial countries appears on continuation of present exchange rates, anticipated average to have been better in the period of floating macroeconomic policies in the subject countries, the rates while that of the smaller industrial countries delayed effects of past exchange rate changes, and a looks worse. At the same time, any notion that floating number of other expected future developments.82 In rates produce prompt equilibration of external pay- short, the aim of the staff exercises has really been to ments balances can be safely discarded. When effective spell out the medium-term balance of payments im- external adjustment does take place, it takes a number plications of past and present exchange rates and of of years and there have been many instances of large expected future macroeconomic policies. If those bal- discrepancies between actual and equilibrium pay- ance of payments developments are deemed unsus- ments balances at prevailing exchange rates. tainable or undesirable in the sense that "underlying" current accounts are quite different from normal capital The Process of External Adjustment flows, then the normative implication is that either planned policies or present exchange rates need to Another route to assessing the adequacy of external change to prevent those undesirable balance of pay- payments adjustment under floating rates is to eschew ments scenarios from taking place. the use of summary indicators of external adjustment Given the focus of this paper, perhaps the key in favor of an analysis of how a greater degree of exchange rate flexibility affects the process of external conclusion arising from these staff exercises is that 84 the period of floating rates has been marked by many adjustment itself. Once again, the answers that emerge instances in which anticipated policies, together with are more indirect and subjective, but they are better prevailing exchange rates, did imply undesirable or able to isolate the independent contribution attributable unsustainable external payments outcomes. Table 4, to the present exchange rate system. The role of the taken from a recent Fund staff study (IMF (1984 b)), exchange rate system is considered here in relation to provides just one representative example of such a the following four external adjustment issues: staff calculation applied to the November 1983 situa- (1) relative price movements versus income move- tion. The key point to note in that table is that the ments as adjustment mechanisms; (2) the implications estimates of normal capital flows given in column (11) of differential adjustment speeds in goods versus asset are quite different from the estimates of underlying markets; (3) the symmetry of adjustment across dif- current accounts given in columns (9) and (10).83 In ferent types of countries; and (4) rules versus discretion short, without changes in macroeconomic policies and/ in adjustment.

81 In Fund terminology, "underlying payments balances" refer to Relative Price Changes Versus Relative actual payments balances adjusted for temporary factors, earlier Income Movements85 exchange rate changes, desired reserve changes, and anticipated future developments (including anticipated future policies). The One of the cornerstones of the case for floating rates equilibrium payments balance can then be defined as a situation where the underlying current account is equal to normal capital is that greater exchange rate flexibility permits the flows. substitution of expenditure-switching policies for ex- 82 Two advantages of taking expected future policies into account penditure-reducing ones, with the result that the cost are that it eliminates the need to estimate a normal cyclical position and that it brings added realism to the exercise (in the sense that of external adjustment (in terms of output and em- private market participants clearly take expected future develop- ployment) is reduced, especially in a context of down- ments into account in assessing exchange rates). 83 In interpreting Table 4, it should be kept in mind that the estimates of normal capital flows, while not based on mechanical formulas, are still subject to rather wide margins of error—especially 84 Yet another approach to assessing the adequacy of external in the recent situation when there have been large changes in fiscal adjustment is to examine the incentives for adjustment; on this positions and/or significant structural changes in capital markets score, see Kafka (1976). (e.g., liberalization of capital flows) in some of the largest industrial 85 Most of the arguments in this subsection draw heavily on countries. Goldstein (1980).

34

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment

Table 4. Underlying Payments Balances on Current Account and Normal Capital Flows 1 (In billions of U.S. dollars) Adjustment for: Scenario I: Scenario II: Underlying Underlying Current Current Current Recent Changes in Medium- Scenario I: Scenario II: Balance Balance Balance Temporary relative relative term Underlying Underlying Including Including Normal Projection distur- price cyclical tenden- Current Current Official Official Official Capital 9 10 1983 bances 2 changes 3 positions 4 cies 5 Balance 6 Balance 7 Transfers 8 Transfers Transfers Flows (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) United States -34.2 0.2 -28.2 -10.2 -7.8 -80.3 -85.5 -5.6 -85.9 -91.1 -5 to -10 Germany, Fed. Rep. 9.0 — 10.9 0.8 -3.1 17.9 20.0 -6.5 11.4 13.5 0 Japan 22.3 -0.6 0.8 -1.6 6.4 27.3 30.0 -2.0 25.3 28.0 -5 to -10 Other industrial countries 1.8 0.1 16.5 -0.2 3.1 21.0 20.7 -8.0 13.0 12.7 +10 to -20 Total industrial countries -1.2 -0.3 — -11.2 -1.4 -14.1 -14.8 -22.1 -36.2 -36.9 — Other countries -72.6 0.3 — 11.2 1.4 -59.7 -59.0 22.1 -37.6 -36.9 — Total11 -73.8 — — — — -73.8 -73.8 — -73.8 -73.8 — Source: IMF, Issues in the Assessment of Exchange Rates of Industrial Countries, Occasional Paper No. 29 (July 1984). 1 All accounts are in constant 1983 prices. 2 In Japan, recent import liberalization measures increased import volumes; adjustments are required to give appropriate balances for the whole of 1983, both for Japan and for its trading partners. 3 Estimates of the effects of changes in exchange rates and domestic price levels that occurred from 1980 through November 1983 but are not reflected in 1983 trade flows. 4 The effect of assumed changes in relative cyclical positions to 1986 on trade and services balances in terms of 1983 dollars. 5 Effects over three years of estimated trend factors, including changes in fuels balances. 6 The 1983 balances adjusted by the factors in columns (2) through (5). 7 A further adjustment is made in column (7) for changes in relative cyclical positions which arise if prospective U.S. demand growth is assumed to be 1/2 percent per annum higher, and prospective demand growth in the Federal Republic of Germany and Japan is assumed to be 1/2 percent lower than in the central scenario. 8 Includes effects over three years of estimated trend in official transfers. 9 The sum of columns (6) and (8). 10 The sum of columns (7) and (8). 11 Reflects errors, omissions, and asymmetries in reported statistics, plus balance with other countries including the U.S.S.R. and other countries of Eastern Europe that are not Fund members. In the table, this asymmetry is assumed to remain constant at the projected 1983 level. ward price inflexibility (Friedman (1953), Meade (1955), . . . consider the probable effect of flexible rates on and Johnson (1973)). Implicit in this conclusion is the foreign trade elasticities. It is convenient to introduce the assumption that, once a country's effective exchange distinction between the actual exchange rate and what, rate moves in the direction required for, say, current in analogy to permanent income, may be called the account adjustment, the resulting relative price effects permanent exchange rate. . . . will act as a powerful aid in securing that adjustment. What matters for trade flows in physical units is mostly While the foregoing argument is unassailable in a the permanent rate. Major changes in the international theoretical sense, critics of floating rates have ex- division of labor require new production facilities, new pressed doubts about the practical efficacy of exchange distribution networks, new sources of supply, and the rate changes as an adjustment mechanism. Those development of new markets. Most firms will try to avoid doubts have two origins—one conjectural and the other making such long-term decisions on the basis of exchange rates which turn out to be only temporary. . . . With historical. The conjectural one is that the uncertainty flexible rates, in view of the slow adjustment of permanent associated with floatingrate s could significantly reduce rates to actual rates, this process will be even slower, the size of price elasticities for traded goods. The and many fluctuations in actual rates will have hardly any historical one is based on the observation that there effect on permanent rates, and thus on trade flows. have been several occasions during the period of floating rates when sizable current account imbalances and by McKinnon (1978, p. 4): have persisted in the face of sizable exchange rate . . . with the advent of floating, the future direction of changes (in the right direction). exchange rate movements has proved highly uncer- The price-elasticity-pessimism thesis has perhaps tain. . . . And it may not be in the interest of merchants been put foward most clearly by Niehans (1975, to engage in active arbitrage in industrial commodities if p. 276). tomorrow's exchange rate is unknown. Hence, the quan-

