_WPS2612 POLICY RESEARCH WORKING PAPER 2642 Does the Exchange Rate The exchange rate regime does make a difference for Regime Affect inflation performance. It is M acroeconomic difficult to infer its effect on growth, but policy variables- Performance? and other variables influencing economic activity-do have different Evidence from Transition Economies effects on growth under different exchange-rate Ilker Doma, arrangements. Kyle Peters Yevgeny Yuzefovich The World Bank Europe and Central Asia Region Poverty Reduction and Economic Management Sector Unit July 2001 |POLICY RESEARCH WORKING PAPER 2642 Summary findings To examine whether a country's exchange rate regime Switching from a floating regime to an intermediate has any impact on inflation and growth performance in regime might not reduce inflation. transition economies, Doma,, Peters, and Yuzefovich * An unanticipated float-when a country whose develop an empirical framework that addresses some of fundamentals make it unlikely to adopt another regime the main problems plaguing empirical work in this strand adopts a floating regime-results in lower inflation. of the literature: the Lucas critique, the endogeneity of Based on their results, it is not possible to infer more the exchange rate regime, and the sample selection about one particular exchange rate regime being superior problem. to another in terms of growth performance. But Empirical results demonstrate that the exchange rate empirical findings do underscore the different effects regime does affect inflation performance. The results that policy variables-and other variables influencing suggest that: economic activity-have on growth under different Transition countries with intermediate arrangements exchange-rate arrangements. might reduce inflation if they were to adopt a fixed regime. This paper-a product of the Poverty Reduction and Economic Management Sector Unit, Europe and Central Asia Region-is part of a larger effort in the region to understand the links between exchange rate arrangements and macroeconomic performance in transition economies. Copies of the paper are available free from the World Bank, 1818 H StreetNW, Washington, DC 20433. Please contactArmanda Carcani, room H4-326, telephone 202-473-0241, fax 202- 522-2755, email address acarcani@worldbank.org. Policy Research Working Papers are also posted on the Web at http:/ /econ.worldbank.org. The authors may be contacted at idomac@worldbank.org or kpeters@worldbank.org. July 2001. (65 pages) The Policy Research Working Paper Seoes disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings ouit qauickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the autbors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Policy Research Dissemination Center Does the Exchange Rate Regime affect Macroeconomic Performance? Evidence from Transition Economies Ilker Domaq, Kyle Peters and Yevgeny Yuzefovich Table of Contents 1. Introduction .....................................................................3 2. A Brief Overview of the Evolution of the Exchange Rate Regimes in the 1990s .....................................7 3. The Relationship between Nominal Exchange Rate Regimes and Economic Performance: A Brief Review of the Literature .................................................................... 12 4. Macroeconomic Performance and the Exchange Rate Regime: Stylized Facts from Transition Countries .................................................................... 17 5. The Empirical Framework .................................................................... 23 6. Empirical Results .................................................................... 26 6.1. The Determinants of the Choice of Exchange Rate Regime ................................................... 26 6.2. The Exchange Rate Regime and the Inflation Performance .................................................... 31 6.3. The Exchange Rate Regime and the Growth Performance ..................................................... 39 6.4. Robustness Test .................................................................... 46 7. Conclusions .................................................................... 46 Appendix 1: Exchange Rate Versus Money Based Stabilization: A Cursory Look at the Transition Experience .................................................................... 50 Appendix 2: Extension of the Heckman Procedure for three Regimes ....................................................... 58 Appendix 3: Simulation Exercise .................................................................... 60 Appendix 4: Description of Data .................................................................... 61 References .................................................................... 62 2 1. Introduction The issue of the appropriateness of exchange rate arrangements has returned to the forefront as a result of the recent crises in Asia, Russia, Brazil, and more recently economic developments in Argentina. More precisely, the debate over fixed and flexible exchange regimes has once again taken center stage. Some claimed that the first round of this debate was won by those advocating flexible regimes: all crisis episodes took place in countries which had adopted a variety of mechanisms for pegging more or less closely to the dollar.' Fixed exchange rates, soft pegs in particular, were blamed for the recent financial meltdowns.2 The advocates of fixed exchange regime, however, have asserted that there are bad fixes and good fixes: a good fix is, for example, full dollarization [Calvo (1999), Hanke and Schuller (1999)]. Clearly, this controversy, which has raged in the economic literature for more than a century, continues unabated. An important recent development in the debate over most appropriate best exchange rate arrangement is the recognition that the choice of the exchange rate regime for developing countries is different from that of developed countries.3 Developing countries are often beset by a lack of credibility and limited access to international markets; they are beset by more pronounced adverse effects of exchange rate volatility on trade, high liability dollarization, and higher passthrough from the exchange rate to inflation. Consequently, benign neglect of the exchange rate is not a feasible option for developing countries. Admittedly, empirical corroboration of the arguments set forth in the literature has been the least explored part of this debate. Contrary to the large number of theoretical and conceptual discussions, relatively few studies have made an attempt to investigate empirically the link lSee Calvo (1999) for a more detailed discussion of this issue. 2 See, for instance, Goldstein (1999). 3 between macroeconomic performance and the exchange rate regime. This is, perhaps, because such an empirical investigation is fraught with difficulties, including the problem concerning the classification of the exchange regime.4 In spite of the growing interest over the link between the exchange rate regime and macroeconomic performance, the burgeoning empirical literature on transition economies has paid little attention to this issue.5 It has largely focused on recovery and growth as well as price liberalization and inflation.6 Some of the existing studies made an attempt to incorporate only the effect of the adoption of a fixed exchange rate regime on inflation and growth with mainly two objectives in mind: (i) to capture favorable confidence effects of nominal exchange rate anchors on velocity; and (ii) to account for the output costs of stabilization associated with the adoption of a particular nominal anchor, namely the exchange rate. Nevertheless, none of the studies made an attempt to investigate explicitly the links between the nominal exchange rate regime and macroeconomic performance. This paper aims to fill this void by investigating empirically the link between the exchange rate regime and macroeconomic performance in transition economies.7 To this end, we develop an empirical framework that addresses some of the main problems plaguing empirical work in this strand of the literature, namely the Lucas critique, endogeneity of the exchange rate regime, 3 Calvo (1999) and Calvo and Reinhart (2000a, 2000b) 4 See, for instance, Ghosh et al (1997), Baxter and Stockman (1989), and Edwards and Savastano (1999) for a review of problems encountered by empirical studies in this literature. Studies by Dombusch (1994) and by Sachs (1996) are among the few papers focusing on the macroeconomic implications of the exchange rate regime and on the choice of the exchange rate in the transition countries. More recently, series of papers- presented at an Association for Comparative Economic Studies panel entitled "Exchange Rate Policies in Transition", in Chicago, January 4, 1998-made an attempt to explore issues related to exchange rate regime in the countries in Transition. Majority of the papers were descriptive in their nature and made no attempt to empirically investigate the macroeconomic implications of the exchange rate regime in transition economies. 