Verifying exchange rate regimes

Public Disclosure Authorized Public Disclosure Authorized POLICY RESEARCH WORKING PAPER Verifying Exchange Rate Regimes X31 2397 One reason intermediate exchange rate regimes have fallen out of favor is that they are not transparent or easyto Jeffrey Frankel verify. A simple peg or a Eduardo Fajnzylber simplefloat may be easierfor Sergio Schmukler market participantsto verify Luis Servgn than a more complicated intermediate regime. Public Disclosure Authorized Public Disclosure Authorized wpv The World Bank DevelopmentResearchGroup Macroeconomicsand Growth July 2000 U POLICY RESEARCHWORKING PAPER 2397 Summary findirngs Credibility and transparency are at the core of the current debate about exchange rate regimes. The steady growth in the magnitude and variability of international capital flows has complicated the question of whether to use floating, fixed, or intermediate exchange rate regimes. Emerging market economies are abandoning basket pegs, crawling pegs, bands, adjustable pegs, and various combinations of these. One of several reasons intermediate regimes have fallen out of favor is that they are not transparent; it is very difficult to verify them. Verifiability is a concrete example of the principle of "transparency" so often invoked in discussions of the new international financial architecture but so seldom made precise. A simple peg or a simple float may be easier for market participants to verify than a more complicated intermediate regime. Frankel, Fajnzylber, Schmukler, and Serv6n investigate how difficult it is for investors to verify from observable data whether the authorities are in fact following the exchange rate regime they claim to be following. Of the various intermediate regimes, they focus on basket pegs with bands. Statistically,it can take a surprisingly long span of data for an econometrician or investor to verify whether such a regime is actually in operation. The authors find that verification becomes more difficult as the regime's bands widen or more currencies enter the basket peg. At the other extreme, they also analyze regimes described as free floating and find that in some cases the observed exchange rate data do validate the announced regime. This paper-a joint product of Macroeconomics and Growth, Development Research Group, and the Regional Studies Program, Latin America and the Caribbean Region-is part of a larger effort in the Bank to understand alternative currency regimes. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact EmilyKhine, room MC3-347, telephone 202-473-7471, fax 202-522-3518, emailaddress kkhine@worldbank.org. Policy Research Working Papers are also posted on the Web at www.worldbank.org/research/workingpapers. The authors may be contacted at jeffrey_frankel@harvard.edu, efajnzyl@ucla.edu, sschmukler@worldbank.org, or Iserven @worldbank.org. July 2000. (64 pages) The PolicyResearchWorkingPaperSeriesdisseminatesthe findingsof work in progressto encouragethe exchangeof ideasabout developmentissues.An objectiveof theseriesis toget thefindingsout quickly,even if thepresentations arelessthanfullypolished.The paperscarrythenamesof the authorsandshouldbe citedaccordingly.Thefindings,interpretations, and conclusionsexpressedin this paperareentirelythoseof the authors.They do not necessarilyrepresentthe view of the WorldBank,its ExecutiveDirectors,or the countriesthey represent. Produced by the Policy Research Dissemination Center VerifyingExchange Rate Regimes Jeffrey Frankel Harvard Universityand NBER Eduardo Fajnzylber Universityof California,Los Angeles Sergio Schmukler WorldBank Luis Serven WorldBank JEL ClassificationCodes:F31, F32, F33, F36 Keywords:verifiability,transparency,exchange rate regimes, basket peg, bands, target zone, floating ' We are gratefulto SebastianEdwards, Miguel Kiguel, Javier Ortiz, Ilan Goldfajn, and participants at the NBER/CEMAInter-American Seminar on Econornicsfor their useful feedback. Changqing Sun performed excellent research assistance work. We also thank Klaus Schmidt-Hebbel,Roberto Steiner, Matias Tapia Gonzalez, Alejandro Werner, and Roberto Zahler for making their data available to us. This paper was prepared as part of the Regional Studies Program of the Latin American and Caribbean Region of the World Bank. E-mail addresses: ieffiey f Iserven(cVworldbank.oiv hrankel(ciTarvard.cdu, efaJnzvl(iucla.edu, sschmukler(iaworldbank.orp, I. Introductionand The CornersHypothesis The choice of exchange rate regime - floating, fixed, or somewherein betweenis an old question in international monetary economics. But the steady increase in magnitude and variability of international capital flows has complicated the question. This is particularly the case for the developing countries that in the 1990s became fullfledgedparticipants in internationalfinancial markets. A major new element in the debate is the proposition that emerging market countries are, or should be, abandoning basket pegs, crawling pegs, bands, adjustable pegs, and various combinationsof these. The currently-fashionableview is that countries are being pushed to the "corners," the extremes of either free floating or firm fixing.The intermediateregimes are said to be no longer viable. This propositionis variously called the hypothesis of the vanishing intermediateregime, the missing middle, or the corners solution. Its life history has gone from birth to conventional wisdom in a remarkably shortperiod of time. The motivation of this paper is the observation that, as fashionable as this propositionhas become, few of its proponents, if any, have offered an analyticalrationale for it, let alone a fully worked out theoretical model. Our aim is to offer a possible theoretical rationale. We seek to introduce the notion of verifiability, and to suggestthat a simplepeg or a simple float may be more verifiable by market participantsthan a more complicatedintermediateregime. Verifiabilityis a concrete instance of the more general principle of "transparency" that is so often invoked in recent discussions of the new internationalfinancial architecturebut so seldommade precise. I L.aMotivation Consider the exchange rate regime that a number of emerging markets had in the 1990s: a band around a central parity that itself is a basket with a rate of crawl. So far as existing theory is concerned, the complexity of this arrangementhas no implications for its credibility. But, in truth, when a central bank announces a regime of this type, the public has no way of verifying quickly, by observing the exchange rate, whether the central bank is doing what it claims to be doing. A centralbank does not earn credibilitymerely by announcinga monetaryregime with a nominal anchor such as the exchange rate, even if its intentions are sincere. The public will judge credibility from data available to it. If the announced exchange rate regime is a simple dollar peg, a market participantneed only check that the exchange rate today is the same as the exchangerate yesterday,in order to verify that the central bank is indeed following its announced policy. If the announced regime is a pure float, a participant can essentiallycheck every month whether the central bank has intervenedin the market by seeing whether its reserve holdings have changed. Under the basket band, by contrast,the market participantneeds more months of data in order to be able to verify that the central bank is indeed implementingthe announcedpolicy. When comparing the corners, simple pegs tend to be more immediately verifiable than floating regimes. Typically, a market participant needs some extra piece of information, like reserves, or more data to check that an exchangerate is truly floating. How many months of data he or she needs is the central analyticalexercise of this paper. We are not claiming that verifiabilityis necessarily the complete story behind the purportednon-viabilityof intermediateregimes. And we are certainly not claimingthat it 2 is the only criterion,or even the most importantcriterion,in the larger debate about fixed and floating exchange rate regimes. Many other factors, whether from the traditional optimum currency area literature or the newer criteria associated with credibility and financialmarkets,need to be taken into account.' Our goal is rather to offer an attempt at what, so far as we are aware, may be the first explicit analytical rationale for the propositionthat intermediateregimes are less viable than the cornerregimes. In this paper, we demonstratethe difficultiesof verifiabilityfor the case of a band around a basket peg. We believe that the same difficulties apply to other intermediate exchange rate regimes, such as a managed float or adjustable peg. One could model a managed float, as a central target and a central bank policy of intervening partially to offset market forces when they push the exchange rate away from that target. But one would have to estimate the central target, and measure somehow the pressure of current market forces in order to figure out to what extent the authorities were intervening to resist them, a difficult econometric exercise. One could model an adjustable peg as a fixed exchange rate with an escape clause: the central bank has an explicit or implicit rule of abandoning the peg when an exogenous shock of a particular size occurs, and when a particular percentage of its foreign exchange reserves have been exhausted. Verifying that sort of rule would be even harder than the others because usually few relevant observationswill occur in the sample period, and even when the adjustment takes place, there is little way in practice of verifyingwhether on the one hand the putative exogenous shock in fact occurred, or on the other hand the government's commitmentto monetary discipline was not sincere in the first place. We choose to explore verifiability for the case of the basket band rather than the other examples because it is a cleaner econometric 'Two recent reviews are Larrain and Velasco (1999) and Frankel (1999). 3 exercise. We also look at countries that are believed to be floating, to offer a contrast to those that are believed to follow basket bands. This paper explores the amount of informationthat it takes for market participants to verify announced exchange rate regimes from observed data. The goal of the paper is to show the difficulty to verify exchange rate regimes and how this varies with regimes. To our knowledge, this is the first paper that performs this type of exercise. We use observed exchange rate data and simulated data to provide empirical estimates. The fact that countries vary their exchange rate regime over time allows us to run this experiment for regimes of different complexity. Regarding bands, the paper confirms the intuitive notion that wide bands are harder to verify than narrow bands. It is often difficult or impossible to estimate the weights of the central parity with only one or two years of data. Regarding regimes announced as free-floating, the paper shows that in some cases the exchange rates observed under such regimes are correlated with those of major currencies. In this sense, they behave similarly to the basket countries. It is not straightforward to verify them, when only using exchange rate data. To complement the tests performed with real data, we run Monte Carlo experiments to obtain more general conclusions and to provide results regarding the amount of information necessary to estimate regimes of interest. Monte Carlo experiments,displayed in the Appendix, confirm that more complex regimes take a larger amount of data to be verified. The Monte Carlo exercise shows the role of a number of factors in determining verifiability:the band size, number of currencies in the basket, the rate of crawl, sample period, periodic adjustments of the central parity. The results 4 confirm the intuitionthat the amount of infornation necessaryto verify the exchangerate regimes increaseswith the complexityof the regime. The rest of the paper is organized as follows. The rest of the introduction introduces the verifiabilityproblem. Section II describes the framework and empirical strategy used to verify exchange rate regimes. Section III presents estimations for the case of exchange rate bands. Section IV shows the results from free-floating regimes. The main conclusionsare summarizedin SectionV. Appendix 1 displays a small Monte Carlo exercise extending the study of regime verification to simulated models, and Appendix 2 gives more details on the construction of the numeraire and the estimated models. I.b IntellectualOriginsof the CornersHypothesis What is known about the origins of the hypothesis of the vanishing intermediate regime? The original reference is believed to be Eichengreen (1994). The context was not emerging markets, but rather the European Exchange Rate Mechanism. The ERM crisis of 1992 and band-wideningof 1993 suggestedto some that a gradual transition to European Economic and Monetary Union, where the width of the target zone was narrowed in steps, might not be the best way to proceed after all. (Crockett, 1994, made the same point.) Obstfeld and Rogoff (1995) concluded, "A careful examination of the genesis of speculativeattacks suggeststhat even broad-bandsystems in the current EMS style pose difficulties, and that there is little, if any, comfortable middle ground between floating rates and the adoptionby countries of a comnmoncurrency." The lesson that "the 5 best way to cross a chasm is in a singlejump" was seemingly borne out subsequently,by the successfulleap from wide bands to EMU in 1998-99. After the East Asia crises of 1997-98, the hypothesis of the vanishing intermediate regime was applied to emerging markets. In the effort to "reform the internationalfinancialarchitecture" so as to minimize the frequencyand severity of crisis in the future, the proposition was rapidly adopted by the international financial establishmentas the new conventionalwisdom. For example, Summers (1999a)2 : "Thereis no singleanswer,but in light of recent experiencewhat is perhapsbecoming increasinglyclear - and will probably be increasingly reflected in the advice that the international community offers - is that in a world of freely flowing capital there is shrinking scope for countries to occupy the middle ground of fixed but adjustable pegs. As we go forward from the events of the past eighteen months, I expect that countries will be increasinglywary about committingthemselvesto fixed exchange rates, whatever the temptations these may offer in the short run, unless they are also prepared to dedicate policy wholeheartedlyto their support and establish extra-ordinarydomestic safeguards to keep them in place." Other high-profile examples include Eichengreen (1999, p.104 -105 ), MintonBeddoes (1999) and Council on Foreign Relations (1999, p.87). The International Monetary Fund has now agreed that countries that get into trouble by following an internediate regime will in the future not be bailed out, though it qualified the scope of Other high-profileexamplesinclude Eichengreen(1999, p.l04-105), Minton-Beddoes(1999) and Councilon ForeignRelations(1999,p.87). 