35

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

titites of goods traded respond sluggishly to exchange perverse episodes for the major industrial countries: rate fluctuations giving rise to a modern version of namely, the 1976-78 period. In particular, worthy of elasticity pessimism. . . . note is the seemingly perverse association between Whatever the plausibility of the Niehans-McKinnon nominal exchange rate changes (row 7) and current argument, there is as yet no firm empirical support for account developments (rows 4-6) in the three largest it. A recent survey by Goldstein and Khan (1982), for industrial countries (the Federal Republic of Germany, example, does not find any consistent tendency for Japan, and the United States). estimated price elasticities to be lower in equations There are at least three reasons why this evidence estimated with 1970s data than in those based on data should be interpreted cautiously, however. from earlier periods. It needs to be acknowledged, (1) Exchange rates are only one component of a however, that the number of studies using data for the country's competitive position. The other component period of floating rates is still relatively small, and that is the behavior of the prices or costs of traded goods composition effects (i.e., changes in the weights of in that country relative to those of its competitors— low and high elasticity goods in total exports or that is, real rather than nominal exchange rate changes imports) can distort the independent effect of the count for trade flows. Divergences from purchasing exchange rate regime on these elasticities. power parity have been much larger during the period Turning to the historical argument, Table 5 docu- of floating rates than under the adjustable peg (Genberg ments perhaps the most striking of these alleged (1978)), but there have nevertheless been episodes

Table 5. Trade Balances, Current Accounts, Exchange Rates, Relative Prices, and Cyclical Income Movements for the Seven Largest Industrial Countries, 1976-78 Fed. Rep. of United United Canada France Germany Italy Japan Kingdom States

Trade balance 1 1976 1.8 -4.9 16.2 -4.2 9.8 -7.0 -9.5 1977 3.1 -3.3 19.6 -0.1 17.3 -3.9 -31.1 1978 4.1 0.1 24.7 2.9 25.6 -3.0 -34.0 Current account 1,2 1976 -4.2 -3.4 3.9 -2.9 3.7 -1.5 4.2 1977 -4.1 -0.4 4.1 2.4 10.9 0.1 -14.5 1978 -4.3 7.1 9.2 6.2 17.5 2.2 -15.5 Percentage change, 1976-78 Nominal effective exchange rate 3 -16.7 -7.8 13.7 -16.2 32.0 -6.0 -10.4 Unadjusted relative wholesale prices 4 3.8 4.6 -9.4 18.7 -12.5 18.5 1.5 Adjusted relative wholesale prices (real exchange rate) 5 -13.6 -3.5 3.1 -0.4 15.3 11.7 -9.0 Percentage change in real GNP/GDP 6 1975 1.2 0.2 -1.6 -3.6 2.4 -1.1 -1.2 1976 5.5 5.2 5.6 5.9 5.3 3.4 5.4 1977 2.1 3.0 2.8 1.9 5.3 1.6 5.5 1978 3.6 3.7 3.5 2.7 5.1 3.9 5.0 Output gap in manufacturing 7 1975 -8.5 -6.9 -11.1 -13.1 -22.3 -9.4 -12.3 1976 -6.2 -3.9 -5.3 -5.9 -17.2 -7.9 -6.4 1977 -7.2 -3.0 -4.2 -7.0 -15.4 -6.5 -2.7 1978 -5.0 -2.6 -4.1 -8.1 -13.1 -6.1 0.0 Sources: IMF, World Economic Outlook (various annual issues); IMF, International Financial Statistics, various issues; and Artus and Turner (1978). 1 In billions of U.S. dollars. 2 Includes goods, services, and all current transfers, both private and official. 3 The Fund's multilateral exchange rate model (MERM) index. A positive figure denotes an appreciation, a negative one a depreciation. 4 Percentage change in ratio of own to competitors' wholesale prices for manufactures. A positive figure denotes deterioration in country's position, a negative one improvement. 5 Percentage change in ratio of own to competitors' wholesale prices for manufactures, adjusted for effective exchange rates. A positive figure denotes deterioration in country's position, whereas a negative one denotes improvement. Since the relationship is multiplicative rather than additive, numbers in the two rows above will not sum to those in this row. 6 Gross domestic product at market prices for France, Italy, and the United Kingdom. 7 Defined as potential output less actual output, as percentage of actual output. A negative figure indicates manufacturing sector is operating at less than normal capacity.