6 See, for instance, Berg et al. (1999), Hemnandez-Cata (1999), Christoffersen and Doyle (1998), Fischer et al. (1998, 2000), and Havrylyshyn et al. (1998). 7 This study will not explore issues related to monetary and exchange rate policy encountered by countries negotiating EU accession-Czech Republic, Estonia, Hungary, Poland, and Slovenia. See Corker et al (2000) and Masson (1999) for a detailed discussion of these issues. 4 and the sample selection problem.8 More specifically, we utilize a switching regression model which is estimated using a two-step Heckman procedure. First, we estimate the equation for the choice of the exchange rate regime by using ordered probit. Second, we utilize a switching regression technique to investigate whether the exchange rate regime has a bearing on inflation and growth performance in transition economies. When tackling this controversial topic in the context of transition economies, however, two issues emerge. First, it is important to make a distinction between the appropriateness of the exchange rate arrangements in the earlier phase of the transition process-money vs. exchange rate based stabilization debate-and the appropriateness of the exchange rate arrangements (in the aftermath of the stabilization) for long-run economic management. Second, one needs to clarify whether the same economic principles of exchange rate policy apply both to market economies and transition economies. Put differently, are transition economies so unique that what characterizes market economies or developing countries has little relevance to them? The first issue, though beyond the scope of this investigation, will be discussed briefly by looking at some stylized facts about the performance of transition countries that adopted different anchors in their stabilization programs.9 In the case of the second issue, the paper, while conceding that transition economies have distinct features-such as extreme forms of central planning which meant price controls, chronic excess demand, and forced saving- compared to other developing countries, will proceed under the assumption that the fundamental tenets of exchange rate policy apply to any and all types economies [Guitian (1994) and 8 The Lucas critigue states that when there is a policy switch the coefficients associated with policy variables should change. This is because the way in which expectations are formed-the relationship of expectations to past information-changes when the behavior of forecasted variables changes. The sample selection problem arises from the fact that countries do not choose their exchange rate regimes randomly. Instead, their choice hinges on a set of fundamentals, which, in tum, affects macroeconomic outcomes such as inflation and growth. Consequently, the use of standard econometric techniques such as OLS or 2SLS will produce biased results stemming from the correlation between the regime choice and the error term in either the inflation or growth equation. 5 Dombusch (1994)1.10 The investigation, however, will attempt to make the necessary modifications to account for the distinct characteristics of transition economies, where possible. The principal conclusions that emerge from our study are: 1. Transition economies that: (i) have lower budget deficits; (ii) are more open (i.e. have a higher ratio of exports plus imports to GDP); and (iii) made more progress in private sector entry and internal markets tend to adopt more stringent exchange rate regimes. While the results suggest that those which have made more progress in opening to external markets and with a reserves to monetary base ratio above 1.34 opt for more flexible exchange rate arrangements. 2. The exchange rate regime does make a difference for inflation performance. The findings imply that countries with intermediate arrangements may achieve lower inflation if they were to adopt a fixed regime. The results also suggest that switching from a floating regime to an intermediate arrangement may not deliver lower inflation since their fimdamentals may be inappropriate for an intermediate regime. However, when a country with an intermediate regime switches to a floating regime, it experiences higher inflation. 3. The results also suggest that the case of an unanticipated float-a situation describing a country where fundamentals make it likely to adopt another regime, but it adopts a floating regime-results in lower inflation. 4. Based our empirical results, however, it is not possible to make any inference about a particular exchange rate regime being superior to the other in terms of growth performance. Nonetheless, empirical findings suggest that policy variables-and also other variables 9In Appendix 1, we try to highlight some stylized facts from the stabilization performance of transition countries under different anchors. 10This conjecture, however, does not imply that the working of a particular exchange rate policy or arrangement is independent of the characteristic of the economy in which it is being pursued. 6 influencing economic activity-do have a different impact on growth under different exchange rate arrangements. The remainder of the paper is organized as follows. Section 2 discusses the overall trend in the evolution of the exchange rate regimes both in general and in the context of transition economies in the 1990s. Section 3 provides a brief review of the literature focusing on the link between exchange rate regimes and macroeconomic performance. Section 4 takes a cursory look at the evolution of key macroeconomic variables in transition economies under different exchange rate arrangements. Section 5 describes the empirical framework. Section 6 reports empirical findings. Finally, Section 7 concludes the paper. 2. A Brief Overview of the Evolution of the Exchange Rate Regimes in the 1990s Following the collapse of the Bretton Woods system of fixed exchange rates in the early 1970s, there has been a gradual shift from fixed to more flexible exchange rates. Initially, many developing countries pegged their currencies either to a single currency (usually the USD or FF) or to a basket of currencies. By the late 1970s, they began to shift from single currency pegs to basket pegs. In the early 1980s, developing countries shifted away from currency pegs towards more flexible exchange rate arrangements." A glance at the evolution of the exchange rate regimes of developing countries and of the transition economies during the 1990s reveals an interesting trend (Figure 1). Since 1994, developing, including transition, countries appear to have shifted away from fixed and independent floating exchange rate regimes towards intermediate flexibility. " i In 1975, for example, 87 percent of developing countries had some type of pegged exchange rate. By 1996, this proportion had declined to well below 50 percent. When the relative size of economies is taken into consideration, the shift is even more pronounced. In 1975, countries with pegged rates accounted for 70 percent of the developing world's total trade; by 1996, this figure had dropped to about 20 percent. 