Exchange Rate Regimes and the Stability of Trade Policy in Transition Economies

Zdenek Drabek World Trade Organization Centre William Reppard CH-1211 Geneve 21 Switzerland FAX: 41 22 739 5762 e-mail: zdenek.drabek@wto.org ABSTRACT This paper examines the interplay b

Staff Working Paper ERAD-98-07 July, 1998

World Trade Organization

Economic Research and Analysis Division

Josef C Brada: Arizona State University

Disclaimer: This is a working paper, and hence it represents research in progress This paper represents the

opinions of individual staff members or visiting scholars, and is the product of professional research It is notmeant to represent the position or opinions of the WTO or its Members, nor the official position of any staffmembers Any errors are the fault of the authors Copies of working papers can be requested from the divisionalsecretariat by writing to: Economic Research and Analysis Division, World Trade Organization, rue de Lausanne

154, CH-1211 Genéve 21, Switzerland Please request papers by number and title

Exchange Rate Regimes and the Stability of Trade

Policy in Transition Economies

Exchange Rate Regimes and the Stability of

Zdenek Drabek World Trade Organization Geneva, Switzerland

Josef C Brada Arizona State University Tempe, AZ 85287-3806 Running Head: Exchange Regimes and Trade Policy

Proofs to: Dr Zdenek Drabek

World Trade Organization Centre William Reppard

CH-1211 Geneve 21 Switzerland

FAX: 41 22 739 5762 e-mail: zdenek.drabek@wto.org

ABSTRACT

This paper examines the interplay between exchange rate regimes and policies and commercial policy in six transition economies In all these economies the rate of protection afforded domestic industry by the exchange rate has been eroded by high rates of inflation and insufficient growth in productivity As a result, there has been pressure on governments to increase trade barriers and each country examined has had recourse to various means of restricting imports We argue that more flexible management of the nominal exchange rate would be a preferable way of dealing with the real appreciation of these countries’ currencies Key Words: exchange rates, trade policy, transition economies

JEL Classification Nos [P33]; [F13]; [F33]

Another alternative is a crawling peg that maintains the real effective exchange rate yet imposessome discipline on domestic monetary policy However, the experience with crawling pegs is also notentirely without problems The crawling pegs may not be flexible enough to completely eliminate theproblems of fixed exchange rates in an inflationary environment For example, Obstfeld (1984) argues,using the example of Argentina, Chile and Uruguay, that the choice of the crawling-peg exchange rateregime in these countries was not effective because it produced dramatic, and ultimately unsustainable,current account imbalances and real exchange rate appreciation One reason was slow labor marketadjustments Thus, once imperfections are introduced into the functioning of product or factor markets,policy prescriptions become more complex Another problem with crawling-peg regimes has been theselection of currency baskets, which introduces an element of arbitrariness into the process of setting thelevel of exchange rate For peggers, the choice of a reference basket of currencies involves decisions that aredependent on trade concentration, the degree of market openness, the size of the country and various otherindicators.6

The literature on the choice of exchange rate regimes provides more guidance regard to the choice

of exchange rate regime in the context of macroeconomic policies, although it touches the relationshipbetween exchange rate regimes and commercial policies only indirectly We can identify two aspects ofmacroeconomic policy making that lead to increased pressures for protection These include governmentpolicies leading to slow economic growth and rising unemployment and those policies relating to inflation

There has been a growing interest in the profession regarding choice of the exchange rate regime as

an instrument of growth-promoting policies Most recently, the discussion has focused on the role ofexchange rate polices in transition and, more specifically, on the role of these policies in stimulating thegrowth of domestic demand and of exports It appears that a consensus is being reached that transitioneconomies with stable exchange rates have arrested and reversed output declines much more quickly thandid those countries that pursued policies of flexible exchange rates However, and equally important, stableexchange rate policies have turned out to be detrimental to the growth of output on a sustained basis Both

of these points have been made forcefully by Sachs (1996), who explains the impediments to economic

growth generated by pegged exchange rates as stemming from various market rigidities in the transitioneconomies that preclude a flexible domestic response to changes in relative prices.8