36

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment when relative inflation rates have offset much of the spondence between the latter and current accounts. effect of nominal exchange rate changes on countries' By all reports, such relative cyclical developments competitive price positions. The 1976-78 period was were responsible for much of both the deterioration in one of these. As shown in rows 8 and 9 of Table 5, the U.S. current account and the improvements in the this relative inflation offset was most pronounced for current accounts of the Federal Republic of Germany the countries with the most abnormal inflation per- and Japan between 1975 and 1978 that are reflected in formance—that is, Italy and the United Kingdom on Table 5 (Lawrence (1978) and Wallich (1978)). This is the high-inflation side, and Japan and the Federal quite a common phenomenon; whenever a country's Republic of Germany on the low-inflation side. In growth rate (relative to its potential) exceeds that cases where the size of real exchange rate changes is abroad, its current account usually deteriorates. modest, a large relative price impact on current account In sum, the contribution made by exchange rate outcomes should not be expected. changes to current account adjustment is at times (2) A second factor that works against a clear hidden from the naked eye, especially over the short association between exchange rate changes and current run, but this contribution is still important. accounts is that relative price changes affect the volumes of imports and exports in the short run Adjustment in Goods Versus Asset Markets differently from in the long run. By now, there is a considerable empirical literature on relative price elas- In this paper equilibrium payments balances, and ticities in international trade (Goldstein and Khan by analogy the equilibrium exchange rate, have been (1982)). This literature suggests that those elasticities defined in terms of a current account equal to normal are significant and reasonably large over the long run capital flows (after correction for temporary factors, (two to three years) in most industrial countries but cyclical effects, etc.). Thus, both normal flows of also that they are much smaller over the short run (up goods and normal transactions in assets are included to one year). A consensus estimate would be that in the concept of "fundamentals." Other observers, short-run price elasticities are only about half as large however, often choose a different definition of "fun- as the long-run ones. Given that the long-run elasticities damentals," ranging from the current account alone, themselves probably lie in the range of -0.5 to -1.0 to all transactions in financial assets, to the behavior for total imports and - 1.25 to - 2.5 for total exports of the major determinants of such trade flows or in a representative industrial country (Goldstein and changes in asset supplies or demand (e.g., real incomes, Khan (1982)), this means that the short-run response relative inflation rates, money supplies, interest rates, of the trade balance to an exchange rate change can etc.). It is thus easy to understand why the adequacy be perverse. For example, Spitaller (1980) shows that of external adjustment or exchange rate behavior under the short-run deterioration or improvement in the trade floating rates—measured as the difference between balance following a depreciation or appreciation, re- actual outturns and that corresponding to the funda- spectively, will probably last about four or five quarters mentals—elicits so many divergent views. and that, at its worst point, it could amount to about Regardless of the precise definition of "fundamen- 8 to 10 percent of the local currency value of imports. tals," however, there is the indisputable basic eco- (3) The third, and probably most compelling, reason nomic reality that adjustments to disturbances in goods why exchange rate changes and current account out- and labor markets typically take much longer than comes do not always move together is that relative those in financial markets. This means that prices in price changes are not the only, or even the most those financial markets (namely, interest rates and important, determinant of current account movements exchange rates) can be expected to bear the brunt of over the short to medium run. Over a one-year period the short-run adjustment to unanticipated economic the combined income elasticities of demand for imports developments. And because these financialprice s have and exports will generally be two to four times larger to compensate for the stickiness of goods and labor than the sum of relative price elasticities (Deppler and prices, their short-run response is often much larger Ripley (1978); Hooper (1978); Goldstein and Khan than the long-run one (after other prices have also (1982)), and there are some indications that income moved). In now popular terminology, these financial movements would still be more powerful even after prices "overshoot."87 three years.86 Thus, if relative income movements This overshooting of exchange rates and other fi- work on trade flows in a direction opposite to that of nancial prices would not perhaps be cause for much relative price factors, there need be no close corre- concern if the overshooting did not last long, or if the

86 See the staff exercise reported in the Fund's Annual Report, 87 For the classic theoretical presentation of overshooting, see 1978 , p. 42. Dornbusch (1976).

37

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

disturbances that initiated it were solely of domestic sequence of shifting more of the adjustment to those origin, or if we could always identify the nature of markets is that real output and employment may suffer those disturbances, or if the induced exchange rate more. The second reason is that pegged rates can give changes did not have a powerful effect on the real side rise to their own brand of overshooting. Machlup of the economy. Unfortunately, experience with float- (1979, p. 76) makes this point as follows: ing rates suggests that none of these conditions is true. The economists who blame particular appreciations and As noted earlier, it would seem that there have been depreciations of currencies on "overshooting in a regime periods of two to three years' duration under floating of floating rates" forget that the discrete and deliberate rates when the structure of payments balances and of adjustments of fixed or pegged exchange rates, especially exchange rates was out of line with fundamentals and the official devaluations, usually involve much wider when this overshooting was most closely connected overshooting. If a currency had long been overvalued by with developments in financial markets. The interna- its fixed parity, and the authorities at last decided on a tional aspect of the problem arises because financial devaluation, they regularly chose a new parity which markets today are so well diversified and integrated undervalued the currency at current levels of prices. that assets denominated in different currencies are Among the justifications for such "excessive" devalua- close substitutes for one another. Hence, anything that tions was usually the argument that the monetary reserves, affects the current rate of return, risk factors, or the depleted in the period of overvaluation, had to be replen- ished in the subsequent period of undervaluation. No one expected future rate of return in a partner country can was ever ashamed of such overshooting in the official have swift and strong repercussions on the home adjustment of the official par value; yet the same observers country's asset prices, including its exchange rate. blame the system of managed floating for allowing the Further, while in some cases it is easy to locate the market to overshoot the "right" exchange rate. What source of the overshooting disturbance in terms of used to be the rule for devaluations is now regarded as large shifts in the stance of monetary or fiscal policy, disorderly in depreciations. in other cases it is not, with exchange rates continuing to move out of line after real interest rate differentials have stabilized or after the current account has begun Symmetry of Adjustment to move in the opposite direction. Last, it has to be As is well-known, the period under the Bretton recalled that although the exchange rate is often viewed Woods system was marked by frequently and strongly as an asset price (i.e., the relative price of two monies), expressed concerns on the part of some countries over it is also a key component of the relative price of the alleged lack of symmetry in external adjustment. national outputs; for this reason, it will have strong Two types of asymmetry were most discussed.88 First, effects on the demand and supply for traded goods, there was the charge that surplus countries were and hence on employment in traded goods industries. subject to a much weaker discipline than deficit coun- This fact is often brought home most vividly after the tries. Second, there were complaints about the special implementation of a program of monetary restraint. In role of reserve centers, especially the United States. the initial stages, the effects of the resultant increase To the Europeans, this special role constituted an in real interest rates on economic activity outweigh unwarranted privilege because the United States alone the relative price effects of exchange rate appreciation could finance payments deficits by liability as opposed and produce an improvement in the current account. to asset settlement.89 On the other hand, to the United Later on, however, while the capital account continues States the special role of the dollar as numeraire of to be aided by the high real interest rate, the current the system came to be seen as a burden because it account deteriorates as the adverse relative price precluded the initiation of exchange rate action as an effects of a higher real exchange rate take their toll on adjustment mechanism. What can be said about the the country's trade flows. validity of these allegations and about their subsequent Having noted the seriousness of the overshooting existence under managed floating? problem, it is equally relevant to point out that the The proposition that balance of payments deficits solution may not lie in pegging of exchange rates for prompted stronger adjustment measures than compa- at least two reasons. One is that such an action may rable surpluses under the adjustable peg system is just transfer disturbances from the exchange market supported in what is probably the most thorough study to goods and labor markets at even greater social cost of the issue. Specifically, after studying balance of (Frenkel and Mussa (1980)). In this respect, at least one advantage of the exchange market is that insurance against unforeseen (short-term) fluctuations can be 88 For a summary of these discussions, see Cumby (1982). 89 The role of the dollar as the primary reserve asset also gave purchased through forward contracts. And because rise to the debate about whether the U.S. payments deficit in the prices are sticky in goods and labor markets, a con- 1960s was "supply determined" or "demand determined."