7 Transition countries have adopted a broad variety of different exchange rate regimes.'2 A large number of countries, from the outset, let their currencies float while maintaining some scope for intervention [Albania (1992), Bulgaria (1991), Romania (1993), and Slovenia (1992) as well as several CIS republics]. Some countries, on the other hand, opted for a fixed regime from the outset [Croatia (1993), Czechoslovakia, later the Czech Republic, and Slovakia, Poland until 1991, and Macedonia (1994)]; three chose the extreme of a currency board (Estonia and later Lithuania as well as Bulgaria). Other countries decided to purse a more flexible approach and introduced a crawling peg system (Poland from October 1991, Hungary from March 1995) or a fixed but adjustable peg (Hungary until 1995). Not surprisingly, the appropriate exchange rate regime for transition economies-both in the case of the initial phase of reform and more advance stage of economic transformation-has stimulated much debate. A recent statement by Vaclav Klaus (1997) on this particular issue is quite telling: The collapse of communism "happened" in the moment when the economic profession believed in fixed exchange rates and in the advantage of anchoring the economy by means of one fixed point-especially in a situation when all other variables undergo large changes and fluctuations. I have to confess that I was originally afraid of introducing such a rigid regime but the first impressions were positive because we succeeded in choosing an exchange rate which functioned well for a very long seventy- six moths. By sufficiently devaluing the crown on the eve of price liberalization we formed something what I later called the "transformation cushion". The exchange rate cushion (as well as the parallel wage cushion) appeared to be crucial for the whole subsequent transformation process. The inflation differential was, in our case, not as big as in some other transforming countries but the appreciation in real terms reached in seventy-six months was almost 80 percent, which was too much. Although we have been constantly checking the remaining thickness of our exchange rate cushion, as we see it now, we-probably in the middle of 1996-missed the most suitable moment for the abolition of the fixed exchange rate regime. The question is, however, whether the subsequent movements of the rate of exchange rate would have been less dramatic than they were in reality in recent months. The vulnerability of an emerging market economy is, in this respect, very high and, probably, unavoidable. 12 See Corker et al (2000) for a more detailed discussion of this issue for selected advanced transition economies. 8 In terms of the choice of the exchange rate regime, he draws a tentative lesson: "A fixed exchange regime should not last too long". Although it may sound trite, the clearest conclusion that has emerged from discussions over this controversial topic was that the adoption of a particular regime is neither a necessary nor a sufficient condition for the realization of desired macroeconomic outcomes. More specifically, it is argued that different exchange rate arrangements can contribute to macroeconomic stabilization provided that: (i) the authorities implement prudent macroeconomic policies consistent with the exchange rate regime in place; (i) the regime is compatible with intial macroeconomic conditions of the country; and (iii) the regime is not altered too frequently so that the necessary credibility can be established.13 It is interesting to note that the recent trend towards intermediate regimes (see Figure 1) is in contrast to arguments set forth by some analysts, who assert that the growing integration of international capital markets over the past two decades requires a clarification of the exchange rate regime.14 They argue that it is not possible to have hybrid solutions endeavoring to reconcile too many objectives. One has to opt for fairly free floating exchange rates or very credibly fixed ones. In short, they conclude that "middle way" solutions, involving fixed but adjustable exchange rates have been rendered more unstable by the growth of capital flows. 13 See, for instance, Radzyner and Riesinger (1997). 14 See, for example, Crockett (1997). However, Fischer (2001) argues that developing countries which are not very exposed to international capital inflows still encounter a wide range of intermediate options. Moreover, it should be noted that Figure I presents the developments up to 1998. As was shown in Fischer (2001), there was a notable decline in the number of countries with intermediate regimes in 1999. 9 Figure 1: Evolution of the Exchange Rate Regimesa All Countriesb Transition Economiesc 100 20 Pegg"d_ -h,rmedib Ixbi -MM Pegged ---- kdependentFloahig ki1rmdUab FI6xbEO, …IdependentFIoaIng 80 15 - 60 10/ 40~~~~~~~~~~~~~~~~~~~~~~ 90 91 92 93 94 95 96 97 98 90 91 92 93 94 95 96 97 98 Source: Exchange Rate Arrangements and Exchange Rate Restrictions, IMF. a:The pegged regimes include single currency pegs, SDR pegs, other published basket pegs, and secret baskets. The intermediate group contains cooperative systems, unclassified floats, and floats with pre-determined range. The float group comprises of floats without pre-determined range and pure floats. b: It consists of 181 countries (as of 1998) reported in the IMF's Exchange Rate Arrangements and Exchange Rate Restriction. c: It consists of Albania, Armenia, Azerbaijan, Belarus, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakstan, Kyrgyz Republic,Latvia, Lithuania, Macedonia, Moldova, Poland, Romania, Russia, Slovak Republic, Slovenia, Tajikistan, Turkmenistan, and Ukraine. This view, however, does not enjoy unanimous support. Specifically, it is argued that there are good reasons for many countries to adopt intermediate regimes in spite of the substantial increase in capital mobility. Proponents of intermediate arrangements contend that presence of a number of safety valves can make such regimes viable, in particular if they are adopted in the context of a broader economic and political integration process.15 They argue that corner solutions tend to be the exception rather than the rule for many countries: currency boards entail very demanding preconditions to be viable, while flexible exchange rate arrangements tend to have considerable disadvantages for small-open economies. 15 See, for instance, Backe (1999). 10 Recent developments in the international monetary and financial environment have had a significant impact on the evolution of exchange rate regimes in at least three aspects."6 First, recent advances in telecommunications and information technology have reduced transactions cost in financial markets and prompted both financial innovations and liberalization and deregulation of domestic and international transactions. As a consequence, there has been a sharp increase in capital mobility. The noticeable expansion of both gross and net capital flows between developed and emerging markets is a case in point. Balance of payments statistics demonstrate that net annual inflows into emerging economies increased from virtually zero in 1989 to reach $307 billion in 1996, before declining to about half that level in 1997 and 1998.17 Second, the increasing integration of emerging market economies into the world economy has enabled them to enjoy the benefits of globalization. At the same time, however, made these countries more susceptible to sudden reversal in capital flows. Private capital flows have emerged as one of the most important elements of adjustment and financing mechanisms in emerging economies. Finally, the launch of the Euro marks the creation of a multi-polar currency system, moving away from dependence on the dollar as the dominant currency of the system. This development has important implications for the system as to whether the exchange rate between major currencies will continue to undergo large fluctuations as occurred in the 1980s and 1990s.18 Indeed, evidence to date-i.e. the evolution of the Euro vis a vis the dollar-appears to suggest that such oscillations between major currencies are likely to resume in the future as well. 16 IMF (2000). 