2 6 the generalizationa bit, for example,by allowinga possible exception for "systemically" important countries. It may be that the Economist (1999, p.l5-1 6 ) is right that "Most academics now believe that only radical solutions will work: either currencies must float freely, or they must be tightly tied (through a currency board or, even better, currency unions)." But the propositionremains to be modeled, let alone proven. It seems intuitively right that these countries, facing finicky international investors and rapidly disappearing foreign exchange reserves, had little alternative but to abandontheir pegs and baskets and bands and crawls and move to a float, unless they were prepared to go to the opposite corner. But what is the rationalefor this proposition? I.c Lack of TheoreticalFoundations What is the analytical rationale for the hypothesis of the disappearing intermediate regime (or the "missing middle")? Surprisingly, none currently exists, to our knowledge. At first glance, it appears to be a corollary to the principle of the Impossible Trinity.3 That principle says that a country must give up one of three goals: exchange rate stability,monetary independence,and financial market integration. It cannot have all three simultaneously. If one adds the observation that financial markets are steadily becoming more and more integratedinternationally,that forces the choice down to giving up on exchange rate stability or giving up on monetary independence. But this is not the 3 Summers (1999b,p. 326)is explicit:"...the coreprincipleof monetary economics is a trilemma: that capitalmobility,an independentmonetarypolicy,and the maintenanceof a fixed exchangerate objective are mutuallyincompatible.I suspectthismeansthat as capitalmarketintegrationincreases,countrieswill be forcedincreasinglyto morepurefloatingor morepurelyfixedexchangerateregimes." 7 same thing as saying one cannot give up on both, that one cannot have half-stability and half-independence. There is nothing in existing theory, for example, that prevents a country from pursuing a target zone of moderate width. The elegant line of target-zone theory begun by Krugman (1991), in which speculation helped stabilize the currency, always assumed perfect capital mobility. Similarly, there is nothing that prevents the government from pursuing a managed float in which half of every fluctuation in demand for its currency is accommodatedby interventionand half is allowed to be reflected in the exchange rate. (To model this, one need only introduce a "leaning against the wind" central bank: reaction function into a standard monetary model of exchange rate determination.) And there is nothing that prevents a country from pursuing a peg that is abandonedwheneverthere is a shock large enough to use up half its reserves. Another justification that has been offered is that when a government establishes any sort of exchange rate target, as did the East Asian countries, its banks and firms foolishly underestimate the possibility of a future break in the currency value.4 As a result, they incur large unhedged dollar liabilities abroad. When a devaluation occurs, their domestic-currencyrevenues are inadequatefor servicing their debts, and so they go bankrupt, with devastatingconsequencesfor the economy. "It follows that in a world of high capital mobilitythere are only two feasible approachesto exchange rate policy. One is not just to peg the exchange rate, but to lock it in - the Argentine strategy....The vast majority of countries will ... have to follow the other alternative of allowing their currencies to fluctuate. If the exchange rate moves regularly, banks and firms will have an incentiveto hedge their foreignexposures..." (Eichengreen,1999,p.105). 8 There is little doubt that the focus on unhedged foreign-currency debt describes accurately why the 1997-98 devaluations were economicallydevastating to East Asia. But the argument, as stated, has some weaknesses. First, it appears to depend on irrationalityon the part of banks and firms. Second, it appears to imply that a country would be better off by gratuitouslyintroducing extra noise into the exchange rate, to deter borrowersfrom incurring unhedged dollar liabilities. This seems unlikely to be right. Third is the point emphasizedby Ricardo Hausmann:because foreigners are unwilling to take open positions in the currencies of emerging-marketcountries, the admonition to avoid borrowing in dollars is to some extent an admonition to avoid borrowing at all. (An admonitionto hedge the dollar exposure is not helpful; someonehas to take the other side of the futures contract, and this will be difficult in the aggregate if foreigners are unwillingto take the open position.) It may well be that this is the right road to go down, that exchange rate volatility is a way to put some sand in the wheels of the excessive capitalmovements, and that a lower volume of total debt is a good outcome. But if this is the argument, the proponents should be explicit about it. In any case, it seems doubtful that this argument couldbe captured by conventionalmodels. A third possible justification is that governments that adopt an exchange rate target, and sometimelater experience a major reversal of capital inflows, tend to wait too late before abandoningthe target. As of 1998, we thought we had learned that the one thing an emerging-market government can do to minimize the eventual pain from a currency crisis is to try to devalue early enough (or else raise interest rates early enough, as would happen automatically under a currency board - anything to adjust, rather than Theversionof thisargumentin Eichengreen(1999,p.104)overstatesthe extentto whichthe EastAsians had"a statedcommitmentto the peg," as most commentatorshavedone as well. In fact few of the East 4 9 try to finance an ongoing deficit). Mexico, Thailand and Korea made the mistake of waiting too long, until reserves ran very low, so that by the time of the devaluationthere was no good way out, no combination of interest rates and exchange rate that would simultaneously satisfy the financing constraint externally and prevent recession domestically. But exiting from an exchange rate target can be difficult politically. The lesson is drawn that, to avoid this difficulty, govermnents should either adopt a rigid institutional fixed-rate commitment (such as the currency boards of Hong Kong and Argentina),or, if not prepared to do that, abandonthe peg early.5 On this basis, when Brazil in the autumn of 1998delayed the seeminglyinevitable jettisoningof the real target, many thought this would be a repeat of the earlier mistakes. Instead, when the devaluation finally came in January 1999, Brazil's trade balance improved sharply, the lack of confidence subsided, and output and employment subsequently performed far better than in neighboring Argentina. Thus it is more difficult to generalize from recent experience than widely believed. Furthermore, if we are to use government reluctance to exit a target arrangement as the basis of a model of the unviability of intermediate regimes,it seems that we would again require some sort of irrationality(or political constraints6 ) on the part of policy-makers. Thus, each of the three argumentsoffered - the impossible trinity, the dangers of unhedged dollar liabilities, and the political difficulty of exiting - contains some importanttruth. But none seems able to stand as a theoreticalrationale for the superiority Asiancountrieshadexplicitdollarpegs. 5 Eventhen we had a counter-example: Indonesiahad widenedthe band rightaway in 1997,andyet that didn't save it. Butone couldarguethat politicalinstabilitywouldhavedone Indonesiain no matterwhat. Taiwandevaluedpromptly,andsufferedlessthanthe others. 6 Governments mayhavean incentiveto postponedevaluationsuntil afterelections.SeeEmestoSteinand JorgeStreb(1998,1999). 10 of the corner solutions over the intermediateregimes. Is the comers hypothesis,then, just a misplaced manifestation of the temptation to believe that the grass is always greener somewhereelse? II. AssessingVerifiability The idea behind verifiability is that the government's announcement of an exchange rate regime is more likely to be credible if market participants can check for themselves from observable data that the announced regime is in fact in operation. Specifically, the goal of our paper is to study how long it takes for financial markets to identify from the data the rules guidingthe interventionbehavior of the authoritiesin the foreign exchange markets The process of verification can be modeled along the lines of statistical inference familiar to econometricians. We are not suggesting that market participantswill literally run OLS or other sorts of regressions, but rather that they must do something sirnilar implicitlyto process the available information. The paper's framework encompasses a broad variety of regimes - simple and basket pegs with bands and crawl as well as floating regimes. However, if a country follows an exact basket peg (i.e., with no band), the problem of statistical inference is of limited interest.7 In practice, however, there is almost always some range of variation in the observed exchange rate data, even if it is only within a narrow bid-ask spread quoted by the banking system, or within the +/- 1% range that constituteda fixed exchange rate under the rules of the Bretton Woods system. Then the problem of statistical inference is 11 not trivial. For bands of substantial width, the statistical inference can in fact be difficult, as we shall see. This is all the more true if one allows for the ever-presentpossibility of shifts in the parameters-basket weights, band width, rate of crawl, or level of parity-or changes in the regime altogether,especiallyif some of these shifts are not announced. In oturempirical analysis,we work with a set of emerging countries, for which we know the announcedexchange rate regimes. We will begin with an analysis of the basket bands followed by Chile and Israel during the late 1980s and 1990s.8 Given that Chile and Israel changed their band parameters over time, we are able to examine them under different regime configurations. Then, we move on to the regimes officially declared as floating followedby Brazil, Mexico, Peru, South Korea, and Thailand. If the currency in question is in fact following a basket band, the question of interest is how many data points are necessary,i.e., how much time must elapse, in order to verify that the regime is in fact in operation. In general, we will consider an anchored exchange rate regime to have been verified if it passes two tests. (1) We fail to reject the hypothesis that the exchangerate is followingthe announcedbasket peg. (2) We can find statistically significantbasket parameters,i.e., can reject the hypothesis that the currency is behaving like any "random" currency. These two tests are informative only if they have adequatepower. To judge the power of the tests, we perform the same tests with a randomly generated variable and with a freely floating exchange rate as the dependent variable. When using these latter variables we should reject the null hypothesis in (1) and fail to reject that in (2). In the case of floating regimes, since there are no announced that case, the announcement of a basket of N major currencies can be verified with N+1 observations, which is the number needed to estimate exactly the basket weights. As noted, however, this does not constitute verificationof an adjustablepeg since we don't observe the terms of the "escape clause." 8 A detailed description of these regimes can be found in Appendix Tables A.1 and A.2. 