38

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment payments developments in nine industrial countries To summarize, the degree of external payments over the 1950-66 period, Michaely (1971, pp. 63-64) adjustment under managed floating may not be as swift concludes: or as predictable as would be desirable, but it is hard Countries whose monetary policy generally responds to to attribute any shortcomings to systematic asymme- changes in the balance of payments tend to make excep- tries across classes or types of countries. tions to this pattern of behavior mainly when they are in surplus. Similarly, compliance of monetary policy with Rules Versus Discretion in Adjustment balance-of-payments requirements in generally noncom- plying countries tends to be found at times of deficits. . . . Yet another frequently heard criticism of the present The loss of reserves is viewed with concern; but their exchange rate system is that it relies too much on accumulation . . .is viewed, in fact, with satisfaction or discretionary policy actions by country authorities to indifference. secure external adjustment. The implication is that a more automatic system that relied on more specific The 1970 report by the Executive Directors on the rules would yield a better result. Further, there is a role of exchange rates in the adjustment of international related criticism that both external adjustment and payments (IMF (1970, p. 38)), similarly suggests that exchange rate stability would be enhanced if there was adjustment pressures on deficit countries were stronger a more formal and more explicit mechanism for co- than those on surplus countries during the 1950s and ordination of economic policies across countries. What 1960s because ". . . reserve accumulation is not sub- can be said about these two criticisms? ject to limit in the same way as exhaustion of reserves To begin with, it is sensible to acknowledge that if or of borrowing facilities."Likewise, the figures on one were to align alternative exchange rate systems mean external imbalances for the larger industrial along a spectrum according to either the degree of countries, shown in Tables 2 and 3, also point in the automatism of the adjustment process or the mix same direction. If the United States is excluded be- betweeen rules and discretion in initiating adjustment, cause of its special position, the countries with the the results would suggest that the present system is largest mean imbalances and with the most persistence closer to the complete discretion pole than to the rules- in those imbalances under the Bretton Woods system only pole. In this sense, the pure gold standard with seemed to be the surplus countries. its automatic specie flow mechanism, the adjustable The fact that payments imbalances and real exchange peg system with its clear implications for the subor- movements have both been more variable under float- dination of domestic monetary policy to the exchange ing rates than before, and the probability that nominal rate except during fundamental disequilibrium, the exchange rate appreciations are harder to sterilize objective indicator system (based, say, on reserve under floating rates than reserve increases were under levels) with its automatic trigger for the initiation of the adjustable peg, have seemingly combined to reduce adjustment actions, mechanistic crawling peg schemes such surplus/deficit asymmetries. Again, Tables 2 and with their automatic adjustment of the exchange rate, 3 suggest that there has been no obvious pattern to or even a pure floating system with its complete mean payments imbalances along surplus/deficit lines. prohibition of all official intervention in the exchange In the case of asymmetries associated with the market—all could be considered less discretionary special role of the United States, the extent of asym- than the present system. In much the same way, it metry appears to have been reduced during the period can readily be agreed that efforts at coordination of of floating rates. The diversification of official reserve economic policies during the period of floating rates holdings away from the dollar toward other currencies represent at most a middle ground along a hypothetical (principally the deutsche mark and the yen) and, most spectrum between completely activist and completely important, the spread of liability settlement of external passive coordination strategies.91 In this context, ef- imbalances to even non-reserve-currency countries, 90 forts have gone beyond the exchange of forecasts and have reduced the special privileges of the dollar. But policy intentions to encompass occasional common by the same token, any special burdens would also actions (e.g., the U.S. dollar support package of seem to have dissipated with the abandonment of November 1, 1978 or the setting of explicit targets for dollar convertibility into gold in 1971 and with the energy conservation efforts within the International sharp increase in the variability of the real effective Energy Agency); agreements on short-term exchange exchange rate of the dollar over the 1973-83 period. rate management policies (e.g., intermittent joint

90 It is perhaps ironic that the greater symmetry of adjustment sought by the Committee of Twenty seems to have been aided not by imposing asset settlement on reserve centers but rather by 91 Following Polak (1981), coordination may be thought of here extending the unwarranted privilege of liability settlement to many as including all international influences on domestic economic non-reserve-center countries. decision making.

39

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM countering of disorderly market conditions); sometime market participants have undoubtedly had difficulty in agreement on medium-term expenditure and/or energy finding a good anchor for longer-term exchange rate policies (e.g., the final agreements of the Bonn eco- expectations. The harder it is to make an informed nomic summit of 1978); and most recently, the exten- judgment about the future course of policies, the more sion of Fund surveillance to a multilateral Group of one can expect the erroneous extrapolation of short- Five format.92 At the same time, coordination efforts term events. By the same token, a country that has have stopped well short of binding agreements on established a credible long-term policy posture can either exchange rate targets or rates of monetary count on the market to be more forgiving of short- expansion. For want of a better label, the present term deviations because such deviations do not ma- system might therefore be characterized as a discre- terially affect the long-term anchor. All of this is tionary and decentralized system, with loose coordi- consistent with the conclusion of the Versailles eco- nation among the main players but with tighter coor- nomic summit meeting of 1982 that the maintenance dination and disaster relief during crises. of the internal and external values of currencies "rests Although the issues of rules versus discretion and primarily on convergence of policies designed to achieve of the optimal coordination of policies are still ones of lower inflation, higher employment, and renewed eco- little agreement, past experience would seem to point nomic growth."93 to the following four conclusions. (2) A second conclusion is that, whatever the mix (1) Whatever the combination of rules and discre- between rules and discretion, effective external ad- tion, a prerequisite for successful external adjustment justment will not be forthcoming unless countries take is the pursuit of stable, credible, and balanced mac- into account the external repercussions of their own roeconomic policies at the national level. Without such macroeconomic policy actions on their trading part- policy behavior, the greater autonomy of policy in- ners.94 The notion that, under floating rates, each struments in more discretion-based systems will not country can decide independently its own policy stance produce greater policy effectiveness, and similarly and mix and let the exchange rate settle all conflicts without it, the policy rules in more rules-based or in the market place is neither realistic nor helpful. It automatic systems will not be observed. This is un- is not a realistic notion because, as argued earlier, doubtedly what led Frenkel (1983 b, p. 112) to con- floating rates are not capable of providing enough clude: insulation from other countries' policy actions to make independent targeting work. It is also not helpful If governments were willing to follow policies that are consistent with the maintenance of a gold standard, then because failure to take into account other countries' the gold standard itself would not be necessary; if, policies is likely over time to induce retaliatory actions however, governments are not willing to follow such by those who are unhappy with the verdict of the policies, then the introduction of the gold standard per market. In the end, therefore, the path to external se will not restore stability, as before long the standard adjustment will be slower and the eventual equilibrium will have to be abandoned. less satisfactory than they would be if some coordi- nation of policies took place. That is, the need for good policy is not diminished by the presence or absence of automatic adjustment But taking external repercussions "into account" in rules. This lesson is especially pertinent to countries setting domestic policies is not the same thing as being that have adopted independently floating exchange dominated by external considerations. It is unlikely, arrangements. Two clear implications of the asset at least for the larger industrial countries, that a fully market view of exchange rates are: (1) that the current centralized and coordinated decision-making system exchange rate will be much influenced by the expected would be either realistic or helpful. It would not be future exchange rate; and (2) that the expected future realistic because, even with the recent narrowing of rate will be much influenced by expected future mac- inflation differentials among the largest industrial coun- roeconomic policies. Since instability in present poli- tries, it is unlikely that these countries would be willing cies generates uncertainty about future policies, it is to fully subordinate domestic policy, particularly mon- easy to see why stable and credible policies are a sine qua non for greater stability in exchange rates. Indeed, 93 See IMF Survey (IMF (1982 b)). The recent report by a one of the main reasons why the adjustment process Commonwealth Study Group on challenges for the world financial and trading system (i.e., Helleiner (1983, p. 31)) also places strong has not worked better under floating rates is because emphasis on better coordination of national macroeconomic policies among the major industrial countries for promoting smooth and 92 See Polak (1981), Dreyer (1982), and Solomon (1982) for accounts equitable adjustment to international payments imbalances. of coordination measures taken during the period of floating rates 94 Although international coordination of policies is often associ- and before. Also see "Evolution of Surveillance Function of the ated with coordination of monetary policies, developments of the Fund Reflects Changing Economic Situation of Members," IMF past few years stand as testimony to the international externalities Survey, Vol. 13 (April 19, 1984), pp. 98-100. of unstable and uncoordinated fiscal policies.