17 See HF (1998). 8 For instance, the appreciation of the dollar against the yen prior to the Asian crisis was considered as one of the contributing factors to the crisis since the exchange rate in most of the crisis countries was rigidly pegged either to the dollar or a basket dominated by the dollar. 11 The above described developments, to a large extent, contributed to the documented trend towards greater exchange rate flexibility and subsequent diminution in the use of the exchange rate to anchor monetary policy. More specifically, the fact that both developing and transition countries are more exposed to currency movements compared to developed countries and that they lack deep financial markets and strong financial institutions suggests that "benign neglect" of the exchange rate is not a feasible option for them. Consequently, many developing and transition countries are, perhaps, inclined to pursue a hybrid arrangement with limited flexibility to the exchange rate-via bands or other limits on fluctuations against some other currency or currencies-but without the rigidity embedded in currency pegs. 3. The Relationship between Nominal Exchange Rate Regimes and Economic Performance: A Brief Review of the Literature Orthodox discussion of the choice between fixed and flexible regimes hinges on the nature of the shocks.'9 Standard models imply that floating rates will be advantageous when disturbances are primarily monetary and foreign, since in this case exchange rate changes can largely insulate the domestic economy. Pegged rates are preferable when shocks are associated mainly with unstable domestic monetary and financial policies as pegged rates will help discipline erratic policy makers. Proponents of flexible exchange rates claim that these regimes are more efficient than fixed exchange rates in correcting balance of payments disequilibria. Furthermore, they underscore that by allowing a country to achieve external balance easily and automatically, flexible rates facilitate the achievement of internal balance and other economic objectives of the country. On the other hand, advocates of fixed exchange rates contend that by introducing a 12 degree of uncertainty not present under fixed rates, flexible exchange rates decrease the volume of international trade and investment, are more likely to lead destabilizing speculation, and are inflationary. Furthermore, one of the main appealing features of floating exchange rates-the ability to absorb shocks-has recently been challenged. It is argued that countries with flexible exchange rates-except those with well-developed and sophisticated markets-are likely to experience a surge in the volatility of the real value of domestic assets due to increased capital mobility. Excessive fluctuations in the real value of domestic assets may, in turn, undermine stability [Cooper (1999)]. The modern literature-also considering the extreme arrangements of flexible and fixed regimes-places great emphasis on the presence of important trade-offs between credibility and flexibility.20 A floating regime enables a country to have an independent monetary policy so that the economy can accommodate domestic and foreign shocks such as changes in terms of trade and interest rates. However, this flexibility is achieved at the cost of some loss in credibility which, in turn, tends to be associated with higher inflation. Fixed exchange rates, on the other hand, reduce the degree of flexibility but bring a higher degree of credibility to policy making. Since, under fixed rates, agents believe that the primary objective of monetary policy is to maintain the parity, they moderate their price and wage expectations, thereby leading the economy to achieve a lower inflation rate. A careful review of the theoretical arguments put forth by each side does not lead to any definitive conclusion that one system is overwhelmingly superior to the other. For instance, contrary to the traditional ranking between fixed and floating regimes, which is based on a loss 19 This ranking is based on a loss function that depends on output volatility. 20 See, for instance, Edwards (1996) and Frankel (1995). 13 function that depends exclusively on output volatility, Calvo (1999b) shows that fixed exchange rates would always dominate flexible regimes if the function being optimized places weight on real exchange rate volatility.21 Furthermore, since shocks could contain both real and nominal components in practice, the choice of exchange rate regime on the basis of the nature of shocks becomes problematic. In fact, recent crises episodes in which shocks have come largely through the capital account-affecting both aggregate demand as well as money demand-lend support to this conjecture and cast doubts about the usefulness of floating exchange rates as a shock absorber. What is the empirical evidence linking inflation and output growth with the exchange rate regime? Although some suggestive stylized fact are beginning to emerge, the evidence is still quite limited. More specifically, studies that have tried to ferret out the influence of exchange rate arrangements on economic performance can be grouped under two categories: country specific studies and multi-country studies. Country specific investigations has had a difficult time unraveling the independent effects of the nominal exchange rate regime on macroeconomic performance: detection of regularity associated with a particular regime in one study was followed by a counter example in another study. Multi-country studies have also found it difficult to make generalizations. For instance, Little et al. (1993) conducted a comprehensive study covering 18 developing countries. They found that while in some countries a fixed exchange rate regime was associated with lower inflation, in other episodes the exchange rate turned out to be an ineffective nominal anchor. Edwards (1993) studied whether, ex ante, the exchange rate regime has an impact on inflationary performance by introducing financial discipline. He employed a sample from 52 21 He also shows that this dominance weakens, but does not vanish, with full indexation to the exchange rate. 14 countries over the period 1980-89. His results showed that countries with fixed exchange rates had lower inflation rates during the 1980s compared to countries with flexible arrangements. Tomell and Velasco (1999), however, challenged Edwards' findings on theoretical grounds, pointing out that a depreciating currency is a more immediate and observable signal of fiscal indiscipline than a decline in reserves that appears with delay and can be concealed. They found empirical support for their position by examining the behavior of 28 sub-Saharan African countries. Ghosh et al. (1997)-one of the most comprehensive multi-country studies-examined the effects of the nominal exchange rate regime on inflation and growth using data from 136 countries during the period of 1960-89. They found that both the level and variability of inflation was markedly lower under fixed exchange rates than under floating exchange rates. However, their findings also suggest that the inflation bias of flexible exchange rate arrangements does not seem to be present among the pure floaters in the sample-particularly among the high and upper middle income ones. This implies that the positive association between exchange rate flexibility and inflation found in the study may not be monotonic.2 Their study failed to find a robust link between growth and currency regimes, probably because investment ratios are higher but trade growth somewhat lower under fixed than under floating exchange rates. However, they found that the variability of real output is noticeably higher under fixed than under floating exchange rates.23 Moreover, a recent study by Hausmann et al. (1999) demonstrated that during the 1990s Latin American countries with fixed exchange rates had greater financial depth-as measured by 22 It should be noted that this finding seems to contradict the conclusion that Quirk (1994) reached in his review of previous empirical literature: there is not much linkage between exchange rate arrangements and inflation. 