7In 12 pegs we only use the second test. Below we specify more explicitly the null hypotheses under consideration. If an announcedregime of basket bands does not pass these tests, one can argue that it is not verifiable, which suggests that the country cannot reap the credibility gains that an anchored exchange rate regime theoretically offers-credibility in the eyes of workers and producers who set wages and prices, and in the eyes of speculatorswho have the ability to attack the central bank's reserves and bring about a crisis. If viability requires verifiability,such a regime may not be viable. In the case of bands, we are especiallyinterested also in seeing how the ability to confirm the announced nominal regime is statistically affected by features such as the width of the band and the number of foreign currenciesin the basket. Our approach focuses on the empirical estimationof the parametersdescribing the exchange rate rule at different sample sizes - e.g., 50, 100, and 200 observations. The point estimates, their precision, and the tests of the above hypotheses constructed using them tell us how well can market participants identify the parameters of the regime when the latter is 10, 20, and 40 weeks old. For this empirical analysis, we need a basic frameworkand a testing procedure. The rest of this sectionis devoted to these questions. IIa Basic Framework We adopt a general formulationto "nest" a number of altemative regimes. We assume that the exchange rate for a given small country is given by a weighted 13 combinationof N foreign currencies, with a possible rate of crawl d and an error term. The exchange rate is: 9 St =c+dxt+'D(W,sit)+Et. (1) where s, is the spot exchange rate of the domestic currency with its value measured in terms of a numeraire that we will explain momentarily;s4,are the spot exchange rates of the major "strong currencies" measured vis-a-vis the same numeraire; d is the rate of crawl, which for now is assumed to be fixed during a given sample period;10 and wi are the weights given to the currencies included in the basket. Dependingon the specification of the basket, (D may take different forms, with the simplest one being the familiar N Simple Pegs, Basket Pegs, CrawlingPegs, CrawlingBaskets This general case captures many possible regimes, including simple pegs, basket pegs, crawling pegs, crawling baskets, target zones, certain forms of managed floating, and free floating. In the case of simple pegs, the value of the currency follows the exchangerates of the foreign currency to which it is pegged, plus the crawling rule, and a stochastic error. The latter is the error allowed or incurred by the government when setting the exchange rate. In the case of simple pegs, N (the number of currencies in the basket) is equal to one. Under basket pegs, N is bigger than one. Crawling pegs imply that d>O. Under crawling baskets,N>l and d>O. 9 The precisemodelsthat we estimatedare describedin Appendix2, whichalsoprovidesa descriptionof the procedurefollowedby Chile and Israel to constructthe basketused as centralparity in their band systems. 14 In the case of an exact peg, the error term would vanish, and an OLS regression of the domestic currency on the foreign currencies to which it is pegged would yield an fe equal to 1. Verificationis a trivial exercise, whetherthe peg is simple or to a basket. This can be easily illustratedby examiningthe behavior of the central parity in band regimes. Central parities behave like simple or basket pegs (with or without crawl), dependingon the regime. Frankel, Schmukler and Serven (2000) report estimations of a version of equation (1) above using as dependent variablethe Chilean peso central parity. In all the cases examined there, the weights of the central parity converge to their announced values almost immediately. In our present context, the pegged regime is verified instantaneously.Thus, in the remainder of this paper we concentrate on the cases of exchange rate bands (target zones) and floating regimes. Target Zones In a regime of target zones, a central parity is defined as a function of a single or multiple foreign currencies and the exchange rate is allowed to float within a prespecifiedband around this centralparity. Whenever it hits the boundary, the government intervenes to keep the exchange rate inside the band. In many cases, governmentsmake intra-band interventionsas well. In a target zone, the log difference between the observed spot exchange rate and the central parity, st, is determinedby the followingequation: -b s, ={b vt if st <-b if s, > b otherwise 10 Onealternativewouldbe to use past domesticor futureinflationratesrelativeto international inflation 15 where s, is defined by equation (1) above and b is the band width. According to theory, the distributionof v can be quite complicated. Even under two simplifyingassumptionsmade by Krugman (1991) in a famous article that generated a sub-field of research on target zones-that the band is 100% credible and that the authorities intervene only at the boundaries-the distribution is not normal, but rather follows a particular S-shape." But extensive empirical investigation of the European Exchange Rate Mechanism in the 1980s and early 1990s established that the spot rate does not in fact obey the predicted distribution. There are a number of likely reasons for this, among them the lack of full credibility of the zones12 and the prevalence of intramarginalintervention. For these reasons we shall assume in our work that v follows instead an autoregressiveprocess, of the form vt = pvt, + ut, where u is iid. In fact, we will focus on the random walk case of p = 1, in accordancewith most time series analyses of exchange rates, which cannot reject the unit root hypothesis.'3 rates?where the authoritiesare believedto be followingan indexationpolicy. " When the spot rate draws close to one edge, speculators are aware that there is a limit on how far it can continue to move in that direction. The expected value will show a regression back toward the central parity. As speculators respond to that expectation, they will push the spot rate away from the margin, even without any intervention. 12 Imperfect credibilitywas in the event justified by realignments in the early 1980s, and especially by the ERM crises of 1.992-93. It is also relevant for the present exercise, which is entirely based on a starting point that assumes imperfect credibility. 13 In unreported results, we found that estimates of p were practically I in most regressions using equation (1). One extension for further research would be to use statistical distributions implied by more sophisticated versions of the target zone theory. Another would be to take the observed statistical distributionfrom historical episodes such as the ERM currencies in the 1980sor 1990s. 16 Managed Floating and Free Floating While there are many possible patterns of exchange rate intervention, our basic framework [equation (1)] only allows us to test whether d or wi are different from 0. In other words, the government is using some form of nominal anchor or crawling peg rule to guide its operations. Other forms of intervention are not nested in our specification14 Hence, we will consider that failure to reject that d=O and wi=Ois a characterization of a pure floating regime.15In such case, the disturbance term accounts for the entire variance of the exchange rate. The Choice of Numeraire The question of what to use as the numeraire to measure the values of the domestic and foreign currencies is a surprisingly subtle one. In the case of exact pegs it makes no difference - so long of course as the same one is used for both dependent and independent variables alike. The correct weights should emerge, with a perfect statistical fit, regardless of the numeraire. But in the general case, the choice of numeraire does make a difference. Past studies have used a variety of numeraires, including the consumerbasket of domestic goods (Frankel, 1993,which emphasized Asian currencies), the SDR (Frankel and Wei, 1995, which emphasized policies of European currencies), the Swiss franc (Frankel and Wei, 1994 and Ohno, 1999) and the dollar (Benassy-Quere, 1999). 14 It is possibleto assume that the governmentfollowsa variance-reducing form of interventionbut, without imposingsome a priori value for the varianceof the underlyingprocess,it is not possibleto identifythe interventionparameter. 15We use the term free-floating to referto a casewherethereis no correlationbetweenthe studiedcurrency and any of the strongcurrencies.It is possibleto arguethat underpure free-floating,marketforcesmight inducesomecorrelationwiththe currenciesof majortradepartners. 17 Upon further reflection,these measures are not quite right. We wish to consider regimes where the central bank monitors a central parity, but routinely allows appreciations or depreciations relative to that parity in response to such factors as inflation, unemployment,trade deficits or surpluses,various market pressures and so on. These factors are only partially accommodatedunder an intermediate regime such as a band or managed float, but they have a role nonetheless. We have not chosen to model explicitly these factors; they are comprised by the error term. The authorities are presumedto be trading off the long-term credibility benefits of stickingrelatively close to their central nominal parity against the monetary-independencebenefits of responding to short-term developments. But in framing this tradeoff, there is no reason for them to think of the departure above or below the centralparity in terms of dollars or a basket of goods, and still less reason to think in terms of Swiss francs. The most useful way to phrase these appreciationsand depreciationsis, rather, in terms of an effective exchange rate, that is, a weighted average of trading partners' currencies. In this paper we measure values of currencies in terms of a weighted basket of the G7 currencies. One possible set of weights is the bilateral trade shares of the smaller country in question. This has a drawback: it leaves out the role of all the other bilateral trade partners, as well as third-country markets and competitors. But most of those are linked to some combinationof the major currencies. Here we adopt the simple approach of using the G7 countries' weights in gross world production. In this way it is hoped that, for example, the large weight of the US will roughly reflect the importance of dollarlinked countries in the trade of Chile or Indonesiabeyond the share of the US in bilateral 18 trade of those two countries.'6 Thus, the exchangerates, both of the major currenciesand the currencies under study, are calculated as the number of units of the currency necessaryto purchase a weightedbasket of strong currencies.'7 II.b Empirical Strategy We use daily data in our empirical experiments.'8 To assess how verifiable different exchange rate regimes are, we use explicit statistical tests that attempt to replicate those implicitly carried out by financial market participants to leam about the actions of the monetary authorities. For countries that have announced a basket band such as Chile and Israel during the sample periods we use (examined in section III below), we seek to establish the amount of information (days) needed to reach a judgment on whether the data support the hypothesis that the exchange rate is following the announcedregime. In the case of currenciesthat have declared their regirne as a pure float - like post-crisis Mexico and Thailand (section IV below) - the purpose of the exercise is to offer a standardof comparisonfor the first set of currencies. A test that fails to reject the announcedregime for the currencies followingbasket bands, has low power if the same test also fails to reject an analogoushypothesis applied to floating currencies. We wish to see whether the public can distinguishthe two sorts of policies statistically, rather than having to rely on the assumption that it can instantaneouslyintuit the true policy of their central bank. 16 A secondadvantageof usingGDPweightsis that one does not need to obtainthe full set of bilateral trade dataand recomputea new set of weightsfor each country. But using bilateraltrade weightsis a possibleextensionforfutureresearch. 17 SeeAppendix2 fora detaileddescription of the constructionof the numeraire. 's Data on major currenciesand some of the emergingcountrieswas extractedfrom Bloombergand Datastream.Data for the case studiesof Chile and Israel was downloadedfrom the respectiveCentral Banks Web pages. 