40

©International Monetary Fund. Not for Redistribution Floating Rates and External Adjustment etary policy, to the goal of a fixed exchange rate. It systems usually turn out to be less automatic in practice would not be helpful because such an assignment rule than in theory and that very specific adjustment or would leave internal balance to be handled by market exchange rate rules run the risk of becoming liabilities forces, or fiscal policy, or incomes policies. It seems when the global environment changes in unexpected wiser to continue to focus monetary policy on the still ways. The first point about automatic adjustment is formidable task of achieving long-run price stability best illustrated by the historical gold standard. Its and sustainable growth. specie flow mechanism is often held up as the ideal in What all this suggests for external adjustment during automatic adjustment. But serious students of the the period of floating rates is that such adjustment historical gold standard seem to agree that internal would have been smoother and exchange rate behavior balance considerations sometimes caused country au- would have been less variable, if there had been a thorities to offset or sterilize the effects of gold flows. more intensive and more regular effort to appraise the Cooper (1982, p. 28), for example, notes that: "off- consistency of countries' intended economic policies setting actions by central banks, in periods of con- at the then prevailing exchange rates and, if, when traction as well as periods of expansion, even took such inconsistencies arose, there had been more com- place often in the heyday of the historical gold stand- promise at the margin by all parties to reduce them. ard." Similarly, in describing the 1914-21 period in The difficulties of achieving such improved coordi- the United States, the report of the Gold Commission nation should not, however, be underestimated. In this (U.S. Congress (1982, p. 67)) states that "the postwar respect, Polak (1981) has pointed out the natural limits increase in the quantity of money occurred because to coordination and even the benefits of decentralized the Federal Reserve System did not observe the rules decision making (via the market) in those cases when of the gold standard but exercised discretion. The negotiation is unsuccessful. He notes that coordination subsequent collapse occurred because the power to can be difficult because, inter alia: (1) exchange rates manage money was not limited by the requirement to (and often interest rates) are by their nature competitive maintain gold reserve requirements." in the sense that one country's gain is frequently the The perils of very specific adjustment rules can be other's loss; (2) the compromise of growth and inflation seen by reference to a number of past systems or objectives at the national level often leaves little room proposals. Perhaps the best example is the Bretton for further compromise on demand policies at the Woods system. As noted by Polak (1981), the Bretton international level; and (3) international decision mak- Woods rules of the game were written so as to place ing on floating rates is inherently more difficult than all exchange rate changes under strict international for infrequent par value changes. In a similar vein, supervision. By so doing, the danger of renewed Solomon (1982), commenting on the problems of se- competitive exchange alterations was minimized. This curing monetary coordination in past economic sum- was just what was suitable in the early years of Bretton mits, mentions that: (1) the perception of independent Woods; but during the late 1960s and early 1970s, these monetary policy may be necessary in some countries same rules became a liability when the need arose for for sustaining confidence that monetary policy will not a greater degree of exchange flexibility. Another ex- be inflationary in the long run; (2) the approach to ample is that in the 1960s the idea of a "crawling formulating and conducting monetary policy differs peg," based on inflation differentials, seemed like the greatly across countries; and (3) there is a logically right antidote for sticky exchange rates and for inef- prior need to coordinate domestic monetary policy ficiencies in central management of exchange rates. with domestic fiscal policy. But after the plethora of real economic disturbances But all these barriers to greater coordination of in the 1970s, that proposal's neglect of the need for policies do not mean that efforts should not be made real exchange rate adjustment now seems like a serious to improve the situation. Volcker (1977, p. 36) lends drawback. To some extent, even the objective indicator support to this position: proposal (based on reserves) considered by the Com- Questions of the policy mix and timing are not rigid and mittee of Twenty appears outmoded in a world in preordained in any of our countries. I would like to think which reserves are determined more by demand than the time has come to be a little more open in putting by supply. In short, the moral would seem to be that those kinds of issues on this and other tables as part of limitations on the ability to forecast the future global the processes of comprehension and consultation. It economic environment put constraints on the useful- seems to me a necessary part of running the floating rate ness of very specific adjustment rules—even if coun- system as effectively as I think it can be—and should tries could be convinced to obey those rules strictly. be—run. In summation, although the present exchange rate (3) The third conclusion regarding rules versus system relies more on discretion in initiating adjust- discretion in adjustment is that automatic adjustment ment measures than do some alternative systems, there