23 A recent IMF study (1997) extends the period to mid-1990s and reaches similar conclusions. This implies that findings of Ghosh et al. (1997) were not greatly influenced by the increased access to international markets enjoyed by developing countries in the 1990s. 15 M2/GDP-lower interest rates, and less effective wage indexation than those with floating exchange rates. Their results also indicated that monetary policy under floating rates has been more pro-cyclical than under fixed rates.24 All in all, one recent review of the empirical literature suggests that empirical investigations in this area suffer from the following problems plaguing empirical work in economics:25 * Cross-country analyses investigating the inflation performance of countries with different regimes are potentially subject to a survival bias. The difficulty is that only countries that have succeeded in defending the peg are included in the fixed exchange rate group. Whereas, countries that adopted a fixed exchange rate, but could not sustain it, are usually grouped under flexible exchange rate regime category.26 • Discrepancies between declared and effective exchange rate arrangements can be an important source of error. * Endogeneity of the choice of the exchange rate regime or reverse causation also constitutes a major problem in empirical studies. It is not clear whether a fixed exchange rate causes lower inflation or whether countries with low rates of inflation adopt this kind of arrangement.27 24 A recent study by Doma9 and Martinez-Peria (2000) investigated the issue at hand from a different aspect by considering the link between banking crises and the exchange rate regimes. 25 Edwards and Savastano (1999). 26 See Aghevli et al.(1991) for more on this. 27 Indeed, this issues is closely related to the ultimate source of inflation: a fiscal deficit. The need to finance a fiscal deficit leads to the excessive growth in money supply, which, in turn, causes inflation. In this context, countries that need to finance a fiscal deficit using seigniorage will opt for an exchange rate system consistent with this target-a flexible exchange rate regime. 16 4. Macroeconomic Performance and the Exchange Rate Regime: Stylized Facts from Transition Countries A wide variety of exchange rate regimes has been adopted in transition countries (Figure 1). Not only have the regimes been different, but also in some countries they have changed since the inception of the reforms. A summary of some stylized facts from three pairs of countries operating under alternative exchange rate regimes 1991-98 is shown in Table 1, which draws on the stated commitment of the central bank (as summarized in the IMF's Annual report on Exchange Rate Arrangements and Exchange Rate Restrictions). In other words, it uses a dejure classification based on the publicly stated commitment of the exchange rate instead of a defacto classification based on the observed behavior of the exchange rate. Both classifications have their own shortcomings. A defacto classification has the advantage of being based on observable behavior, but it does not capture the distinction between stable nominal exchange rates resulting from the absence of shocks, and stability that stems from policy actions offsetting shocks. More importantly, it fails to reflect the commitment of the central bank to intervene in the foreign exchange market. Although the de jure classification captures this formal commitment, it falls short of capturing policies inconsistent with the commitment, which, in turn, lead to a collapse or frequent adjustments of the parity. Following Ghosh et al (1997), we classify exchange rate arrangements into three categories: pegged; intermediate; and floating regimes.28 The pegged regimes include single currency pegs, SDR pegs, other published basket pegs, and secret baskets. The intermediate group contains cooperative systems, unclassified floats, and floats with pre-determined ranges. Thefloat group comprises of floats without pre-determined range and pure floats. 28 To this end, we draw on the various issues of the IMF's Exchange Rate Arrangements and Exchange Rate Restrictions. 17 The analysis presented in Table 1 shows that countries with intermediate flexibility had better growth performance, compared to those that pegged and floated. In terms of inflation performance, countries with pegged exchange rates had the lowest inflation, whereas those with floating rates experienced the highest inflation during the period under consideration. Not surprisingly, countries with floating rates had considerably higher monetary growth compare to those with fixed or intermediate regimes-an observation confirming the conventional discipline argument arising from the impact of fixed regimes on the dynamics of money creation. Moreover, countries that pegged or adopted intermediate exchange rate arrangements exhibited noticeably betterfi scal discipline compared to those that adopted floating rates. Countries with fixed exchange rate regime appear to have higher current account deficits compared to those adopting intermediate and flexible regimes. However, once the outlier observation, Azerbaijan (1992), is excluded, countries with flexible exchange rate regimes have higher current account deficits than those with fixed and intermediate regimes. Finally, countries with fixed and intermediate regimes have higher ratios of reserves to base money than those with floating exchange regime. These are, of course, simple observations without controlling for many relevant factors. It is, therefore, not possible to conclude how much of the better macroeconomic performance was in fact due to the particular exchange rate regime adopted and how much was due instead to other important factors. Figure 2 provides more detailed information by presenting the evolution of selected key economic indicators of transition countries operating under alternative exchange rate regimes over the period 1991-98. 18 Table 1. Exchange Rate Regime and Macroeconomic Performance: Transition Economies Pegged Intermediate Flexibility Independent Floating Growth Performance Mean -0.40 0.73 -7.81 Median 3.24 2.30 -8.20 Inflation Performance Mean 71.02 228.12 933.70 Median 14.05 19.50 116.00 Inflation Performance' Mean 0.20 0.26 0.54 Median 0.12 0.16 0.54 Unemplovment Performance Mean 8.97 10.61 9.16 Median 9.65 9.05 8.45 Budeet Balanceb Mean -3.53 -3.55 -9.74 Median -1.90 -3.10 -7.50 Broad Money Growth Mean 38.85 111.04 286.79 Median 20.40 29.10 92.15 Broad Money Growth' Mean 0.22 0.27 0.50 Median 0.17 0.23 0.48 Current Account Deficitb Mean -5.21 4.72 -3.86 Median -5.02 4.40 -7.70 Reserves to Base Money Mean 1.03 1.13 0.73 Median 1.13 0.98 0.60 Source: EBRD, IMF, WB, and authors' calculations. Note: The sample, over the period of 1991-98, consists of Albania, Arnenia, Azerbaijan, Belarus, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakstan, Kyrgyz Republic,Latvia, Lithuania, Macedonia, Moldova, Poland, Romania, Russia, Slovak Republic, Slovenia, Tajikistan, Turkmenistan, and Ukraine. a: To reduce the importance of outliers, the inflation rate (it) is transformed to: n/(1+it). Clearly, as X- oo, the inflation rate will approach to 1. b:Asa%ofGDP c: Similar transformation to reduce the irnportance of outliers for inflation is also performed for broad money growth. 19 Figure 2. Exchange Rate Regime and Macroeconomic Performance: Transition Economies Growth Performance (Mean) Growth Performance (Median) 10 1 0 0. 15 . , - . .. . -15 s .. . . . ~~~~...... . ... ....... . .... .................!.... ...... . . t. s . - h,fl E F.exbIr h,rmucdaisa Flexbity -20 -20 .F..n......... : ....... -25_ . ; . ', '. . -25 1991 1992 1993 1994 1995 1996 1997 1998 1991 1992 1993 1994 1995 1996 1997 1998 Inflation Performance (Mean) Inflation Performance (Median) 10 ID -M Pegg-.d7 fge kirMidWf* exlb4 - lermldiiae F)sx bl4 Os - - aa------- nB ...