19 To make this approach operational, we summarize the exchange rate regime in terms of the basket weights in equation (1). Tests of hypotheses about the exchange rate regime then are just tests of hypotheses regarding the basket weights. The tests we performare the following. Test 1 (TI): Market participants test whether the weights obtained from empirical estimation of equation (1) are equal to the announced weights. Conditional on the announcementbeing true, we expect that this null will not be rejected. The null and alternativehypotheses are: HO:wi=announcedweights;HA: w1• announcedweights. Test 2 T2): The second test inquires more generally, whether we can reject that the currencyis freely floating. We assume that market participantsdo not know what the governmentis doing, for examplebecause the governmenthas not explicitly announceda regine, or else they do not necessarily believethe announcedexchange rate regime. The null hypothesis is that the value of the currency follows a random walk with or without drift. Therefore, we think of market participantsas testing if all the weights on the strong currenciesarejointly equal to zero. Formally, HO: w= 0 ... and ... WN = O; HA: wI#O ... or... wN• O. One problem with this approach is that Test 1 might fail to reject the null due to lack of power --e.g., if we work with too short a time sample. Market participants know instinctively that a failure to reject the regime is an informative finding only when that test would be capable of rejecting the regime in the case where it was false. To see if Test 1 has power, we complement it with another experiment in which we replace the dependent variable with fictitious data and with a floating exchange rate. Then, we test 20 the null hypothesisthat the weights are equal to the ones announced by Chile and Israel. We perform this experiment for the cases in which Test 1 fails to reject the null hypothesis. In this experiment,we expect to reject the null, given that we are using false weights. Analogously,to check that Test 2 is not rejecting the null hypothesis when it is true (i.e., it is not making a Type I error), we perform a similar experimentfor Test 2. In this case, we should fail to reject the null hypothesisof Test 2.19 To estimate equation (1) and carry out inference on its parameters, a variety of procedures are potentially applicable. In this paper we report results using a "naive" estimationprocedure, which we implicitlyassume to be the one that market participants apply to process the observed data. Specifically,we compute OLS estimates of equation (1) in firs.t differences. We do this for all the exchange rate regimes explored in the paper.20 While nmrecomplex models, such as those derived from the recent target zone literature, might offer some advantage in terms of consistency, their estimation would also require a vastly larger amount of data. Therefore, we work with these relatively simple specificationsto carry out our tests and illustrate the point of the paper. As independent variables for the basket band regimes, we use those currencies that were included in the announcedbasket. In some cases we found that some of these currencies were strongly correlated over some periods (particularly the Deutsche mark and French franc, both included in the Israeli basket), so that the estimations were plagued by severe multicollinearityand identificationof the specific weights was almost Alternatively to T2,whichchecksthenullthatallweightsarezero,weusedanothertestofthenullthat allthestrongcurrencies havethesameweight,obtaining verysimilarrejectionfrequencies. 20 The estimated modelsaredescribedin Appendix2. Wealsoexperimented withmoregeneralerrorcorrectionmodelsallowingfor long-runequilibriumand short-rundynamics,with the long-term relationship linkingthelevelof thedomestic exchange ratewiththelevelofthestrong-currency exchange rates.Ingeneralprecision waspoor,andthelong-run equilibrium poorlyidentified. Tosavespace,wedon't reporttheresultshere. 19 21 impossible. To solve this problem, we opted for computing also estimates of a "restricted" model combining the most highly correlated currencies, using the ratio of their announcedweights. We return to this below. III. VerifyingExchange Rate Bands Using the frameworkand empirical approach just described, this section focuses on the verifiabilityof the exchange rate bands followed by Chile and Israel over recent years. TI.aChile A number of successiveexchange rate regimes have been in place in Chile since the early 1980s.21 In 1982, Chile had a crawling peg vis-a-vis the US dollar, with daily devaluationsfollowing the difference between domestic and external inflation. The peg to the dollar continued until 1992, with bands of varying width around the central parity and with realignmentsof the central parity. In 1992, the government decided to adopt a target zone around a basket peg. The weights on the currencies deflning the centralparity changed over time and there were realignments,but the central parity was always tied to the US dollar (US$), the Deutsche mark (DM), and the Japanese yen (JY). Finally, in September 1999 the centralbank decided to float the peso. The entire period of exchange rate bands can be broken down into a number of sub-periodsdistinguishedby differentlevels of the central parity, basket composition and band width. To analyze the verifiability of Chile's exchange rate band system, we focus 21 A detailedchronologyof the exchangerate systemin Chileis presentedin TableA.1 in the appendix. 22 on seven of those sub-periods,selected on the basis of a minimum duration (specifically, those comprising at least 249 daily observations,amounting to approximately one year). The relevant parameters characterizingthese sub-periods are summarized in Table l.a. The first three sub-periods involve a peg to the US dollar with a band, while the last four involvea basket peg with a band. Figure 1 displays the observed exchange rate in terms of the weighted basket numeraire, along with the announcedbands. The figure shows that the trend of the peso has been to depreciate over time, with significant appreciations and depreciations on several occasions, and highlights the fluctuationsof the exchange rate within the band, as well as the gradual widening of the latter. In some periods, like 1991-92,the exchange rate is close to the lower band. In other periods, like 1994-95, the exchange rate fluctuates farther inside the band. After suffering pressure on the peso, the authorities decided to narrow the band from 12% to 3.5% in September 1998 to show their commitmentto the value of the peso. The band was widened back to 8% in December 1998. Table l.b reports the results of the verifiabilitytests using the Chilean exchange rate data, based on first-differenceOLS estimates of the basic equation. For each of the seven sub-periods under consideration, the table presents the cumulative rejection frequencies of the null hypotheses of Test 1 and Test 2 at increasing sample sizes - 50, 100 and 200 observations. For example, a rejection frequency of 100 for 50 observations in Test 2 means that in 100 percent of the estimationswith sample sizes smaller than 50 we can reject the null hypothesis that the weights are equal to 0.22In addition, the table also reports point estimates and standard errors of the weight of the US$ in the basket 23 defining the central parity (to save space, we omit the estimated weights of the other currencies). Finally, the last two columns of the table give a measure of the precision of the estimates, in terms of their mean absolute error - that is, the sum of absolute deviations of the estimated weights from their announcedvalues.23 For periods 4-7, when the central parity is defined by a basket of several currencies, the table presents two sets of results. The first set is based on an unrestricted version of the model, in which we attempt to estimate the individual weights of all currencies in the basket. As already mentioned, however, this procedure may run into difficulties due to the high correlation among some of these currencies in some subperiods, and therefore we also present results from a restricted model version combining the most highly-correlated currencies in the proportions dictated by the announced weights. In the Chilean case, this involved combining in such fashion the US dollar and the yen.24 The results in Table l.b show a clear difference between periods 1-3 and 4-7, regardless of whether we use the restricted or unrestricted model in the latter sub-periods. In the former sub-periods, the point estimates of the US dollar weight approach fairly quickly the announced weight (equal to one), especially in periods 1-2. In these two periods the estimated weights are not statistically different from the announced value (Test 1), but are statistically different from zero (Test 2) for any sample size. In turn, in sub-period 3, with an increased bandwidth (equal to 5%) relative to periods 1-2, the point estimate of the US dollar weight still approaches its announced value, although at a 22 Thefirst estimationstartswiththe minimumnumberof observations requiredto estimatethe models. Since the annoumcedweights sum up to 1, no rescaling is required. The correlation between the first differences of these two currencies exceeds .85 in some of the subperiods of analysis. 23 24 24 somewhatslower pace than in periods 1-2. However,we also find a higher rejection rate in Test I and a somewhatreduced rejection rate in Test 2. On the whole, these results tend to suggestthat the wideningband makes verificationmore difficult. In contrast, for periods 4-7, characterizedby a currency basket and much wider bands, none of the estimates in Table lb - whether restricted or unrestricted - appears close to the announced values even after a reasonably large number of observations. Precisionis much poorer than in the earlier periods, although the restricted estimates are in general substantiallymore precise than the unrestrictedones (see the last two columns in the table). In any case, both sets of estimatesappear clearlybiased; indeed, some point estimates are even negative. As a result, while Test 2 generally rejects the null of zero weights, Test 1 also rejects the announcedweights in most samples,and this applies both to the restricted and unrestrictedestimates. These results can be more easily understood with the help of Figure 3, which presents scatter plots of the observed exchange rate of the Chilean peso against the central parity. In the first three sub-periods, the points cluster along the 45-degree line, reflectinga relatively close match between the peso and the central parity. As the band widens in the last four periods, the peso can fluctuate further away from the central parity. This is particularly apparent in periods 5-7, whose scatter plots display little or no clusteringalong the 45-degreeline. Thus, it is not surprisingthat in the first three periods the basket weights defining the central parity can be estimated fairly precisely from the observedexchange rate data, while this is not the case in later periods.25 25 The scatterdiagram,alongwith Figure1, also providessomecluesfor the relativelypoorerverifiability of the third period vis-a-vis the first two. In the early part of the third period (approximately 50 observations), the exchangerate waspracticallypeggedto the upperpart of the band,but startingin early 1990 the peso started appreciating until it finally reached the lower band. This marked break in the 25 On the whole, the results for Chile strongly suggest that the widening of the band, together with the adoption of multiple instead of simple pegs, make verification of the announcedregime more difficult using simple econometric estimates. 26 III.b Israel Israel presents another interesting experience of basket band with weight changes and progressive widening of the band. During the periods on which we will focus, the band included the same five currencies [US dollar (US$), Deutsche mark (DM), British pound (BP), French franc (FF), and Japanese yen (JY)] and the bandwidth rose from 3% to 15%. The Bank of Israel had already introduced an exchange rate band around a basket in 1976.27 It lasted for a year before being replaced with a floating exchange rate, followed in tum by a dollar peg in 1985 and a basket peg in 1986, with basket weights determinedby trade shares and subject to relatively frequent revisions.28 At the beginning of 1989, the Bank of Israel reintroduced a band system by allowing the exchange rate to fluctuate within a region of ±3% around the currency basket defined by the five currencies already mentioned. The band was later widened to trajectory of the peso, clearly visible from the scatter plot in Figure 3, is behind the poorer performance of Tests 1 and 2 in the third period that is apparent from Table lb. 26 One could object that the contrast between the results we obtain for the earlier and later periods of Chile's band regime might be due instead to some underlying change in the behavior of the strong currencies or in the way the peso moved within the band. However, the intuition that verifiability is more difficult with wider bands and baskets with more currencies is confirmed by the Monte Carlo experiments in Appendix 1, which are not subject to those objections. 27 For a detailed account of the exchange rate policy in Israel, see http://www.bankisrael.gov.il and Appendix Table A.2. 