Pegged exchange rates also have tended to stimulate domestic demand for consumer goods andthus contributed to overheating, which itself may have had different origins For example, the growth ofdomestic demand in the Czech Republic accelerated in the early 1990's following a rapid rise in wages.However, the concurrent sharp appreciation in the real effective exchange rate has added considerably to thegrowth of domestic demand for imported consumer goods Moreover, the pegged exchange rate combinedwith inflation higher than abroad has forced up the level of interest rates, thus attracting foreign capital,which also stimulated the growth of domestic aggregate demand

None, of this, of course, tells us anything directly about changes in commercial policy It does say,however, that exchange rate policy can stimulate or retard domestic growth and thus affect the intensity ofprotectionists pressures Rapid growth of domestic output is likely to reduce calls for extra protection whileslow growth is likely to stimulate them If growth is not accompanied by balance of payments difficulties

or with a rise in unemployment, protectionist pressures are likely to be relatively small

There is also a growing consensus that stable exchange rates have performed an extremely usefulrole in stabilizing transition economies (Sachs 1996), although the evidence is not conclusive The inability

to suppress inflation to the level of West Europe and, as a result, the inability to lower the level of interestrates and to reduce foreign capital inflows have intensified inflationary pressures in some transitioneconomies At the same time, high interest rates have facilitated the financing of current account deficits.Thus, the pressures for additional protection from a more flexible exchange rate policy have been relativelymild This has tended to delay important policy decisions

Finally, other factors also play a role For example, Ades et al (1993) argue that the initial rate of

inflation matters They find that exchange rate-based stabilizations have been relatively less successful inhigh-inflation countries where initial booms have been followed by severe recessions The impact has beenthe opposite in low inflation countries that started with recessions but were able to recover later In eithercase, exchange-rate-based stabilizations have been associated with recessions and thus with increased callsfor protection.9

3 Empirical Evidence

While the potential sample of transition economies is quite large, we focus our analysis on sixcountries Four of these, the Czech Republic, Hungary, Poland and Slovankia have been among the betterperformers in terms of economic liberalization and stabilization The remaining two, Bulgaria andRomania, illustrate the experience of countries less successful along these lines The former four countriesare considered to have a trade regime based on “standards and performance norms of advancedindustrialized countries."10 For this sample of countries, we seek to show to what extent commercial policychanges, particularly in the form of increased protectionism, have been used as a substitute for exchangerate realignment We do not expect a one-to-one correspondence between exchange rate appreciation and atightening of commercial policy Rather, we expect that only when and if appreciation of the exchange rateresults in the deterioration of trade performance will authorities be tempted, or pressured, to consider arevision of commercial policy.The mirror image of increased protection through higher tariffs is a change inthe exchange rate policy An increased REER will reduce domestic protection and reduced REER willincrease it It follows, therefore, that devaluation of the nominal exchange rate will increase domesticprotection

Before turning to the empirical findings, the remaining issue to be addressed is the question whatkind of REER is relevant for our comparisons REER can be computed with either producer prices orconsumer prices in the denominator If both prices indices moved more or less in parallel, the issue wouldmoot However, this is often not the case, and the discrepancies are usually very large in the transitioneconomies.11 For this reason we use both measures We have chosen 1992 as the base period to measurethe changes in REER The reason for not selecting an earlier year is that many price liberalization measureswere not completed until the end of 1991 Using the earlier years, therefore, would be meaningless In order

to demonstrate the emergence of protectionist pressures, however, we do report some measures taken bythese countries already during 1990-1992

It should be also noted that we avoid as much as possible any exchange rates as undervalued orovervalued Undervaluation and overvaluation remain a theoretical concept that is approximated in practice

almost exclusively by considering the deviations of actual exchange rates from purchasing power parity.The relevance of the latter as a yardstick for equilibrium is highly dubious The evidence regarding thevalidity of PPP as a standard of equilibrium in the external balance shows that deviations of actualexchange rates from PPP-based rates are more the exception than the rule.1Alternative measures such aschanges in REER are partial-equilibrium concepts that are fully suitable for our purposes.