- O ....... ..... . .... ... ......._ /.... \.. i...... ... . ... ... ...: A' ' ' --OA . ..... ...... ...... .. ......O 02 02 OR~~~~~~~~A 91 92 93 94 95 96 97 98 91 92 93 94 95 96 97 98 Current Account Deficit (Mean)' Current Account Deficit (Median)' kI U Pd xi~ , , . , -b FbxPd 15- _ -oaf 9 .... o .. .. ....... -2 .- , L ...12_.--t. ~... . . ..... .. . . .~~~~~~~~~~~~~~~~~~~~~~.. .... ........ . .. ..... . ..... 10_ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. tS..~ ........ . ....... .. .. -_ . , , , , '. ' . ! , 0 . i~~~~~~~~~~~~~~~~~~~~~~~~ ~ -5_ | | w |-4 1991 1992 1993 1994 1995 1996 1997 1998 1991 1992 1993 1994 1995 1996 1997 1998 a: as a % of GDP. 20 Figure 2. Exchange Rate Regime and Macroeconomic Performance: Transition Economies (Conduded) Reserves to Monetary Base Ratio (Mean) Reserves to Monetary Base Ratio (Median) 16. 15_ r!-t Pegged Pe4qed 1 A _ n h*rrWedie Fiexib.,. re .IA -Flonig --- Floaing 02. 02FT 91 92 93 94 95 96 97 98 91 92 93 94 95 96 97 98 Budget Deficit (Mean) Budget Deficit (Median) 20 20 Pegged . :1 l Pegged _ . hbrrnadiae FbxibIitiec . . . - h rnads Flexibil Fbating. Floahng 1 . 5 - - - - - 10_ . :'8%s ... 10.:,,""" . 5 . 5 0 0 1991 1992 1993 1994 1995 1996 1997 1998 1991 1992 1993 1994 1995 1996 1997 1998 Government Expenditures (Mean) ' Government Expenditures (Median) a 60_ 60. ! ! t~~~~~~~:l 1ggted r -t- Fegged 55 _nk- j .hrmedieb Fexibiity 55 _ ,trrrnedi Ftxibity _ . ......... ~~~~~- -- - ---__ ttrg_.......-------- ---_ - Fbing 45 - V45 ! .. 40 ~~~~~~~~~~~~~~~~~~~~40 1991 1992 1993 1994 1995 1996 1997 1998 1991 1992 1993 1994 1995 1996 1997 1998 a: as a % of GDP. 21 From this analysis, one can identify at least five key patterns from the behavior of the macroeconomic variables in question. First, countries with intermediate regimes appear to have experienced smaller contractions of output and faster recovery. Second, the lowest inflation throughout the period under consideration is observed in countries operating under fixed regime. However, the difference between inflation under floating and intermediate regimes tapers off overtime. Third, countries with floating regimes clearly experience the highest budget deficit compared to those operating under fixed or intermediate regimes. The fiscal performance of countries with fixed regimes, however, is not noticeably better that those with intermediate regimes. It is interesting to note that the relatively poor fiscal performance of countries with flexible regime arises from weak revenue collection not from excessive spending. This could be due to the so called "reverse Tanzi effect" arising from increases in tax collection caused by immediately stabilizing prices-a phenomenon observed in successful exchange-rate based stabilization programs. Fourth, contrary to the experience of other emerging markets where the current account deficit tends to be higher in countries with more stringent exchange regimes, it appears that, in the case of transition economies, countries with floating regimes experience, on average, higher current account deficits. Finally, it appears that the ratio of international reserves to monetary base tends to be somewhat lower in floating countries, though the difference is not very significant. 22 5. The Empirical Framework It may be possible to underpin some of these stylized facts from a cursory look at the evolution of selected key economic indicators under alternative regimes. However, it is not possible to identify the independent effects of the nominal exchange rate regime on economic performance without a thorough analysis in which macroeconomic/financial fundamentals and institutional arrangements-affecting both economic performance and the choice of the exchange rate regime-are controlled for. In an attempt to examine the impact of the exchange rate regime on macroeconomic performance, empirical studies often employ exchange rate dummies in reduced form equations for inflation and growth. The coefficient estimate of a particular exchange regime dummy is, in turn, deemed to reveal the effect of the exchange rate arrangement on the dependent variable. One of the major drawbacks of this approach is that at the time of the regime switch the coefficients associated with policy variables also change-a phenomenon referred to as the Lucas critique. One approach to avoid this problem is to estimate each equation representing different exchange rate regimes separately and then to test for the equality of coefficients. This approach, however, would fail to capture the causal link between macroeconomic fundamentals and the exchange rate regime-the ability of an economy and also policymakers' desire to implement certain exchange rate regimes under given fundamentals. Moreover, existing studies, to the best of our knowledge, fail to address the issue of the sample selection problem. The sample selection problem arises from the fact that countries do not choose their exchange rate regimes randomly. Instead, their choice hinges on a set of fundamentals, which, in turn, affects macroeconomic outcomes such as inflation and growth. Consequently, the use of standard econometric techniques such as OLS or 2SLS will produce 23 biased results stemming from the correlation between the regime choice and the error term either 29 in the inflation or growth equation. It should be noted that addressing the sample selection problem will also address the issue of the endogeneity of the choice of the exchange rate regime. This is not achieved by instrumenting the dummy variable for the exchange rate regime a la Ghosh (1997). Instead, it is achieved through the assumption of constant covariance between the error term in the structural equation and the normally distributed random variable whose realization determines the exchange rate regime. In an attempt to address the above mentioned problems plaguing empirical work in this literature, we propose an empirical framework which is based on a switching regression technique. To this end, the investigation employs the following standard formulation of switching regression: Yi = XiBj+uj if Vi < Zjr+al, i=l ................... I, (1) Yi=XiB2+u21 ifZ1y+zaZiY+ 0f2, i=l 1...................I3 (3) ujj is iidN(O, j),while vi is iidN(O, 1), cov(ui,vi)=avj=1,2,3 where (1), (2), and (3) correspond to respective regimes. The only difference with respect to the standard switching regression model is that we employ the same set of regressors in each equation in order to be able to test the equality of the coefficients across the regimes. The regime is determined by the realization of normally distributed random variable v, which is not observable. We, however, know in which of three areas it is realized. Therefore, a>, a2, and y 29 To be more precise, this bias arises from the correlation between the error term of the latent variable capturing the regime choice and the error of the structural equation. 24 can be estimated by ordered probit approach. It should be noted that Z should not contain a constant term since a, and a2 are already in the model. Given the following equations: E(u,, vi < Z1Y + a,) = -'-' F(Z +a,) = -c1(Zir + a,) = -avli (4) E(ujl vi > Ziy + a2) = C3V - f(Z,y + a2) = a3A3(Zi7+a2) = C3v3i( 1_-F(Zy7+ a2) (5) E(u11 IyZ +a, F' ' 12 1 12 l FC2J-F(61i) where h2, - F ) , = ZaY-+ &,, c2U = Zi t + a2, e2, are residuals obtained from the second step Heckman procedure along with 62&, , , &2and y are coming from the ordered probit regression in the first stage, 12 is the number of observations in the second regime. Next, we derive the assymptotic VC matrix of the estimated parameters. Drawing on Heckman (1979), we can write: 40 2 02 )4N(O, B WV), where 58 B=piimIX2I X2 X2h = milxx2 X j2 B = plimI2( 22 ,2)=plimI2( 2 2) I,I2 -OD k2X2 ^2k lkZe ^X2 h2h2 Y = p 1im(1 + 2) 1,1, 4C0 = a2 (XV X Xi hj) it hiX2i h2i2 0 C22vE L X2iX2j' X2i2jS II, kh2X2j h21h21) E2 2 cbf(c f()2- Ee22 U22 2_ Clif(Cli)-ci(2) 7i = C2 = 1 42 - F(c2j) - F(c1i) a~~~k akh2Z Ahi8210 Dh2j Z ah2jf{h2A Laa, aa2\ ay J L O; aa2c ar' Oh2, (c21) (h2, -CA2 la2 F(C21) - F(cl1) - -k i c1f(c2i) - hc2if(c2i) Oac2 F(c21) - F(cli) aha, _ /2_ c1jf(c2,) - c2f(c2i) ar F(C21)-F(cli) In the above expressions Q is a variance-covariance matrix of (&i &2 ,r), I is the total number of observations. Expression in square brackets is a stacked vector. 