28 The number of units of each currency in the new basket was originally determined according to its share in trade during the previous calendar year and to international cross rates. Since then, the trade shares were revised annually and when significant changes produced, the weights and units in the basket were recalculated. The number of units of each currency in the basket is kept constant, but its weight - 26 5% in March 1990, 7% in May 1995, and then gradually since June 1997, to reach 15% by the end of that year.29 Most importantly, since December 1991 a pre-announced, constant rate of crawl was added to both the midpoint and the band - a system known as a crawling band. Figure 2 shows the evolution of the Israeli exchange rate and the exchange rate band. One feature that stands out is the frequency of realignments of the central parity, particularly in the early years of the band. For the analysis of verifiability, we divided the sample into different sub-periods characterized by different bandwidth, basket weights and/or rate of crawl of the exchange rate band. Table 2.a lists the periods under consideration,their beginning and ending dates, and the relevant parameters of the band. In the case of Israel, collinearity among basket currencies is more of an issue than in Chile due to the larger number of currencies and, especially, to the simultaneous inclusion of the French franc and DM in the basket.30 Thus, for the restricted model estimation we combined the DM with the franc and the US dollar with the yen, using in each period the ratio of announced weights. The empirical results for Israel are reported in Table 2.b, which is analogous to Table l.b for Chile. It is apparent from the table that the exchange rate system can be unambiguously verified by our procedure only in the third sub-period (labeled 2.2 in the table), when the announced weights cannot be rejected by Test 1 and zero weights are clearly rejected by Test 2 - particularly when using the restricted model estimates. In the understoodas the sharein the totalcostof the basket- can changedailyaccordingto changesin crossrates (seeAppendix2 formore details). 29 AppendixTableA.2 providesmore detailson the developments of exchangerate policy in Israelsince 1986. 30 The correlationbetween these two currencies exceeds .98 in some of the sub-periodsunder consideration. 27 other sub-periods, the unrestricted estimates wander off very far from the announced values, and lead to rejection of both null hypotheses in the majority of cases, even though their precision is extremely poor. In turn, the restricted estimates are much more precise, and generally closer to the announced weights. In general, they lead to rejection of the null of zero weights for sufficientlylarge samples in all sub-periods, but tend also to lead eventually to rejection of the announcedweights except in period 2.2. Like in the case of Chile, the scatter plots presented in Figure 4 help understand these empirical results. Period 2.2 is the only one in which the observed exchange rate behaved in a fashion similar to the central parity. During this period, which coincides with the introduction of a crawl in the path of the central parity, the exchange rate hovered around the midpoint of the band, and the boundaries were never reached. In contrast, during periods I and 2.1 the frequent level adjustments to the band already mentioned are reflected in the disconnected scatter plots of Figure 4. From the perspective of verifiability, these jumps make identification of the basket weights more difficult. Finally, in the wider-band periods 3-6, the scatter plots are more reminiscent of those correspondingto the multiple-currencyperiods of the Chilean band: they show little correspondencebetween the central parity and the observed exchange rate. In summary, one interpretation for the poor verifiability results in the Israeli case probably lies in the additional complexity induced by the presence of five mutually correlated currencies in the central basket. Even after reducing to three the number of regressors, identification is still poor. The sharp discontinuous changes in the central parity in the earlier periods, and the augmented band width in the later ones, are also likely obstacles to the verification of the regime. 28 III.c Is the Test Informative? We conclude this section with a reassessmentof the robustnessof our findings for the cases in which we achievedverification (periods 1-2 in Chile and 2.2 in Israel). We do this by constructing a randomly generatedvariableand using it to replacethe observed exchange rate as dependent variable in the empirical estimation and testing procedures performed earlier. The results are reported in Table 3, from which it is apparent that Test 1 rejects the announcedweights in most cases, and Test 2 fails to reject the zero weights in all the cases. This suggeststhat problemswith test power are not behind the success in verifying the exchange regime in these episodes. As a final exercise to reassess the robustness of our findings, we replace the Chilean peso and the Israeli shekel with the Swiss franc. We choose the Swiss franc because Switzerland had a floating regime during the periods of interest. Table 3 shows that we reject the null hypothesisthat the weights are equal to the announcedweights. As reported before, for the same periods, the estimations with the Chilean peso and the Israeli shekel fail to reject the announced weights. Therefore, one can conclude that periods 1-2 in Chile and period 2.2 in Israel are verifiable. Table 3 also shows that the Swiss franc rejects the null hypothesis that the weights are equal to zero. This rejection does not mean that the Swiss franc is not freely floating during the periods under consideration. Exchange rates are correlated for other reasons than government intervention. This tends to yield rejections of zero correlation. The next section of the paper exploreswhether it is possible to fail to reject free floating using the methodologyapplied for band regimes. 29 IV. Verifying Free-FloatingRegimes We now turn to the verification of free-floating regimes. The concept of verifiability has a different meaning under floating regimes. Under these regimes, governments do not make a commitment to a nominal anchor. There are no exchange rate rules to be verified, except that the exchange rate is floating or that the government is not interveningin the market. Applying the methodology used for exchange rate bands, market participants can check if the exchange rate is uncorrelatedwith major exchange rates. A rejection of no correlation is not a rejection of a free-floating regime. Using observed exchange rate data, it is generallydifficult to fail to reject that weights are equal to zero, either because governments intervene or because exchange rates co-move. Rejecting zero weights is not necessarily a sign of intervention. However, failing to reject zero weights is a clear sign of no intervention (or no pegging to other currencies). We rejected zero weights in the case of the Swiss franc above. Now, we move to the case of emerging markets. Before proceeding with the estirnations,note that there are other alternative ways of verifying free-floatingregimes. Market participants can essentially check every month whether the central bank has intervened by seeing whether its reserve holdings have changed. Also, banks usually know who is participating in the market, so they can tell the difference between a system where the central bank never intervenes and where it intervenes occasionally. These methods require some type of additional information, 30 beyond observed exchange rates. In this paper, for ease of comparison with the previous section, we stickto verifiabilityjust using exchange rate data. For the verification of floating regimes, we focus on Brazil, Mexico, Peru, South Korea, and Thailand during specific periods in the 1990s.These countries provide a good opportunity to compare periods of intervention with periods of free-floating. Brazil, Mexico, South Korea, and Thailand suffered exchange rate crises in the 1990s, which forced them to abandon previous exchange rate arrangements and adopt systems officially described as free-floatingby their respective authorities. Therefore, for these countries we perform the same statistical tests for periods labeled as free floating and for periods during which other regimes were in operation - namely periods of managed floating, bands, crawlingpegs, and basket pegs. Peru, on the other hand, is an interesting case because despite declaring a free-floatingregime for the entire decade, several papers have noted that the Peruvian exchangerate has remained surprisinglysteady. (See Calvo and Reinhart, 2000, Hausmann, Panizza, and Stein, 2000, and Edwards and Savastano, 1999, for characterizationsof floating regimes.) As in other cases, the observed pattern calls into questionwhetherthe governmentis in truth followingthe regime that it says. Following the methodology used in the previous section, we estimate our basic equation (1) for each of these countries over the periods noted in Table 4 and dictatedby data availability over the 1990s. We allow for a constant rate of crawl, include as regressors the five major currencies (US dollar, yen, DM, British pound and French franc),and as before estimate the model by OLS in first differences.3 ' As we are only testing the hypothesis of all the weights being equal to zero (a standard goodness of fit test), we are not concernedby the potential multicollinearityproblem. 31 31 In free-floatingregimes, we expect to fail to find any peg of the exchange rate visa-vis foreign exchange rates, so we would expect Test 2 not to reject the null of zero weights. Table 4 reports the percentage of observationsfor which the test does reject the null hypothesisthat weights are equal to zero. If the exchangerate is in fact free-floating, one would expect to find low values in the table, meaning that we only find a relationship between the local currency and strong currencies in very rare occasions. On the other hand, when the central bank follows a specific rule relative to one or several strong currencies, we expect to find large values in the tables (mostly rejections of the hypothesisthat weights are equal to zero). For ease of comparison,the shaded areas in Table 4 correspondto periods labeled as free-floating. The results displayed in the table show that in the case of pegs, bands, and managed floating regimes we reject in almost every sample the null hypothesis that the weights are equal to zero. The only exceptionis the case of Brazil, during part of her period of managed floating. On the other hand, in the episodes declared as free-floating we generallyfail to reject the same hypothesis. There are two exceptions, however: Peru and most of the post-Tequilaperiod in Mexico. The samples used for the tests reported in Table 4 start at the beginning of the year except when a specific date is known for the transition to floating. As an alternative approach, Figure 5 shows the rejection percentage (right-hand scale) of the zero-weights hypothesis in rolling samples of 100 observationsduring the 1990s, together with the exchange rate vis-A-visthe US$ (in the left-hand scale). The pattern is similar to that shown in Table 4. It is possible to see how rejection rates fall dramaticallyright after a major devaluationin the three countriesaffected by the 32 late 1990scrisis: Brazil, Korea, and Thailand.In the last two cases, we observe a similar pattern in the sense that after approximately one year has elapsed since the large devaluation,rejection rates appear to rise again. In the Mexican case, rejection rates fall only during short periods after the late 1994 crisis. Much interest has been devoted to the Mexican free-floating regime that followed the Tequila episode of 1994, particularly during the stable period starting in late 1995. Edwards and Savastano(1998) found that the volatility of the exchange during 1996 was not smaller than that of other currencies widely considered as free-floating,but there seemed to be some form of feedback rule from the exchangerate to monetarypolicy. Finally, it is also clear from the figure that periods of marked stability of the exchangerate are matchedby rejectionsof the zero-weightshypothesis. Examples of this are the Brazilian band period (1995-1998), the Korean, Mexican, and Thai pre-crisis periods, and the Peruvian free-floatingregime of the 1990s. V. Summaryand ConcludingRemarks The new conventionalwisdom is that intermediateexchange rate regimes, such as baskets, crawls, and bands, are no longer viable. Accordingto this proposition, countries are being pushed to the "corners," the extremes of either free floating or firm fixing. We have argued that a theoretical rationale for this proposition is currently lacking; none of the candidatesoffered- the impossibletrinity, the dangers of unhedged foreignliabilities, or governmentreluctance to abandon ship in time - is quite up to the job. We offered such a rationale, by introducing the notion of verifiability, By verifiabilitywe mean the ability of a market participant to infer statisticallyfrom observed data that the exchange 33 rate regime announcedby the authority is in fact in operation. Verifiabilityis an instance of transparency,a means to credibility. Our point is that a simple regime such as a clear dollar peg, or even a free float, may be more verifiable by market participants than a complicatedintermediateregime. In this paper we have made a first attempt at assessingempiricallythe verifiability of various exchange rate regimes. We first focused on the verification of exchange rate bands, drawing from the experiences of Chile and Israel. In the case of Chile, when the band was relatively narrow and the peg involved only the dollar, verification is relatively easy to achieve. But from 1992 to 1999, when the band became wider and the peg involved additional currencies, our simple statistical procedures fail to achieve verification. In the case of Israel, whose basket involved five currencies, two of which were very strongly correlated, we only achieve verification in a period of