Real exchange rates for the sample countries are reported in Table 1 The Czech Republic provides

a particularly dramatic example of the effects of an excessively rigid exchange rate policy Between 1992and 1997, the REER appreciated by 36 percent on the basis of producer prices and by 54 percent on thebasis of consumer prices This means that the protection to Czech industries provided by exchange ratepolicy declined by the corresponding amounts during the period under consideration, assuming that was nocommensurate improvement in fundamentals in particular through the growth of productivity The latter hasnot been the case, (Begg, 1998).13 Thus, our conclusion is that Czech exchange rate policy led to asubstantial decline in the protection afforded the country's industries

A similar pattern can be observed for the other countries in our sample Each has seen its REERappreciate Using the producer price-based indices, the most serious appreciation has taken place inSlovakia followed by Romania and Bulgaria In contrast, both Poland and Hungary have experienced arelatively mild appreciation of their REER.14 The main reason for the relatively better performance of theREER in Poland and Hungary is that their exchange rate policies have reflected, to a greater extent,domestic rates of inflation This contrasts not only with the performance of the other countries but also, inthe case of Poland, with the policies pursued during 1990-1992 when a significant appreciation of theREER took place.15 Using the consumer price-based indices, the most serious appreciation of REER tookplace in Bulgaria and Romania, with Slovakia a close third Once again, both Poland and Hungary haveexperienced more modest appreciation

Thus, protection of industries afforded by the level of the exchange rate declined dramatically in theCzech Republic, Slovakia, Bulgaria and Romania in the brief period following the reforms of the early1990's The question is whether this has been reflected in worsening trade performance or in attempts bythese countries to offset the effect of their exchange rate policies through changes in commercial policies

Trade performance as measured by the changes in the trade balance is as expected, with a dramaticdeterioration in the trade balance, particularly in the countries with the sharpest appreciation in the REER.Following a brief period of relatively balanced trade, the Czech Republic experienced a sharp deterioration

in its trade balance in 1994 that continued until the end of 1996 A similar pattern can be observed forPoland Slovakia maintained a relative trade balance until the end of 1995 when a trade deficit began toemerge and then deteriorated further in 1997 Romania's trade position also deteriorated sharply in thecourse of 1996, and, by mid-1997, the country was running a large and growing trade deficit Hungary hasmaintained a fairly stable trade deficit throughout the most recent period with a slight improvement at theend of the period Bulgaria's trade balance has been considerably influenced by the financial crisis of thesecond half of 1996, which lead to a collapse of imports By the end of the first quarter of 1997, theannualized import level declined by almost 27 per cent in current dollar terms

The link between changes in REER and trade performance is very tight in some countries includingthe Czech Republic and Slovakia; in others the relationship is more tenuous Because the impact ofexchange rate changes on trade flows is felt with a time lag, the imprecise nature of the relationship is notsurprising Moreover, the relationship was probably also influenced by the speed with whichcompetitiveness was lost in individual countries, and this, in turn, was also a function of productivityimprovements, which depended crucially on the success of industrial restructuring This point is particularlyevident in the performance of Hungary The relatively stable REER in Hungary has been associated with afairly constant level of trade (im)balance and, most recently, with some improvement in the trade balance.This improvement is particularly impressive in the light of the fact that the initial devaluation in Hungarywas the least dramatic in the whole region The case of Poland is somewhat puzzling; the deterioration inthe trade balance has been associated with a fairly constant level of the REER measured by producer prices,although the REER measured by consumer prices has been sharply rising, perhaps suggesting that the lattershould have been the basis for the crawling peg mechanism if a stable trade balance was the policyobjective

Initial devaluations have not been decisive in maintaining a sustainable trade balance The sharpestdevaluations took place in the Czech Republic and Slovakia and in Poland between 1989 and the end of

1991 Their currencies were devalued by about 43 percent and by 50 percent respectively Hungarydevalued by only 11 percent during the same period While these devaluations were clearly helpful in theearly phase of the reform by stimulating trade reorientation, the competitive advantage they conferred wasquickly lost because they were not supported by measures leading to a rapid improvements in productivity.