59 Appendix 3: Simulation Exercise In the simulation exercise, we place the values of right hand side variables of one regime into the structural equation of another. The resulting value would be the expected inflation if the country were to run the other regime. It should be kept in mind that when conducting this kind of simulations the tern associated with the covariance in the simulated equation should be substituted by the corresponding term from the original equation. For instance, if we were to simulate intermediate countries to find our how they would perform under fix regime, the original equations for fix and intermediate regimes are the following: E(Yj I Xi, Zi, fix) = XAh, - , (Z,i + al,) E(Yj X, Zj,interm.) = XjB2 + 52, f(Zj +a,)-f(Zj9+ 2) F(Zj + &2) -F(Z1jp+ t,) where observations denoted by the notation i represent fix regimes, while those withj stand for intermediate arrangements. Now, we substitute the intermediate observations into the equation for fix and obtain the simulated inflation from the following equation: kE(Yj I Xj,Zj,fix) = XjBI - vFZ +a)-(Zj+a Note that after covariance, we have a term from coming form the original equation for intermediate regime. Substituting only Zj into the "fix" equation would be incorrect. Finally, from the last expression, we obtain the expected values of inflation or growth if a country under intermediate regime were to run fix. 60 Appendix 4: Description of Data Data on GDP growth, inflation, budget deficit are obtained from various EBRD reports. Data on international reserves and monetary base were taken from IFS. Unfortunately, for the majority of transition countries financial statistics in the beginning of transition were unavailable. Consequently, those observations could not be included in the regression analysis. Liberalization indices are taken from De Melo et al and updated for 1998 from the EBRD report. Classification of the exchange rate regimes is obtained from the IMF's Exchange Rate Arrangements and Exchange Rate Restrictions. Moreover, observations corresponding to the year 1997 for Bulgaria were excluded on the grounds that Bulgaria accepted currency board in the middle of the year while experiencing extremely high inflation prior to the month and, as a result, for the entire year. Since this particular year, 1997, would be treated as a fixed regime in the annual data, it would become an obvious outlier. We also excluded years of war and severe regional conflicts involving the following countries: Armnenia, Azerbaijan, Georgia, Moldova, Macedonia, Croatia. In addition, Tajikistan, Turkmenistan and Uzbekistan were excluded from the sample due to data problems. Time period for Czech Republic and Slovak republic was considered after they separated to avoid inconsistency in the data. The countries and the periods for the regression analysis are the followings: Albania 1993- 98, Armenia 1995-98, Azerbaijan 1995-98, Belarus 1995-98, Bulgaria 1995-98, Croatia 1995-98, Czech Republic 1994-97, Estonia 1993-98, Georgia 1994-98, Hungary 1992-98, Kazakhstan 1994-98, Kyrgyz Republic 1994-98, Latvia 1994-98, Lithuania 1994-98, Macedonia 1996-98, Moldova 1993-98, Poland 1992-98, Romania 1992-98, Russia 1995-98, Slovak Republic 1994- 98, Slovenia 1993-98, and Ukraine 1994-98. The series employed in the regression analysis contains 24, 54, and 35 observations under fixed, intermediate, and floating regimes, respectively. 61 References Aghelvi, B., M. Khan, and P. Montiel (1993), Exchange Rate Policy in Developing Countries: Some Analytical Issues, IMF Occasional Paper, 78, Washington: International Monetary Fund. Aizeman, J. (1991), Foreign Direct Investment, Productive Capacity and Exchange Rate Regimes, NBER Working Paper, No. 78. Baxter, M. and A. Stockman (1989), Business Cycles and the Exchange Rate-Regime: Some International Evidence, Journal of Monetary Economics, vol. 23, pp. 377-400. Backe, P. (1999), Exchange Rate Regimes in Central and Eastern Europe: A Brief Review of Recent Changes, Current Issues and Future Challenges, Focus on Transition 2/1999, ONB. Begg, D. (1997), Monetary policy during transition: progress and pitfalls in Central and Eastern Europe, 1990-6, Oxford Review of Economic Policy (U.K.), vol.13. pp. 33-46. , (1966), Monetary policy in Central and Eastern Europe :Lessons after halfa decade of transition, IMF Working Paper, WP/96/106. Berg, Anndrew, Eduardo Borensztein, Ratna Sahay, and Jeromin Zettelmeyer (1999), The Evolution of Output in Transition in Central and Eastern Europe, IMF Working Paper, 99/73, Washington: International Monetary Fund. Calvo, G. (1 999a), Testimony on Full Dollarization, presented before a Joint Hearing of the Subcommittees on Economic Policy and International Trade and Finance, U.S. Congress, April. , (1999b), Fixed versus Flexible Exchange Rates: Preliminaries of a Turn-of- Millennium Rematch, www.bsos.umd.edulecon/ciecalvo.htm. , (1998), Capital Flows and Capital Market Crises: The Simple Economics of Sudden Stops, Journal of Applied Economics, vol. 1, pp. 35-54. and C. Reinhart (2000a), Fear of Floating, mimeo, University of Maryland. and C. Reinhart (2000b), Fixingfor Your Life, mimeo, University of Maryland. Christoffersen, Peter, Peter Doyle (1998), From Inflation to Growth: Eight Years of Transition, IMF Working Paper, 98/100, Washington: International Monetary Fund. Corker, Robert, Craig Beaumont, Rachel van Elkan, and Dora Iakova (2000) Exchange Rate Regimes in Selected Advanced Transition Economies - Coping with Transition, Capital Inflows, and EUAccession, IMF Policy Discussion Paper, 00/3 Cooper, Richard (1999), Exchange Rate Choices, mimeo, Harvard University. Crockett, A. (1997), Why is Financial Stability a Goal of Public Policy?, in Maintaining Financial Stability in a Global Economy, Federal Reserve Bank of Kansas City, Jackson Hole Wyoming, August 28-30, 1997. De Melo Martha, Cevdet Denizer, Alan Gelb. (1997) From Plan to Market : Patterns of Transition, World Bank Economic Review, Vol. 10, Washington: The World Bank. 62 Doma9, I. and M.S. Martinez Peria (2000), Banking Crises and Exchange Rate Regimes: Is There a Link?, World Bank Policy Research Working Paper, No. 2489. Dornbusch, Rudiger (1994), Exchange Rate Policies in Economies in Transition, in Approaches to Exchange Rate Policy Choices for Developing and Transition Economies, edited by R. Barth and C. Wong. Dreyer, J.S. (1978), Determinants of Exchange Regimes for Currencies of Developing Countries: Some Preliminary Results, World Development, vol.6, No. 2, pp. 437-445. Edwards, Sebastian (1996), The Determinants of the Choice Between Fixed and Flexible Exchange-Rate Regimes, NBER Working Paper, No: 5756. (1993), Exchange Rates as Nominal Anchors, Weltwirtschaftliches Archiv, Vol. 129, No 1, 1-32. Edwards, S. and M. Savastano (1999), Exchange Rates in Emerging Economies: What do we know? "hat do we need to know?, NBER Working Paoer, No: 7228. Eichengreen, B. and Paul Masson (1998), Exit Strategies: Policy Options for Countries Seeking Greater Exchange Rate Flexibility, IMF Occasional Paper, No: 168. Fischer, Stanley, (2001), Exchange Rate Regimes: Is the Bipolar View Correct?, IMF, mimeo. Fischer, Stanley, Ratna Sahay, and Carlos A. Vegh (1997), From Transtion to Market: Evidence and Growth Prospects. In S. Zecchini (eds), Lessons from the Economic Transition, Kluwer Academic Publishers. Fischer, Stanley, Ratna Sahay, and Carlos A. Vegh (1996a), Stabilization and growth in transition economies : the early experience. Journal of Economic Perspectives, Vol. 10, No.2, Spring, 45-66. (1996b), Economies in Transition: The Beginning of Growth . AEA Papers and Proceedings, Vol. 86, No.2, 229-233. Fischer, Stanley, and Ratna Sahay (2000), The Transition Economies After Ten Years. IMF Working Paper, 00/30. Frankel, J. (1995), Monetary Regime Choice for a Semi-open Country, in Capital Controls, Exchange Rates and Monetary Policy in the World Economy, edited by S. Edwards. Ghosh, A. A. Gulde and J. Ostry (1997), Does the Nominal Exchange Rate Regime Matter?, NBER Working Paper, No: 5874. Goldstein, M. (1999), Safeguarding Prosperity in a Global Financial System: The Future International Financial Architecture Report of an Independent Task Force, Washington DC: Institute for International Economics for the Council on Foreign Relations. Guitian, M. (1994), The Choice of an Exchange Rate Regime, in Approaches to Exchange Rate Policy Choices for Developing and Transition Economies, edited by R. Barth and C. Wong. Hanke, S. and K. Schuller (1999), A Dollarization Blue Print for Argentina, CATO Institute, February. 