41

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

is little to suggest that performance would have been The basic thesis is simple enough and has not better over the past ten years if adjustment rules had changed much in 40 years.95 Simply stated, it is that been used instead. A more promising avenue would most exchange rate fluctuations under floating rates be to try to achieve a better anchor for longer-term will be unexpected and that such an increase in exchange rate expectations by inducing countries both exchange rate uncertainty to risk-averse suppliers and to pursue more stable and credible macroeconomic purchasers will, ceteris paribus, reduce the volume of policies and, in framing these policies, to take more trade and perhaps also increase the price of traded explicit account of their repercussions on other coun- goods.96 But this direct effect is by no means the only tries. avenue or channel by which exchange rate fluctuations can affect trade and investment. Other indirect, but nevertheless important, effects can take place as ex- Floating Rates and World Trade and change rate fluctuations induce producers and traders Investment to alter the structure of their output (traded versus nontraded goods) or their investment (diversification In 1963-72 the average annual increase in the volume versus less diversification) in order to reduce their of world trade was 81/2 percent. During the decade of overall risk.97 In addition, exchange rate fluctuations floating rates (1973-82), world trade growth slowed to could induce a variety of adjustment costs if producers less than half that figure (4.1 percent) and it was and workers do shift resources between economic absolutely stagnant during 1980-82. Also, the steady activities in response to what later turn out to be postwar momentum toward trade liberalization, marked temporary relative price signals.98 Finally, if transitory by successive rounds of multilateral trade negotiations or excessive movements in exchange rates are per- (the Dillon Round in 1960-61, the Kennedy Round in ceived as being a primary cause of the decline of 1964-67, and the Tokyo Round in 1973-79) and by the certain export- or import-competing industries, they formation of free trade areas in Europe (the European can generate pressures for protectionism that, if suc- Economic Community and the European Free Trade cessful, often prove difficult to eliminate, even if Association), had come to a full stop by the end of exchange rates later move in the opposite direction. 1979. Identifying channels by which exchange rate uncer- Indeed, despite repeated commitments at successive tainty might affect the volume, prices, or structure of economic summit meetings to the goal of maintaining world trade and investment is not, however, the same a free and open trading system, the last three years as finding good evidence in support of these effects. have witnessed an increase in both pressures for An evaluation of issues relating to the implications of protectionism and in restrictive trade measures them- exchange rate uncertainty for world trade—including selves: an assessment of existing empirical evidence and the

Protectionist pressures and protective actions that were 95 taken or intensified since the conclusion of the Tokyo For example, in discussing why freely fluctuating exchange rates would be undesirable, Nurkse (1945, pp. 5-6), states: "For Round affect sectors accounting for more than one fifth one thing, they create considerable exchange risks, which tend to of world trade in manufactures, including iron and steel, discourage international trade. "Likewise, in assessing freely floating automobiles, textiles, and clothing. In addition, industrial exchange rates, the report by the Fund's Executive Directors on countries apply restrictions at the border or use other the role of exchange rates in the adjustment of international payments (IMF (1970, p. 42)) argues: "The fluctuations in exchange rates that measures that affect or distort trade in temperate zone occurred, and the absence of any limits on the scope for potential agricultural products accounting for one third of inter- fluctuations, would involve damaging uncertainties for international national trade in agriculture, including sugar. (Anjaria and trade." others (1982, p.2)) 96 The volume effect is unambiguous because the backward shifts induced in the supply curve and the demand curve each reduce the However, only three years after the conclusion of the volume of trade. However, the price effects of increased uncertainty Tokyo Round, the pressures for protectionism are high could in principle go in either direction. See, for example, Hooper and Kohlhagen (1978). and, in the opinion of some, higher than at any other 97 For example, if relative price uncertainty is a major factor in time in the postwar period. (Bergsten and Cline (1983, decisions on resource allocation, increased exchange rate uncer- p. 2)) tainty could result in a gradual shift away from traded goods industries (which are more exposed to such uncertainty stemming Against this disappointing recent background of from exchange rate fluctuations) and toward nontraded goods in the trade and protectionism, and given the aforementioned service and other sectors. A multinational firm, in deciding where to locate new investment, needs to be aware not only of technical increase in the variability of both nominal and real factors affecting cost but also of uncertainties of currency relation- exchange rates during the period of floating rates, it ships. This can lead to a diversification of investment, even at some is not surprising that there has been renewed concern cost in terms of efficiency, in order to minimize the risks arising from currency instability. about the possible adverse effects of exchange rate 98 See the discussion above under Floating Rates and Unemploy- variablity on world trade and investment. ment.

42

©International Monetary Fund. Not for Redistribution Floating Rates and World Trade and Investment presentation of some new evidence—is contained in a well be that the measures of variability used are inadequate recent Fund staff study on exchange rate volatility and measures of uncertainty; that other factors overwhelm world trade (IMF (1984 a)). the impact of variability in the estimating equations; or The findings of this Fund staff study are summarized that the presence of statistical problems . . . interferes below. with the effectiveness of statistical tests. It may also be (1) The primary determinant of the growth in the that the lags with which greater variability in the exchange rate regime affect trade flows are longer and more variable volume of world trade is the growth in world real than imagined by previous investigators. output. Specifically, each 1 percent change in the growth of world real output was associated over the (3) Third, the evidence from surveys of participants 1959-82 period with approximately a 2 percent change in international trade points in the same direction102— in the volume of world trade in the same direction that is, it suggests that exchange rate uncertainties (see Table 9 in the Appendix). This suggests that the have not had an important impact on either the volume major factor underlying the slower average growth or cost of international trade or investment, at least rate in world trade during 1973-82, as well as the relative to other sources of instability in the system. standstill of trade in 1980-82, was the slower growth Again, however, the Fund staff study cautions that in world real economic activity during those periods.99 such evidence has to be interpreted with care because This finding applies not only to world trade but also the samples in these surveys are generally small, to the trade of important country groups (e.g., OECD because they cover predominantly large diversified real imports), to bilateral as well as aggregate trade firms which are perhaps better able to cope with relationships, to other time periods, and to other exchange rate variability than smaller firms, and be- empirical studies (e.g., Blackhurst and Tumlir (1980), cause some respondents may have implicitly assumed Anjaria and others (1982), and Bergsten and Cline that pegged rates would have been just as unstable. (1983)). Also, it is worth observing that the same surveys (2) Once the influences of real output and relative usually showed that firms had taken some specific traded goods prices on trade flows are accounted for, actions to cope with the risks created by greater there is little econometric evidence that exchange rate currency volatility, including, inter alia, creation of variability has had a statistically significant negative new departments to monitor foreign exchange expo- independent impact on the volume of trade.100 Again, sure, allocation of more of senior management's time the results appear to be quite robust—applying to both to exchange exposure decisions, greater use of forward aggregate trade flows over time and individual country cover, more systematic distribution of liquid assets trade flows over time, to trade patterns across countries among different currencies, diversification of produc- at a given point of time, and to real as well as nominal tion sources, increased sharing of the exchange risk exchange rates. Thus, even though both exchange rate in currency invoicing, etc. In this connection, the variability and exchange rate uncertainty have clearly development of interbank markets and of various increased in moving from the adjustable peg system futures markets (e.g., the International Money Market to managed floating, any independent adverse effects created by the Chicago Mercantile Exchange, the New of this increased variability and uncertainty on trade York Futures Exchange, the London International volumes have thus far eluded conventional trade equa- Financial Futures Exchange, etc.) has made manage- tions. However, as pointed out in the Fund staff study ment of forward cover substantially easier for enter- (IMF (1984 a, p. 36)):101 prises. Similarly, the buoyancy of long-term invest- ments under floating rates has been aided by the more The failure to establish a statistically significant link intensive use of various techniques (e.g., parallel loans, between exchange rate variability and trade does not, of swap arrangements, simulated dollar loans, etc.) for course, prove that a causal link does not exist. It may covering longer-term exchange risk. All of this suggests that even if increased exchange rate uncertainty has 99 In 1963-73 world output (excluding construction and services) not deterred firms from engaging in foreign trade and rose by 6 percent annually versus 3 percent for 1974-80, 1 percent in 1981, and 0 percent in 1982. investment, and even if they have not passed on any 100 Exceptions to this general conclusion can be found in the increased hedging costs to consumers, it has changed econometric studies by Coes (1981), Thursby and Thursby (1981), rather significantly the ways in which firms conduct and Cushman (1983). However, the Fund staff study (IMF (1984 a)) suggests that these studies establish an empirical link between business. exchange rate variability and trade volumes that are only subject to particular conditions. (4) Indirect effects of exchange rate variability on 101 It should also be kept in mind that the Fund staff study (IMF the structure of domestic output—and thereby on the (1984 a, p. 37)) does not address the question of ". . . whether or not prolonged shifts in underlying conditions that cause sustained 102 The surveys covered in the Fund staff study (IMF (1984 a)) departures from some medium-term trend in exchange rate rela- include: Duerr (1977), Fieleke (1978), Group of Thirty (1980); and tionships are harmful for trade." Blin, Greenbaum, and Jacobs (1981).