relatively narrow band in which the centralparity does not experience sharp realignments, and only when using a restricted specificationinvolving a reduced number of currencies. In widerband periods, and in narrow-band periods with frequent realignments, our procedures again fail to achieve verificationof the regime. This is precisely the result we expected. On the whole, the results suggestthat higher bandwidth,as well as the adoption of multiple instead of simple basket pegs, and frequent parity realignments, all make more difficultthe econometricverificationof the announcedregime. The finding that Chile and Israel fail to reject the announced weights for some particular periods seems to be an informative test. For the same time periods, we reject those weights when we replaced the peso and shekel by a randomly generated variable 34 and by the Swiss franc. This means that for narrow bands we are able to verify the announcedexchange rate regime. We also examined the verifiability of regimes self-declared as free floating in several Latin American and East Asian countries in the 1990s - Brazil, Mexico, Peru, South Korea, and Thailand. Even though there are different ways to verify floating regimes, we followed the same methodology used for bands. We tested whether supposedlyfloating exchange rates are correlated with major exchange rates. Our tests do not show significantevidence against the hypothesis that the exchange rates of these countries are indeed floating, with the exception of Peru and part of the post-Tequila period in Mexico. In these cases, we find some evidencethat the exchange rate is in fact moving along with some weighted combination of strong currencies. This appears to agree with the conclusions reached by other researchers. In sum, we fail to reject freefloating regimes, although cross-currency correlations tend to reject free-floating- even when governments do not intervene. Whether these findings can be extended for long periods of free floating, after the high volatility in the aftermath of crises has vanished,is a questionfor future research. The analysis in the main text was complementedby means of Monte Carlo tests reported in Appendix 1 assessing the effects of bandwidth and number of currencies in the basket on the time needed to verify exchange rate bands. On the whole, the results agree with the above findings. As expected, when the range of variability of the exchangerate is relatively large, the number of observationsneeded to verify the regime increases considerablywith the width of the band. The number of observationsneeded to differentiatethe crawling basket from a random variablein at least half of the samplesis 35 under 100 days when the band width is 2%, as it was for Chile from 1985 to 1987, but is over 500 days when the band width is 10%, as it was for Chile from 1992 to 1998. Regarding the role of the number of currencies in the basket, we find that moving from a single-currency parity to a 3-currency basket increases the amount of data needed to distinguish the basket from a random currency by an extra year's worth of observations (assuming a 10% band, and again using the criterion of finding statistically significant weights at least half the time). If we are right that it is hard for a central bank to establish credibility for its proclaimed monetary regime without verifiability, then our results confirm that complicated combinations of baskets, crawls, and bands, are less likely to satisfy skeptical investors than are simpler regimes. We thus offer a possible and much-needed rationale for the hypothesis of the vanishing intermediateexchange rate regime. 36 References Benassy-Quere, Agnes, 1999, "Exchange Rate Regimes and Policies: An Empirical Analysis," in Exchange Rate Policies in EmergingAsian Countries, edited by Stefan Collignon,Jean Pisani-Ferry,and Yung Chul Park (Routledge:London), 40-64. Calvo, Guillermo and Carmen Reinhart, 2000, "Fear of Floating," mimeo, University of Maryland. Council on Foreign Relations, 1999, Safeguarding Prosperity in a Global Financial System: The Future International Financial Architecture, published by Institute for InternationalEconomics, Washington,DC, 1999. Crockett, Andrew, 1994, "Monetary Policy Implications of Increased Capital Flows," Changing Capital Markets:Implications for Monetary Policy. Symposium sponsored by Federal ReserveBank of KansasCity, Jackson Hole, August 1993. The Economist,"Global Finance: Time for a Redesign?"January 30, 1999,p. 1-18. Edwards, Sebastian and Miguel Savastano, 1998, "The Morning After: The Mexican Peso in the Aftermath of the 1994Currency Crisis" NBER Working Paper No. 4661. April. Edwards, Sebastian and Miguel Savastano, 1999, "Exchange Rates in Emerging Economies: What Do We Know? What Do We Need to Know?" NBER Working Paper No. 7228, July. Eichengreen,Barry, 1994, International Monetary Arrangements for the 21st Centurv, BrookingsInstitution, WashingtonDC. Eichengreen,Barry, 1999, Toward a New Financial Architecture: A Practical Post-Asia Agenda, Institute for InternationalEconomics,Washington,DC. Frankel, Jeffrey, 1993, "Is Japan Creating a Yen Bloc in East Asia and the Pacific?" in Regionalism and Rivalry: Japan and the U.S. in Pacific Asia, edited by Jeffrey Frankel and Miles Kahler, Universityof ChicagoPress, Chicago, 1993, 53-85. Frankel, Jeffrey, 1999, "No Single Exchange Rate Regime is Right for All Countriesor at All Times," Graham Lecture,Princeton UniversityPress. NBER Working Paper No. 7338. Frankel, Jeffrey, Sergio Sclmukler, and Luis Serven, 2000, "Verifiability and the Vanishing IntermediateExchangeRate Regime," workingpaper. Frankel, Jeffrey, and Shang-Jin Wei, 1994, "Yen Bloc or Dollar Bloc? Exchange Rate Policies of the East Asian Economies" in Macroeconomic Linkages: Savings. Exchange Rates, and Capital Flows, NBER - East Asia Seminar on Economics, Volume 3, Takatoshi Ito and Anne Krueger, editors, University of Chicago Press, 1994. Frankel, Jeffrey, and Shang-Jin Wei, 1995, "Emerging Currency Blocs," in The International Monetary System: Its Institutions and its Future, edited by Hans Genberg, Springer, Berlin, 1995, 111-143. Hausmann,Ricardo, Ugo Panizza, and Emesto Stein, 2000, "Why Do CountriesFloat the Way they Float?" mineo, Inter-AmericanDevelopmentBank. Krugman, Paul, 1991, "TargetZones and Exchange Rate Dynamics," Quarterly Journal of EconomicsCVI, 669-682. Lanrin, Felipe, and Andres Velasco, 1999, "Exchange Rate Policy for Emerging Markets: One Size Does Not Fit All," July, forthcoming, Essays in International Finan , PrincetonUniversity Press. 37 Minton-Beddoes,Zanny, 1999, "From EMU to AMU? The Case for Regional Currency Blocs," Foreign Affairs. Obstfeld, Maurice, and Kenneth Rogoff, 1995, "The Mirage of Fixed Exchange Rates," NBER Working Paper No. 5191. Stein, Emesto and Jorge Streb, 1998, "Political Stabilization Cycles in High Inflation Economies,"Journal of Development Economics, 56, 159-180,June. Stein, Ernesto and Jorge Streb, 1999, "Elections and the Timing of Devaluations," working paper, InterAmericanDevelopmentBank. Summers, Lawrence, 1999a, testimony before the Senate Foreign Relations Subcommittee on International Economic Policy and Export/Trade Promotion, January 27. Summers, Lawrence, 1999b "Building an International Financial Architecture for the 21 st Century," Cato Journal, 18, no. 3, 321-330. 38 Appendix 1: Monte Carlo Simulations We turn now to the Monte Carlo simulations, which offer a more general testing ground for verifiability of intermediate regimes. For our experiments,we generate 1,000 samples according to the simple model described by equation (1), using for the baskets actual data on the exchange rates of the major currencies (valued in terms of the GDPweighted numeraire). We use daily data between February 1986 and September 1999. The parameters of the data-generating process are c (level of exchange rate), d (yearly rate of crawl), w,...w3 (weights on US$, DM, and JY), a (standard deviation of the error term), and to (initial observation). We use a log linear version of equation (1). The log error term is generated as i.i.d. normal with mean zero. Based on this basic framework, we study the effect of different model specifications on the amount of time to reject our proposed null hypotheses. For each sample, we calculate the number of observations necessaryto obtain 10 rejections of the null hypothesis that both the weights and the rate of crawl are zero (Test A) and the null hypothesis that the weights are zero (Test B). Role of Band Size Clearly, it should be harder to verify a basket regime with a wide band than one with a narrow band, and harder to verify a basket regime with a loosely managed float (i.e., a small tendency to intervene when the exchange rate drifts from the parity) than another with a tightly managed float (a strong tendency to intervene). To verify the role of band size in determining tne amount of information needed to reject the proposed null hypotheses, we generate sets of 1,000 samples. Each set has a different standard deviation of the underlying disturbance (a), representing different band sizes. 39 For this exercise, we generate the samples using a level parameter equal to 1, a rate of crawl of 1% per year, and equal weights for all major currencies, and starting from observation 1 (2/24/1986). We let the standard deviation cavary from 1% to 10%. In this regard, recall that 2% was the width of Chile's band from mid-1985 to 1987, and 10% was the width of the band during the period 1992-97. For purposes of comparison, 21/4% was the width of the ERM target zone followed by many European countries up until 1992 (and still followed today by Denmark), 6% is the width of the ERM target zone followed by Italy and the United Kingdom up to 1992, and 15% is the width of the ERM zone for France and others from 1992 until the beginning of EMU in January 1999. The results appear in Appendix Figure 1. The graphs plot the quantiles of Test A and Test 2 against the standard error (a) used to generate the samples. Each line corresponds to one quantile, and depicts the number of observations needed to achieve rejection of the null hypothesis (at the 5% level) in x% of the 1,000 samples-where x is the quantile in question. As expected, the graphs show that, for both tests, the number of observations needed to reject the null of zero weights and rate of crawl in any given percentage of the samples rises steadily with (. This is reflected by the fact that the lines correspondingto the various quantiles have positive slopes. In other words, wider bands make it more difficult for investors to reject specific hypotheses conceming the weights of the central parity-they need more time to get an accurate assessment of the parameter values. And the additional time needed is not negligible. For Test B, for example, the number of observationsneeded to reject the null in 50% of the samples ranges from under 100 days 40 for an (old-) EMU-sized band (2% width) to over 500 for a Chilean-sized one (10% width). Role of Number of Currencies in Basket Intuitively, the larger the number of unknown parameters that need to be estimated, the harder it should be to verify that the data match the announced policy regime. This applies not only to the number of currencies in the basket, but also to the presence of a non-zerorate of crawl. To verify this assertion, we next examine the impact of different basket sizes on the amount of informationneeded to reject the nulls underlying Tests A and B. For this purpose, different numbers of currencies were included in the Data Generating Process (DGP). We construct a simple peg (the US dollar), a two-currencybasket (the US dollar and the Deutschemark), and a three-currencybasket (the dollar, the Deutsche mark, and the Japanese yen). In each basket the currencies are equally weighted. The other assumptions are like in the previous exercise. The results are portrayed in Appendix Figure 2. To avoid cluttering the pictures, only the medians of Test A and Test B (defined as before) are presented. They are plotted against alternative values of the standard deviation of the random disturbance assumedin the simulation. As expected, increasing the number of currencies in the basket shifts the quantile lines upward, reflecting the fact that for any given value of the standard deviation more observationsbecome necessary to reject the null hypotheses. As before, the increase in information requirements is sometimes substantial. For example, with a bandwidth of 41 10% (as observed in Chile in recent times), moving from a single to a 3-currency basket raises the 50% quantile of Test B by over 200 observations-implying that an extra year of data becomes necessaryto reject the null hypothesis. Role of Rate of Crawl What about the rate of crawl? Intuitively,its value should have little consequence for Test B, which is concerned only with the basket weights. However, for Test A it can make a big difference-rates of crawl further away from zero must help reject the null hypothesismore quickly, since the latter involvesa zero rate of crawl. This is verified in Appendix Figure 3, which shows the effects of different rates of crawl on the verificationtime, as reflected by the 50% quantile of Test A and Test B. For a given value of a, we generate different samples assuming increasing rates of crawl. As expected, the time to reject Test A (measured by the