63 Hausmann, R, M. Gavin, C. Pages-Serra, and E. Stein (1999), "Financial Turmoil and the Choice of Exchange Rate Regime," Inter-American Development Bank, mimeo. Havrylyshyn, Oleh, Ivailo Izvorski, and Ron van Rooden (1998), Recovery and Growth in Transition Economies, IMF Working Paper, 98/141, Washington: International Monetary Fund. Heckman, J. (1979), Sample Selection Bias as a Specification Error, Econometrica, vol. 47. Pp. 153-161. Heller, H.R. (1978), Determinants of Exchange Rate Practices, Journal of Money, Credit, and Banking, vol. 10, pp. 308-32 1. Hernandez-Cata, Ernesto (1997), Liberalization and the Behavior of Output During the Transition from Plan to Market, IMF Working Paper, 97/53, Washington: International Monetary Fund. Heybey, Berta, and Peter Murrell (1999), The Relationship between Economic Growth and the Speed of Liberalization during Transition, Journal of Policy Reform, Vol. 3, 121 - 137. Holden, Merle, Paul Holden (1981), Policy objectives, country characteristics and the choice of exchange rate regime. International Review of Economics and Business (Italy); 28:1001-14 Oct./Nov. 1981 Honkapohja, S. and P. Pikkarainen (1994), Country Characteristics and the choice of the Exchange Rate Regime: Are mini-skirts followed by maxis? in Exchange Rate Polices in the Nordic Countries, CEPR, London, edited by Akerholm and Giovannini, pp. 31-53. International Monetary Fund (1997), Exchange Rate Arrangements and Economic Performance in Developing Countries, Chapter 4 of the World Economic Outlook, Washington, D.C., October 1997. (1997), Fixed or Flexible? Getting the Exchange Rate Right in the 1990s. (1998), World Economic Outlook, May. (2000), Exchange Rate Regimes in an Increasingly Integrated World Economy, April. Klaus, V. (1997), Promoting Financial Stability in the Transition Economies of Central and Eastern Europe, in Maintaining Financial Stability in a Global Economy, Federal Reserve Bank of Kansas City, Jackson Hole Wyoming, August 28-30, 1997. Levy-Yeyati, E. and F. Sturzenegger (1999), Classifying Exchange Rate Regimes: Deeds vs. Words, mimeo, Universidad Torcuato Di Tella. Little, I., R. Cooper, W. Corden, and S. Rajapatirana (1993), Boom, Crisis, and Adjustment: The Macroeconomic Experience of Developing Countries, Oxford University Press for The World Bank. Masson, P. R. (1999), Monetary and Exchange Rate Policy of Transition Economies of Central and Eastern Europe after the Launch of EMU, IMF Policy Discussion Paper 99/5. 64 Mills, T. and G.E. Wood (1993), Does the Exchange Rate Regime Affect the Economy?, Federal Reserve Bank of St. Louis Review, July/August, pp. 3-20. Mishkin, F. and M. Savastano (2000), Monetary Policy Strategies for Latin America, NBER Working Paper. No: 7617. Obstfeld, M. and K. Rogoff (1995), The Mirage of Fixed Exchange Rates, Journal of Economic Perspectives, vol.9., pp. 73-96. Popov, V. (2000), Shock therapy versus gradualism: The end of the debate (explaining the magnitude of transformational recession),Comparative Economic Studies; Spring 2000, pp. 1-57 Quirk, P. (1994), Fixed or Floating Exchange Rate Regimes: Does it Matter for Inflation?, IMF Working Paper, No: 143. Radzyner, 0. and S. Riesinger (1997), Exchange Rate Policy in Transition-Developments and Challenges in Central and eastern Europe, Focus on Transition, Oesterreichische Nationalbank. Rizzo, Jean-Marc (1998), The economic determinants of the choice of an exchange rate regime: a probit analysis. Economics Letters (Netherlands!; 59, No. 3:283-87, June 1998. Sachs, J. (1996), Economic Transition and the Exchange Rate Regime, AER Papers and Proceedings, vol. 86, No. 2, pp. 147-152. Sahay Ratna, Carlos Vegh (1996) Inflation Stabilization in Transition Economies: An Analytical Overview. Journal of Policy Reform, Vol 1, 75-198. Selowsky, Marcelo, and Ricardo Martin (1997), The Transition From Socialism: Policy Performance and Output Growth in the Transition Economies, AEA Papers and Proceedings, Vol. 87, No. 2, 350-353. Tornell, A. and A. Velasco (1999), Fixed versus flexible Exchange Rates: Which Provides More fiscal Discipline?, Forthcoming, Journal of Monetarv Economics. (1998), Fiscal Discipline and the Choice of a Nominal Anchor in stabilization, Journal of International Economics vol. 46, pp. 1-30. (1995),. Fixed Versus Flexible Exchange Rates : Which Provides More Fiscal Discipline? NBER Working Paper, no. 5108 Vegh, Carlos (1992), Stopping High Inflation: An Analytical Overview, IMF Staff Papers, 39, 626-695. Velasco, A. (1999), Exchange Rate Polices for developing Countries: What have We Learned? What do We still Not Know?, Studies on International Monetary and Financial Issues for the Group of Twenty-four. _and L. F. Cespedes (1999), Exchange Rate Arrangements: A Developing Country Perspective, unpublished manuscript. 65 Policy Research Working Paper Series Contact Title Author Date for paper WPS2628 Monopoly Power and Distribution in Hanan G. Jacoby June 2001 P. Kokila Fragmented Markets: The Case of Rinku Murgai 33716 Groundwater Saeed Ur Rehman WPS2629 Bridging the Digital Divide: How George R. G. Clarke July 2001 P. Sintim-Aboagye Enterprise Ownership and Foreign 37644 Competition Affect Internet Access in Eastern Europe and Central Asia WPS2630 Parallel Imports of Pharmaceutical Mattias Ganslandt July 2001 L. Tabada Products in the European Union 36896 WPS2631 Pension Reform in Hungary: Roberta Rocha July 2001 L. Gross A Preliminary Assessment Dimitri Vittas 37030 WPS2632 Human Capital and Growth: The Sebastien Dessus July 2001 K. Mazo Recovered Role of Education Systems 39744 WPS2633 Bank Privatization in Argentina: George R. G. Clarke July 2001 P. Sintim-Aboagye A Model of Political Constraints and Robert Cull 37644 Differential Outcomes WPS2634 Chile's Regional Arrangements and Glenn W. Harrison July 2001 L. Tabada the Free Trade Agreement of the Thomas F. Rutherford 36896 Americas: The Importance of Market David G. Tarr Access WPS2635 Optimal Use of Carbon Sequestration Franck Lecocq July 2001 V. Soukhanov in a Global Climate Change Strategy: Kenneth Chomitz 35721 Is There a Wooden Bridge to a Clean Energy Future? WPS2636 Processes, Information, and Javier Campos-Mendez July 2001 G. Chenet-Smith Accounting Gaps in the Regulation Antonio Estache 36370 of Argentina's Private Railways Lourdes Trujillo WPS2637 Technical Efficiency Gains from Antonio Estache July 2001 G. Chenet-Smith Port Reform: The Potential for Marianela Gonzalez 36370 Yardstick Competition in Mexico Lourdes Trujillo WPS2638 On Financing Global and International Todd Sandler July 2001 S. Kpundeh Public Goods WPS2639 Public Policy toward Nongovernmental William Jack July 2001 H. Sladovich Organizations in Developing Countries 37698 WPS2640 Where Has All the Foreign Investment Harry G. Broadman July 2001 S. Craig Gone in Russia? Francesca Recanatini 33160 Policy Research Working Paper Series Contact Title Author Date for paper WPS2641 Is Russia Restructuring? New Harry G. Broadman July 2001 S. Craig Evidence on Job Creation and Francesca Recanatini 33160 Destruction