43

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM

level and pattern of international trade—are extremely Yet another counter-example to the thesis of exchange difficult to trace. Exchange rate uncertainty is usually rate protection is the steel sector. The restrictiveness only one of many considerations in each individual of trade policy in this sector has been increasing in decision to invest at home or abroad, expand output, the United States and the European Community since merge with another enterprise, and so forth. Consistent the mid-1970s, notwithstanding the exchange rate fluc- with this proposition, the data do not reveal any tuations experienced by their currencies over this obvious association between exchange rate variability period. Trade policies in high technology industries and either the trend rate at which resources are being (where strategic consideration seems to dominate de- shifted from traded to nontraded sectors, or interna- bate) and in the automobile sector (where the energy tional concentration in export production, or in the crisis drastically shifted the preference to small fuel- rate of real nonresidential gross fixed capital formation efficient cars at the expense of U.S. automakers) also relative to GNP.103 It is not clear whether the continuing do not quite fit the mold.104 trend toward larger firmsan d to increased international None of this evidence should be taken to imply that investment represents a rational reaction to increased exchange rate factors are unimportant in the generation uncertainty about relative factor and product prices in of protectionist pressures; rather, it suggests that the different markets, or instead a longer-run tendency pressures, and the extent to which authorities yield to (from the time before floating rates) attributable to the them, depend on a host of factors in addition to effects of greater international integration of markets exchange rates. Bergsten and Cline (1983) have re- and to the impact of technological progress on the cently given some specificity to this point by outlining nature of production processes. Similarly, the failure the features of a general model of protection. In their of the volume of business fixed investment (relative model, the degree of pressure for new trade restrictions to GDP) to decline during the period of floating rates will be greater, inter alia, the higher is the aggregate as a whole in the face of presumably greater uncertainty unemployment rate in the host country, the larger is may indicate that effects of uncertainty were out- the increase in unemployment over the recent period, weighed by the need to invest more heavily in energy the smaller and less generous are existing trade ad- exploration and conservation (after the larger rise in justment programs, the higher is the degree of eco- the relative price of energy). nomic interdependence (proxied by the ratio of imports (5) The argument that prolonged deviation of actual to GNP in the host country), the higher is the ratio of exchange rates from equilibrium exchange rates gen- imports to consumption, the larger is employment in erates protectionist pressures is certainly plausible, import-competing industries, the higher is the level of but there are too many other forces at work to accept general government intervention in the host country, an unequivocal relationship between exchange rates the higher is the degree of overvaluation of the host and the stance of trade policy. For example, many of country's exchange rate, the higher is the extent of the current protectionist measures have been sector trade restrictions maintained by major foreign coun- oriented or country specific rather than general and tries, and the greater are negative perceptions of the have been influenced by long-lasting shifts in compet- General Agreement on Tariffs and Trade (GATT) itiveness arising from factors other than exchange rate machinery for resolving trade disputes. On the other shifts. Such a case is clothing and textiles, where side of the coin, Bergsten and Cline (1983) expect the restrictions have been directed against developing pressures for new trade liberalization to be greater, countries with a comparative cost advantage in this inter alia, the higher is the existing inflation rate, the sector and where these restrictions have become pro- higher is the economy-wide ratio of exports to GNP, gressively more severe over a quarter of a century and the greater is the perception of the GATT system almost irrespective of changes in the exchange regime to handle trade policy problems effectively. Of course, or of the degree of exchange rate variability. Another the major obstacle to actually estimating such a model example is protection of the agricultural sector, which is the development of an index of protection to serve is fairly entrenched in most industrial countries and as the dependent variable. Nevertheless, just the which is motivated to a significant extent by socio- specification of such a model is sufficient to demon- political concerns. Here, it is hard to relate the high strate how hard it is to separate exchange rate influ- and (as some observers would argue) increasing level ences from other influences on protection. of protection over the past two decades to particular exchange rate movements among the major currencies. 104 In addition, there are examples of moves toward trade liber- alization being introduced in the face of large exchange rate varia- 103 At the same time, the Fund staff study (IMF (1984 a)) did find bility (e.g., the Tokyo Round of Multilateral Trade Negotiations some suggestive evidence that the variance or volatility of resource in 1973-79 and the later stages of the Generalized System of shifts in and out of the foreign sector was greater in 1974-82 than Preferences, which was phased in by the industrial countries during in 1960-70. 1966-76).

44

©International Monetary Fund. Not for Redistribution The Adaptability of Floating Rates