left scale) declines steadily as the rate of crawl rises away from zero, while the time to reject Test B (measured by the right scale) shows only modest variation. Role of Period The power of these tests depends on the precision of the parameter estimates, itself given by the noise-to-signal ratio-or the relative size of the variances of the dependent and independent variables. When the variance of the dependent variable is large relative to the variance of the independentvariable, the estimates are imprecise and it is difficult to reject a given null hypothesis. Since these relative variances are not 42 constant over time, the verifiabilityof a given model may depend on the specific time period over which it is observed. This can be assessedusing data from differenttime periods to carry out the Test A and B. Since our experiments use actual data on the hard currencies, any differences in time to reject Test A and B across replications,using hard-currency data from different time periods, should be attributed to changes over time in the variance-covariancematrix of the hard currencies. The results of such an experiment are reported in Appendix Figure 4, which shows the median values of the time to reject Test A and B, obtained when the simulations use hard-currency data from different periods in 1986-96 and assuming a three-currencybasket with equal weights. To facilitatethe interpretationof the results, we also show in the figure a measure of the variance of the hard currencies-specifically, the inverse of the average of their standard deviations. As the graph shows, variability of the hard-currencyexchange rates was particularly high in the first and fourth periods considered. This results in a clear reduction in time to reject Test A and B in such periods, relative to the rest. 43 Appendix2: Constructionof Numeraire and EstimatedModels In this appendix,we describe how we constructedthe weighted basket numeraire and the precise models we estimated in the case studies of Chile, Israel, and the floating regimes. Constructionof the WeightedBasket Numeraire The numerairewas constructedusing the bilateral exchange rates of seven strong currencies weightedby the GDP share in 1992. The specificunits of each currency in the basket were chosen so that the basket is valued in 1 US dollar on January 2, 1990. The value, in US$, of the weighted basket (WB) at a given point in time is: VBt=aI+a2DMt+a3BPt+a4FFt+a5JYt+a6CD,+a7 II (Al) such that all the exchange rates are expressed in US$ over local currency. IL stands for the Italian lira and CD for the Canadian dollar. Using 1991 GDP at market prices (constant 1995 US$) data from the World DevelopmentIndicatorsreport, we defined the followingweights: Currency US$ DM BP FF JY CD IL Weight 13.01% 5.79% 8.34% 28.25% 2.97% 5.92% 35.72% These weights represent the share of the cost of each currency in the total value of the basket at the reference date (in this case 1/2/1990).Based on this definition, we can calculatethe units of each currency (a, ... a7 ): w= a, / WBo a a, = w *WBo W2= a2 DMJ/ WBO - a2 = (A2) W 2 * WBo / DMo 44 W3= a3 BPo/WBo - a3 = W3* WBo/ BPo The resulting units are the following: Currency US$ DM BP FF JY CD IL Units (a,) 0.3572 0.2192 0.03566 0.4803 40.707 0.0499 91.245 Using these units and equation (Al), we obtained the value of the weighted basket at any point in time. In order to obtain any exchange rate as a function of this numeraire, we just multiply the exchange rate of the local currency in terms of the US$ by WBt. Estimation of Basket Weights in Case Studies of Chile, Israel andfree-floating regimes In the two case studies undertaken in this paper, the baskets were, in fact, constructed in a similar way to our weighted basket numeraire. When the basket is defined for the first time, some strong currencies are selected. Initial weights are calculated according to trade weighs. The units of each currency that are used for the calculation of the basket from that moment on are defined according to the procedure described above. The units remain constant over time, but the actual weights of each currency depend on the bilateral exchange rates movements. In order to complete the definition of the exchange rate regime, a path must be defined for the value of the basket (B1).In some cases, this value is to remain constant,to increase at a constant rate or to vary with internal or external inflation rates. The local exchange rate, in the case of a basket peg, is determined by the equality of the 45 predetermined path for the value of the basket (A) and the actual value of the basket, giventhe units chosen, the bilateral foreignexchange rates and the local rate: + b3 St*BPt+ b4 St*FFt+ b5 St*JYt, Bt = b1 St + b2 St*DMA where St represents the local currency (in this example the Israeli Shekel vis-a-vis the US$). Using a formula analogousto (A2), we can expressthe previous equation in terms of the original weights: W5 Bj/Bo=w, St/SO+w2 (StDMK)/(SODMO)+W 3 (St*BPt)/( SOBPO)+W 4 (St*FFt)/( SOFFo)+ (St*JYt)/(SoJYo). Rewriting the previous expressionwith the local currencyon the LHS, we have: SOSt= wl BJBt + w2 BWBt*DMt/DMO + w3 B&Bt*BPt/BPo + W4 BO/Bt*FFt/FFo + w5 B3Bt*JYt/JYO. Finally, multiplyingboth sides of the previous equation by WB/WBt we obtain an equation where all the exchange rates are expressed in termnsof the numeraire. Redefiningvariables, we obtainedthe followingequation: Y, = w, XUSt + w2 XDMK+ W3 XBPt+ W4 XFFt + W5 XJYt. (A3) In the case of basket bands, the actual value of the basket is allowed to fluctuate around the predeterminedpath, usually with a given percentage above or below (the band width). In those cases, we refer to the reference path as central parity. Equivalently,the band defined for the basket implies an analogous band for the local exchange rate vis-avis the numeraire. In our analysis, we try to recover the original announced weights from the observed exchange rate, the bilateral exchange rates between the strong currencies and the predeterminedpath for the central parity. The movements of the observed exchange 46 rate inside the band give rise to an error terrn. A stationarity assumptionis certain to fail in a time series for the level of the exchange rate. A simple way to deal with this is to work with first differences. The basic equation we estimate in this paper, expressed in first differences,is the following: AY,= do+ w, AXUSt+ W2AXDM+ W3AXBPt+ W4AXFFt+ w5AXJYt+ et. (A3) For the Chilean case, in the first three periods we included only the US$ and in the followingfour periods, the US$, the DM, and the JY. As described in the next section, we finally used a restrictedversion of equation (A3) for the case studies of Chile and Israel but in the case of the free-floating countries, we estimated equation (A3) without worrying about the multicollinearityproblem. Dealing with MulticoUinearity As mentioned in the text, strong correlation between the included regressors (particularly between AXDMt and AXFFt and between AXUSt and AXJY, in some periods) gave rise to a significant multicollinearityproblem. In order to deal with it, we combinedpairs of regressors,using the ratio of announcedweights: AYt = do + wI (AXUSt + W50 / WIO AXJYt) + w 2 W3 AXBPt+ et (AXDMt + W4 0 / W20 AXFFt)+ (A5) where w1 O, w20 , w40 and w50 represent the announced weights (which are known constants). With this specification, we were able to identify only the following parameters:do,wI, w2 and W3. 47 Figure 1: Chilean Exchange Rate and Exchange Rate Band Chilean Peso Relative to Weighted Basket - 1986-1999 1051 Period band width =2% 95 Period 2 band width=3% Period 3 band width=5% 85 75 I I 65 55 I 45 45 ) I 1 Period 5I Period 6 1 band band width width Period t t 4 |band 2 width =10% 25 T 0 __ '0 '0 N N N I0,0 0 00 a,0 I' 0- °,, =10% T Period 7 band width =12.5% =10% 'c0 N ,0 N 0 0b O i Figure 2: Israeli Exchange Rate and Exchange Rate Band 1989 - 1999 - New Israeli Shequel Relative to Currency Basket 5.95 4.95 Period 2.2 With Crawl Band width=5% Period 2.1 No Crawl Band width=5% Period I band witdth =3% basket peg 5.45 4.45 3.95 3.45 2.95 2.45 Period S expanding band width -15% Period 3 Period 4 band band width width =7% - 7% z r 2.45 1.95 1.45 oo o? O? O? 1 M O? fiO a, C0 ~~~~~~~~~~~~~~~~~~i AD oo I Q~~~~~~J Cf ON a,i 3N a',N ON :tQnt F ooo£o af I i I bo I U) I~ I I U m g I I UQt ONI I U a U Figure 3: Chilean Peso and Central Parity - Scatter Plots (Chilean Peso / Weighted Basket) 280 550 260 240 500 200 720 4500 ~200 45 ~~~~~~~~~~~~U ~~~~~~~~~400- C.) 180x4 1600 Period I 140 Period 4 350 .3.50. 140 160 180 200 clkm 220 240 260 280 3 0 400 Peso 450 550 500 hdkmnPeso 270 520 500 260.P 480 1830 /=8 440 - 240 230 PUid240-PTo 230 240 250 260 270 420 440 460 480 500 520 CNueanPeso 450. 400 520 e -500- ~350 - 5.~~~~~~~~~~~~~Z 480, 5300 5 250 460 440- ~~~~~ ~ 2DO 200 250 300 Period 3 350 400 40Period 450 420 6 440 460 Csiean PesoCh] 48'0 500 52'0 Pa 500 460 -4 440 420 ~ ~~~~~~~Period 7 40 400 420 440 46'0 Cames Peso 480 500 Figure 4: Israeli Shekel and Central Parity - Scatter Plots (Israeli Shekel / Weighted Basket) 2.05 3.8 2.00/ 3.7 1.95 3.6 f g 1.90 j<f/ u 1.85 °g 3.5 u 1.80 3.4 3.3 Peiod I 1.75 1.75 1.80 1.85 1.90 1.95 2.00 Indi Shdcd Period3 2.05 3.2 3.2 2.6 4.04 2.4. 3.8. 3.3 3.4 3.5 I&WiSh*d 3.6 3.7 3.8 A~~~~~ 2.2Pcid2. 1.8 cro 2.0 2.2 2.4 2.6 3.2 3.4 Lsndi Shdckd 2.25 3.8 4.0 3.6 ~~~~~~~~~~~~~~~~~5.0 4.0 3.5 2.0 2.0 3.6 Isqdi Sh&dd 2.5 ~~~~~~~~~~~~~4.5.Period 5 Period 2.2 3.0 IsJi Shd&d 2.5~~~~ 3.5 4.0 3.0 3.0 3.5 4.0 Isndi Shdcd 5.0 3.5. 4.8 2.0 4.5 / ~~~~~~~~~~~~~~3.0 4.4 4 < + / X 3.8d 2.2 3.0 Period6 4.0 4.2 4.4 1rIlai Shkd 4.6 4.8 5.0 Figure 5: Exchange Rates Against US Dollar and Percentages of Rejections Rejection Rate Corresponds to Testing HO:Weights=O Model Tested Is First Differences of Domestic Currency on Major World Currencies (Exchange Rates expressed as Domestic Currency / US$ ) S. Korean Won Brazilian Real 1.8 0.9 1.8 S.g 0 1.4 0.7 1.2 0610 z~~~~~~~~~eet 0 2 Rejection Rate(ih Scale) . n Rat (Rgh Scale) e Ra0.0 1eruvlan 0.8 0.4 10 0.8 0.303 0.: 0.2 00 . Exchange Rate 1 0I0.1 0 Thai Baht Mexican Pes 12 1 80 0.8 Rejection Rate(RghtScale) 0 ~~~~~~~~~~~~~~ 0 0.8 0s 40 0.T 0.6 0.8 30 0.0 0.j s 8 4-04 2004 0.3 - -nca-nge,Rate 0.3 0.202 2 -10 0.1 0 0 A Peruvian Sol 41 0.8 3. 0.7 2.5 2 0.8 EcagRea0.8 1.6 0O4 0.3 0.2 0.0 0 RejectionRate(RightScale) 0.1 0 0.1 0 0 Table l.a: Chilean Exchange Rate Description of Exchange Rate Regimes Begin Period End Number of Band Observations Width (+/-) 434 2% Weights of Central Parity U.S. Deutsche Japanese Dollar Mark Yen 100% 0% 0% 1 February24, 1986 January 4, 1988 2 January 5, 1988 June 5, 1989 340 3% 100% 0% 0% 3 June 6, 1989 April 2, 1991 449 5% 100% 0% 0% 4 July 1, 1992 October 31,1994 580 10% 50% 30% 20% 5 November 30, 1994 November 30, 1995 236 10% 45% 30% 25% 6 December 1, 1995 January 20, 1997 264 10% 45% 30% 25% 7 January 21, 1997 June 24, 1998 326 12.5% 80% 15% 5% Onlyperiodswithat least250 observationsare listed. Duringtheseperiodstherewereno changesin the exchange rate regime. The bands' width,the weightsof the centralparity, and the level of the centralparity were held constant. Theperiodsexcludedincludediscretedevaluations/revaluationsof the centralparity. For more details about the exchangerate regimesin Chile,see Appendixtable. The announcedweightscorrespondto the relative importanceof the respectivecurrencyin the firstday whenany new weightis defined.Withrelativemovements betweenthe foreigncurrencies,those weightsvary with time.The estimationprocedure,however,is designedto estimatethe initialweight. Table 1.b: Chilean Exchange Rate Percentage of Observationsfor Which Null Hypothesis Is Rejected (1%) Period Obs OLS First Differences OLS First Differences Unrestricted Model Xiw-wol Restricted Model Test 1 Test 2 Ho: Ho: Point Estimate Test 1 Test 2 Ho: Ho: Point Estimate Weights Weights Wus$ (s.e.) Weights weights Wuss (s.e.) = =0 ~~flflOUflC ~flflOUflCannouncMoe OLS OLS First First Differences Differences = = 50 100 200 0 0 0 100 100 100 0.94 (0.06) 0.90 (0.07) 0.92 (0.05) 0.06 0.10 0.08 50 100 200 0 0 0 100 100 100 1.27 (0.19) 1.09(0.13) 1.00 (0.07) 0.27 0.09 0.00 50 100 200 21 22 :34 92 97 98 0.80 (0.09) 0.85 (0.05) 0.90 (0.07) 0.20 0.15 0.10 4 50 basket 100 width=10% 200 100 100 100 100 100 100 1.10(0.15) 1.10 (0.09) 1.04 (0.06) 100 100 100 100 100 100 1.08 (0.15) 1.09 (0.08) 1.04 (0.06) 0.90 1.08 1.06 0.87 0.90 0.87 68 86 94 68 86 94 1.01 (0.55) 1.67 (0.28) 1.19 (0.22) 68 86 94 79 91 96 1.38 (0.26) 1.02 (0.09) 1.07 (0.08) 1.37 1.69 1.19 1.07 0.95 0.97 50 100 200 58 82 91 50 78 90 0.37 (0.44) 0.81 (0.25) 1.02 (0.17) 45 76 89 45 76 89 0.95 (0.21) 1.10 (0.12) 1.12 (0.08) 1.44 1.31 1.27 0.86 0.98 1.00 50 100 200 13 63 82 100 100 100 1.08 (0.14) 1.14(0.11) 0.83 (0.16) 26 68 85 100 100 100 0.97 (0.07) 1.03 (0.04) 0.95 (0.07) 0.38 0.44 0.38 0.25 0.33 0.31 announc 1 US$ band width=2% 0 Precision Unrestricted Model Restricted Mdl Model WUS$1 2 US$ band width=3% Wuss=l 3 US$ band width=5% WUSS=1_ Wus==0.5 5 50 basket 100 width=10% 200 Wuss=O.45 6 basket width=10% Wus$=0.45 7 basket width=12.5% Wus$=0.8 In periods 1-3, only the US$ was considered in the estimation. Precision is calculated as the sum of absolutes deviations of the estimated weights at 50, 100 and 200, with respect to the announced weights. In the restricted model, for periods 4 to 7, the JY and the US$ were combined in one variable, using the relative announcedweights. See Appendix 2 for details. Table 2.a: Israeli Exchange Rate Description of Exchange Rate Regimes Period Begin End 1 January3, 1989* 2.1 March 1, 1990 2.2 December 17, 1991 3 May31, 1995 4 April30, 1996 5 June 18, 1997 February 28, 1990 December 16, 1991 May30, 1995 April 29, 1996 June 17, 1997 December 31, 1998 Number Band of Width Observations (+/-) 291 3% 443 851 222 U.S. Dollar 60% 5% 60% no crawl 5% 60% crawling 7% 54.8% Weights of Central Parity Deutsche Japanese French Mark Yen Franc 20% 5% 5% 10% 20% 5% 5% 10% 20% 5% 5% 10% 24.2% 1% 5.6% 8.3% 273 7% 60.3% 21% 5.6% 5.1% 8% 374 15%** 60.3% 21% 5.6% 5.1% 8% The basket was introducedwith the presented weights in August, 1986, but the exchange rate was allowed to vary around a 3% band in January 1989. ** Widening band designed to reach 15% by end of 1997. * British Pound Table 2.b: Israeli Exchange Rate Percentage of Observationsfor Which Null Hypothesis Is Rejected (1%) Period Obs OLS First Differences OLS First Differences Unrestricted Model Test 1 Ho: Test 2 Ho: Precision £|w-woI Restricted Model Point Estimate Test 1 Test 2 Ho: Ho: Point Estimate Weights Weights Wuss Weights weights = aniounc =°0 (s.e.) =0 Wuss (s.e.) Unrestricted Restricted 50 100 200 29 69 86 92 97 98 -1.37 (0.40) -2.52 (0.25) 0.43 (0.02) 0 0 44 89 95 98 0.64 (0.10) 0.45 (0.10) 0.43 (0.02) 3.94 6.19 0.48 0.06 0.29 0.27 50 100 200 100 100 100 100 100 100 -5.10 (0.24) -5.30 (0.15) 0.15 (0.07) 0 0 46 0 7 40 0.31 (0.34) -0.12 (0.25) 0.09 (0.06) 11.49 11.98 1.33 0.79 0.91 0.68 50 100 200 0 0 2 89 95 98 -0.54 (0.36) 0.01 (0.31) 0.22 (0.13) 0 0 0 97 99 99 0.53 (0.07) 0.57 (0.08) 0.55 (0.05) 2.42 1.55 0.80 0.14 0.11 0.07 3 50 basket 100 width=7% 200 63 84 93 100 100 100 -1.28 (0.44) -0.62 (0.30) -1.52 (0.21) 39 67 85 100 100 100 0.60 (0.15) 0.84 (0.08) 0.73 (0.08) 5.23 3.89 6.37 0.29 0.47 0.36 100 100 100 100 100 100 -3.59 (0.45) -2.89 (0.32) -2.97 (0.25) 0 55 76 0 55 79 0.53 (0.25) 0.55 (0.17) 0.47 (0.12) 11.35 9.54 9.30 0.82 0.55 0.40 100 100 100 100 100 100 -5.54 (0.46) -4.52 (0.36) -4.36 (0.28) 0 15 60 0 51 77 0.91 (0.30) 0.97 (0.17) 0.96 (0.10) 15.39 13.18 12.62 0.67 0.62 0.50 1 basket width=3% Wus$=0.6 2.1 basket width=5% (w/o crawl) Wuss=0.6 2.2 basket width=5% (with crawl) - announc OLS OLS First First Differences Differences Model Model Wuss=0.6 Wus$=0.5 4 8 4 50 basket 100 width=7% 200 Wuss=O.603 5 50 basket 100 width-15% 200 expanding band Wus5 =0.603 Precision is calculated as the sum of absolutes deviations of the estimated weights at 50, 100 and 200, with respect to the announced weights. In the restricted model, the DM and the FF on the one hand, and the US$ and the JY on the other, were combined using the relative announced weights, to form new variables. See Appendix 2 for details. Table 3: Swiss Franc and Randomly GeneratedVariable as Dependent Variable Percentage of Observationsfor Which Null HypothesisIs Rejected (1%) Random Swiss Franc Period Obs Test 1 Test 2 Test 1 Test 2 Ho: Ho: Ho: Ho: Weights= Weights Weights= Weights announced = 0 announced = 0 Announcement: 50 100 100 97 0 Chile-PeriodI 100 100 100 99 0 200 100 100 99 0 Announcement: 50 100 100 89 0 Chile-Period2 100 100 100 95 0 200 100 100 98 0 Announcement: 50 100 100 76 0 0 100 100 90 Israel-Period2.2 100 200 100 100 95 0 The rejectionpercentageswere recalculatedfor the referredcountryperiods,replacingin each case the local currency with the Swiss franc and a randomlygenerateddata. Fictitious data were generatedfollowingan AR(I) processwith parametersobtainedby fittingan AR(l) model to the originaldependentvariable. Table 4: Floating Exchange Rate Regimes - First Differences Linear Model Percentage of Observations for Which HO:Weights=O Is Rejected (1%) obs. 20 50 100 150 200 Brazil 1995 1996 1992 1993 1994 Mana ed Floatin2 0 0 0 0 0 0 0 41.8 2.2 0 62.4 1.4 0 72.3 1 0 73.2 87.9 92.2 94.2 90.9 97.6 98.9 99.3 99.5 1997 Band 100 100 100 100 100 Mexico 1994 99 obs. 1990 1991 1992 1993 CrawlingPeg 20 100 72.7 100 100 _ 100 0 50 100 92.7 100 100 150 100 200 100 196 _ 0 1998 1999 Free Fl 0 0 0 0 90.9 97.6 98.9 99.3 99.5 99 17 Sf eeE1oati;> ; 18.2 0 100 100 95.1 2 2 2 8 96.7 100 100 97.8 604 5 68 484 97.9 100 100 98.6 7, 16 7. 75 98.4 100 100 99 6 34 78 01 9. - Peru 99 obs. 209. 50 94 52 150 58 IS $6 17 198 94 9.0 76 7. ~ 6A 5. 1. 37 752 8. 94 21 8.5 200 ~, obs. 1990 1991 1992 20 50 100 150 200 100 100 100 100 100 90.9 97.6 98.9 99.3 99.5 100 100 100 100 100 99 7. 93 9. . 74 S. Korea 1993 1994 1995 Mn"agedFloating 63.6 45.5 54.5 90.2 85.4 87.8 95.6 93.4 94.5 97.2 95.7 96.5 97.9 96.9 97.4 1996 1997 98 19 18.2 61 82.4 88.7 91.6 18.2 26.8 67 78.7 84.3 0 00 00 142 3.6 Thailand obs. 20 50 100 150 200 1990 100 100 100 100 100 1991 90.9 97.6 98.9 99.3 99.5 1992 81.8 95.1 97.8 98.6 99 1993 1994 Basket Peg 100 100 100 100 100 100 100 100 1995 1996 1997(I' half) 19("'af Fe 100 100 100 100 100 63.6 90.2 95.6 97.2 100 100 97.9 10000 10000 100 1 98 19 0 0 0 9.90 41.8 ll 0 1 1. . 0 Appendix Figure 1: Monte Carlo Simulations-Role of Band Size Quantiles of Test A (Weights=Rate of Crawl=O) 350 300* 250- ~200150 o 100 50- 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 Sigma 5% 0 25% *0°/o 0 75% W95% Quantiles of Test B (Weights=O) 1200 1000 - 800600 O 400- 200 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 75% x* 95% 0.09 0.1 Sigma -5% - 25% - 50% : Parameters of estimations: 500 samples; weights on dependent variables 1/3 for US$, DM, and JY; initial observation February 24, 1986; constant=-1;rate of crawl=0.10; sigma={0.01; 0.028; 0.046; 0.064; 0.082; 0.1 }. Quantile values are calculated for the first 10 rejections. Appendix Figure 2: Monte Carlo Simulations-Role of Number of Currencies in Basket Quantiles of Test A (Weights-Rate of Crawl=O) 350 - 300 250- 200 150 10050 O-- 0.01 0.048 0.086 0.124 0.162 0.2 Sigma + US only - A US + DM + JY US + DM Quantiles of Test B (Weights=O) 700 600 - 500 0g 400- 300- 2001000 0* I - 0.01 I 0.048 0.086 0.124 0.162 0.2 Sigma US only - US + DM - US + DM + JY Parameters of estimations: 500 samples; initial observation February 24, 1986; constant=,; rate of crawl=0.10; sigma={0.01; 0.048; 0.086; 0.124; 0.162; 0.2}; weights on dependent variables are 1, 1/2, and 1/3, for 1, 2, and 3 currencies in the basket respectively. Quantile values are calculated for the first 10 rejections. Appendix Figure 3: Monte Carlo Simulations-Role of Rate of Crawl Median Valules for Tests A & B Test A Test B 400 - 650 350 - 600 55 300 0 5 250200 - 500 150 - 450 100 50 - 400 350 0 0 0.1 0.3 0.2 0.4 0.5 Crawl TestA -- TestB TestA:Weights=Rateof Crawl=OTestB: Weights=O Parametersof estimations:500 samples;initialobservationFebruary24, 1986;constant=l;rate ofcrawl= weightson dependentvariablesare equalto 1/3for each {0.01;0.108;0.206;0.304;0.402;0.5};sigma=O.1; currencyin the basket.Medianvaluesare calculatedfor the first 10rejections. Appendix Figure 4: Monte Carlo Simulations-Role of Period and Variability of Regressors Median Values for Test A 180 160 140 120 100 80 60 40 20 0 Mar-86 Mar-88 Mar-90 Mar-92 Inverse Variance of US$, DM, JY - Mar-94 Mar-96 Median Test A Median Values for Test B 180 160 140 120 100 80 60 40 20 0 Mar-86 Mar-88 _Inverse Mar-90 Mar-92 Variance of US$, DM, JY Mar-94 - Mar-96 Median Test B Parameters of estimations: 500 samples; weights on dependent variables 1/3 for US$, DM, and JY; constant=1; rate of crawl=O.O0;sigma=0.005. Median values are calculated for the first 10 rejections. "Inverse Variance" is the inverse of the average standard error of the three currencies, for the first 50 observations of each respective period. Appendix Table A.1: Exchange Rate Policy in Chile 19982-1999 Date September, 1982 * August 1, 1984 June, 1985 January 5, 1988 June 6, 1989 * * * * * * April 3, 1991 January 23, 1992 * * * March, 1992 July, 1992 November, 1994 November 30, 1994 December, 1995 January 21, 1997 June 25, 1998 September 16, 1998 * * * * * * . * * * * December 23, 1998 * * January 1, 1999 September 2, 1999 . . * Policy Daily devaluationsin line with domestic inflation in the preceding month minus an estimate of external inflation Band of +/- 0.5% Widening to 2% Widening to 3 % Widening to 5% Accelerate the rate of real depreciation,which was achieved by reducing the estimate of internationalinflation Adjustment of central parity: previous month inflation minus estimated international inflation 2% revaluation of central parity Band widened to 10% (from +1-5%) Discrete 5% revaluation of central parity Managed floating is authorized Central parity: 50% U.S. dollar, 30% Deutsche mark, 20% Japaneseyen Central parity: 45% U.S. dollar, 30% Deutsche mark, 25% Japanese yen 9.66% revaluation of central parity 2% revaluation; 2% annual revaluation 4% revaluation of central parity New band: +/- 12.5% New weight: 80% U.S. dollar, 15%Deutsche mark, 5% Japanese yen 2% annual revaluation New asymmetricband: +2%, -3,5% New band: +/- 3.5% The band is widenedprogressivelyuntil it accumulates and additional 1.5% in each extreme , such that by the end of the year the band would be +/- 5% New estimates of annual internationalinflation from 2.4% to 0% for the rest of the year The relevant internal inflation rate is the inflation target and not past inflation New band: +/-8% No change in other parameters (centralparity adjusts only with internal inflation and the band continue widening daily by 0,013575%) Deutschemark is replacedby the euro, with the same weight Free floating with managed interventiononly in exceptional cases Release of new informationregarding interventionsin the foreign exchange markets Source:CentralBankof Chile,HusseyandMorande(1996),andVergara(1994) Appendix Table A.2: ExchangeRate Policy in Israel 1986-1999 Date August 1, 1986 * * January 3, 1989 * June 23, 1989 * * March 1, 1990 * * September 10, 1990 March 11, 1991 December 17, 1991 * November 9, 1992 * * * * * July 26, 1993 May 31, 1995 * * * * * * April 30, 1996 June 18, 1997 * * August 17, 1]998 * * * o January 4, 1999 * * Policy Beginning of basket peg without crawl Initial weights: 60% US$, 20% DM, 10% BP, 5% FF, 5% JY Central parity is devaluated 13% in a week A ±3% band is introduced Midpoint raisedby 6% Midpointraised by 6% Band widenedto ±5% Midpointraisedby 10% Midpointraisedby 6% Introductionof crawlingband Midpointraisedby 3% Slope of band 9% Midpointraised by 3% Slope reduced to 8% Midpointraised by 2% Slope reduced to 6% Midpointraised by 0.8% Band widenedto ±7% No change to slope Weights. 54.8% US$, 24.2% DM, 8.3% BP, 5.6% FF, 7.1% JY Weights: 60.3% US$, 21% DM, 8% BP, 5.1% FF, 5.6% JY Band widenedto reach ±15%by end of year Slope of lower limit 4% Slope of upper limit 6% Slope of lower limit 2% Slope of upper limit 6% DM and FF are replacedby Euro Weights: 61.4% US$, 8.9% BP, 5.2% JY, 24.5% Euro Source:Bankof Israel,"ForeignCuirencyExchangeRatesIn Israel 1999,"January2000. Policy Research Working Paper Series Title Contact for paper Author Date WPS2378Disintegrationand Trade Flows: Evidencefrom the FormerSoviet Union SimeonDjankov CarolineFreund June 2000 R. Vo 33722 WPS2379 Indiaand the MultilateralTrading Systemafter Seattle:Towarda ProactiveRole Aaditya Mattoo ArvindSubramanian June 2000 L. Tabada 36896 WPS2380Trade Policiesfor Electronic Commerce Aaditya Mattoo Ludger Schuknecht June 2000 L. Tabada 36896 WPS2381Savingsand theTermsof Trade underBorrowingConstraints Pierre-Richard Agenor JoshuaAizenman June 2000 T. Loftus 36317 WPS2382 Impedimentsto the Developmentand ThorstenBeck Efficiencyof FinancialIntermediation in Brazil June 2000 E. Mekhova 85984 WPS2383 NewFirmFormationand Industry Growth:DoesHavinga Market-or Bank-BasedSystemMatter? June 2000 E. Mekhova 85984 WPS2384Are Cost ModelsUsefulfor Telecoms DanielA. Benitez Regulatorsin DevelopingCountries? Antonio Estache D. MarkKennet ChristianA. Ruzzier July 2000 G. Chenet-Smith 36370 WPS2385The Rise,the Fall,and ... the EmergingRecoveryof Project Financein Transport Antonio Estache John Strong July 2000 G. Chenet-Smith 36370 WPS2386Regulatorsand the Poor:Lessons fromthe UnitedKingdom RichardGreen July 2000 G. Chenet-Smith 36370 July 2000 G. Chenet-Smith 36370 ThorstenBeck RossLevine WPS2387The Longand WindingPathto Private AntonioEstache Financingand Regulationof Toll ManuelRomero Roads John Strong WPS2388The Roleof Specialand Differential ConstantineMichalopoulosJuly 2000 Treatmentfor DevelopingCountriesin GATTand the WorldTradeOrganization L. Tabada 36896 WPS2389Vietnam:On the Roadto LaborIntensiveGrowth? Patrick Belser July 2000 H. Sutrisna 88032 WPS2390The SocialRateof Returnon InfrastructureInvestments DavidCanning Esra Bennathan July 2000 H. Sladovich 37698 Policy Research Working Paper Series Title Contact for paper Author Date MartinRavallion July 2000 P. Sader 33902 WPS2392 What FactorsAppearto DrivePrivate Dipak Dasgupta CapitalFlowsto DevelopingCountries?Dilip Ratha AndHow DoesOfficial Lending Respond? July 2000 S. Crow 30763 WPS2393Will the EuroTriggerMoreMonetary Unionsin Africa? PatrickHonohan PhilipR. Lane July 2000 A. Yaptenco 31823 WPS2394 Tax Evasion,Corruption,and the Remunerationof Heterogeneous Inspectors WalyWane July 2000 H. Sladovich 37658 WPS2395 Decentralizingthe Provisionof Health WilliamJack Services:An IncompleteContracts Approach July 2000 H. Sladovich 37698 WPS2396Aid Dependenceand the Qualityof Governance:A Cross-Country EmpiricalAnalysis July 2000 P. Sintim-Aboagye 38526 WPS2391Are the PoorProtectedfrom Budget Cuts?Theoryand Evidencefor Argentina StephenKnack