To summarize, it cannot be doubted that an increase in industrial countries, significant new natural resource in uncertainty in such a key price as the exchange rate discoveries, several serious regional conflicts, and is of major consequence for the conduct of international sometimes large intercountry differences in inflation trade and investment. This is not the same thing, rates, in monetary policies, and in policy mixes—all however, as saying that floating rates act as a strong have been accommodated without either suspending deterrent to trade and investment. This latter statement the operation of exchange markets or implementing needs to be modified to recognize that (1) exchange wide-scale restrictions on trade and capital flows. rate variability is only one dimension of the total Indeed, in such an environment managed floating may uncertainty associated with international transactions, well have been the only system that could have (2) exchange rate changes often reflect uncertainty in functioned continuously. their basic underlying determinants, and—perhaps most On a less general level, there would seem to be at important—(3) firms and individuals can and appar- least four factors behind the present system's relatively ently do resort to a wide variety of methods to reduce high adaptability or resiliency. the impact of exchange rate uncertainty on their (1) Because the present codes of conduct (i.e., the operations. By and large, existing empirical work Articles of Agreement) permit Fund members to have suggests that if exchange rate variability does adversely a wide choice of exchange arrangements, it is possible affect international trade and investment, it does so in to accommodate different national preferences with a manner that is too subtle and indirect to be captured respect to flexibility of exchange rates and to the mix in conventional trade volume equations alone. On the of domestic economic policies—that is, those countries important issue of protectionism, volatility and longer- for which the benefits of a fixed exchange rate are term disequilibria in exchange rates represent just one viewed as far outweighing the costs can opt for pegged more misguided rationale for seeking and/or granting arrangements, whereas those who feel that exchange protection. Until more is known about the determinants rate flexibility is an indispensable policy instrument of actual protection, one cannot adequately assess the can opt for floating rates. Between the two extremes, independent role played by exchange rates in that there is room for adjustable pegs as in the European politico-economic process. It is not too early, however, Monetary System and for heavier and more frequent to conclude that where and when protection is used exchange market intervention within the countries to offset exchange rate movements, the results are classified as "independently floating." While this may sure to be detrimental to the international adjustment be a less desirable situation than one where all countries process because resource allocation will suffer, needed could agree on the appropriate degrees of exchange structural changes in the economy will be delayed, rate flexibility and of economic policy convergence retaliation by other countries will be encouraged, and and would act on those common agreements, it is an inappropriate pattern of exchange rates will be likely to be preferable to a situation where strong perpetuated. differences in view exist and yet the same degree of flexibility and convergence is imposed on all partici- pants. In this connection, one probable reason for the The Adaptability of Floating Rates viability of the European Monetary System is that, Ten years is probably too short a time to proclaim even given the joint political commitment to its success, any system as "adaptable" or "robust." For example, its architects foresaw the need to accommodate some the international gold standard operated from the 1870s intercountry differences (e.g., wider margins for the to 1914 and then was briefly revived in the late 1920s. Italian lira and special financial measures for the less Great Britain, however, was on a legal gold standard prosperous members). By the same token, the Bretton beginning in 1816 and on a de facto one from 1717 on Woods system operated successfully while there was (Cooper (1982, p. 3)). Even the immediate predecessor a common view about the assignment of responsibilities of managed floating—the adjustable peg system— for adjustment and for exchange rate action between lasted about 25 years. By the standards of past ex- the reserve center country (i.e., the United States) change rate systems, floating rates are of relatively and the others, but it broke down when such agreement recent vintage and their capacity to adapt has not been was no longer forthcoming. fully tested. (2) A second related reason for the present system's Nevertheless, given the events of the past decade, viability is that it permits decentralized "market- it is easy to be impressed by the resiliency of the based" decisions to act as a safety valve when more present system. Two major changes in the price of centralized decisions about adjustment responsibilities energy products and the associated large changes in and exchange rate alignments do not prove possible. current account positions, a number of important bank Again, this probably represents a second-best solution failures, many changes in economic policy strategies to one where countries could always coordinate poli-

45

©International Monetary Fund. Not for Redistribution III • EVALUATION OF THE PRESENT EXCHANGE RATE SYSTEM cies successfully and where reliance on the sometimes the viability of the system is enchanced but not without short-sighted and sometimes narrow view of the market cost. was not required. But the fact that the market does (4) A fourth factor that is important for the resiliency "take a view" precludes the need to convene a series of the present system is that the move toward greater of high-level policy meetings when conflicts inevitably use of exchange rates in the adjustment process has arise.105 Also, at least over the long run, it is by no not been accompanied by the atrophy of the global means clear that the market is a poorer judge of the financing mechanism, either official or private. Given right pattern of rates than are the country authorities. the shocks to the system in the 1970s, the combination of exchange rate movements and expenditure-reducing (3) The present system provides enough "flex" in policies could not alone have produced an acceptable exchange rates to avoid what was perhaps the fatal solution; financingwa s absolutely necessary to achieve flaw in the Bretton Woods system—namely, the an economically optimal and politically acceptable incompatibility of fixed exchange rates and narrow speed of adjustment. This feature of the present system margins with high international mobility of capital. was most dramatically illustrated by the recycling of Given the fact that the financial resources available to oil revenues, but there are many other examples, private market participants are always much greater including some where official financing was critical than those of central banks, and given the rapidity (e.g., recent loans by the Fund and the Bank for with which the market view on a given exchange rate International Settlements to major borrowers among can change, any system that places stringent limits on the developing countries). Again, however, the reduc- short-run exchange rate movements is susceptible to tion of one type of risk increased others. In this case, successful speculative attack. This is why some ob- the same expanded role of commercial bank lending servers doubt the feasibility of re-establishing an ad- in balance of payments financing that proved so useful justable peg system.106 Williamson (1979, p. 26) for in the oil crisis posed new risks to the system later on example, argues: when that lending did not take adequate account of the changed circumstances of some major borrowers . . . the fundamental development that has undermined (i.e., slower growth in export markets, weaker terms the feasibility of the adjustable peg is the growth of capital of trade, much higher real interest rates). mobility. Since that growth is most unlikely to be reversed . . . monetary reformers should draw the conclusion In summary, it would not be safe to say that the that, whatever other exchange-rate regime they may present system is absolutely immune to collapse. recommend in particular situations (fixed rates, a crawling Nevertheless, because the present exchange rate sys- peg, a managed float, or a free float), the adjustable peg tem essentially rose directly from the ashes of its is not a viable option now or for the future. predecessor, it carries with it various "escape valves" (principally, more flexibility in exchange rates and Of course, that same "flex" in exchange rates that more decentralized decision making) that make it less provides a defense against "hot money" and that vulnerable to the same problems that plagued its increases the riskiness of "one way bets" by specu- predecessor (albeit at the cost of creating some new lators can become a liability if exchange rate move- ones). Relative to the Bretton Woods system, the ments go far beyond fundamentals. In other words, present exchange rate system is more heterogeneous, more flexible, more pluralistic, more market oriented 105 Kenen (1979, p. 163) makes this same point: "In the flexible and slightly more forgiving of idiosyncratic policy rate system we have found a device for achieving exchange rate changes without raising them to the highest level of public policy behavior. None of these characteristics would neces- both domestic and international." sarily be desirable in an ideal world of consistently 106 On this point, de Vries (1980) makes a distinction between the disciplined and coordinated macroeconomic policies, Bretton Woods type of adjustable peg system (under which exchange of rapidly stabilizing private speculation, and of perfect rates were changed infrequently and by large amounts) and the adjustable peg system (the snake) operated by some European foresight. But in the imperfect real world in which countries in 1972-78 (under which exchange rates were adjusted exchange rate systems have to operate, these same frequently and in small amounts). He considers the former but not characteristics as a group represent a workable and the latter to be incompatible with high international mobility of capital. reasonable second-best solution.

46

©International Monetary Fund. Not for Redistribution