In sum, exchange rate policy has influenced trade performance A loss of competitiveness caused by anappreciation in REER is associated with a deterioration in the trade balance The appreciation in REER was

to some extent offset by improvements in productivity but only partially

In order to see the countries' responses to the deteriorations of their trade performance, we turn totheir trade policies The most salient features of the policies are summarized in Table 2 We can see fromthe table that all of these countries have experienced not only instability in their exchange rate policies butalso an instability in commercial policy In the case of the Czech Republic, the appreciating REER hascontributed to the emergence of a rapidly deteriorating trade and current account imbalance and to the needfor an introduction of import restrictions in April, 1997 These restrictions took the form of foreignexchange deposits amounting to 20 percent Under pressure from the European Union and from other WTOmember countries, the Czech government eliminated the deposits starting from September 1997 In themeantime, the Czech crown was devalued, and this relieved the competitive pressures on Czech industry

By July 1997, the REER had dropped by 10 percent from its peak in March 1997, moving it back to thelevel of about July 1996

The experience of other countries is very similar Bulgaria introduced an import surcharge in June

1996, and Romania has maintained an import surcharge since May 1992 The Slovak Republic introduced

a 10 percent import surcharge in March 1994 that it eliminated by June 1996, but it re-introduced thesurcharge, amounting to 7 percent, in July 1997 Hungary and Poland did not avoid these additional importrestrictions either; Hungary introduced an import surcharge in March 1995 and Poland in December 1992.Poland eliminated the surcharge by January 1997 and Hungary by July 1997 With the exception of theCzech Republic, over time, each of these countries has also introduced measures affecting particularindustries For example, Slovakia introduced so-called quality certificates and import deposits, and Polandadopted measures to protect its car industry Poland and Hungary also introduced measures to protect

agriculture Hungary has global quota on consumer goods imports, and Poland introduced import quotas oncertain electronic products Romania introduced a system of reference prices for six products in 1994 toprotect domestic producers These were abolished in July 1995 when a new tariff was introduced

We do not claim to demonstrate a causal relationship between the introduction of the protectionistmeasures and exchange rate policies Nevertheless, the timing of the protectionist measures described inTable 2 is more than coincidental As Chart 2 shows, the import-restrictive measures were introduced in alltransition economies at the time when the trade deficit increased to levels that were menacing At the sametime, the deteriorating trade imbalance was also closely connected to appreciating real effective exchangerates, as noted above In these cases, the countries' international competitiveness had suffered, causingincreasing pressures for an increase in import restrictions

4 The Czech Case

It follows from the previous discussion that protectionist pressures arise from three aspects ofexchange rate policies; from an appreciating exchange rate that is not offset by an increase in productivity,from a level of the exchange rate that is neither conducive to international competitiveness nor to economicgrowth; and from a rigid exchange rate policy that encourages speculative inflows of foreign capital.Empirical assessments of the appropriate level of the exchange rate are always difficult to make, but there is

a growing consensus among domestic and foreign observers that Czech exchange rate policy has failed onall three accounts This represents a dramatic switch in the opinion of the majority of economists who untilrecently have praised the Czech experience as an example of a well-managed system.16

(i) The Exchange Rate Level There has been a great deal of interest in the literature and in policycircles whether the exchange rate of the Czech Republic has been undervalued or overvalued.17The empirical questions remain, of course, speculative since the existence of differences betweenthe actual exchange rate and the equilibrium exchange rate cannot be demonstrated conclusively.18(ii) However, an answer can be provided by looking at the performance of the current account, andthe figures have been quite revealing, a dramatic deterioration in both the trade and current accountbetween 1994 and 1997

Despite the conceptual problems involved, some observers have attempted to calculate thedistortion of the actual exchange rate The conclusions are virtually the same for all of these studies, asignificant undervaluation of the exchange rate in the early 1990's and an equally significant loss of thisadvantage over time.19 By 1997, some economists have indicated that the exchange rate began to beovervalued.20 Given the large current account deficit at the end of 1996, above 10 percent of GDP, afunctional overvaluation of the exchange rate was possible.21 The devaluation of May and June 1997moved the level of REER down somewhat but only to the level of mid-1996 Perhaps this was notsufficient because the 1996 level may have left Czech industry highly exposed to foreign competition (ii) Rigidity of the Exchange Rate Regime There are reasons to believe that the exchange rateregime of the Czech Republic had been pegged inflexibly for too long As we have noted above, fixedexchange rates stimulate capital inflows while more flexible exchange rates regimes tend to discouragethem A fixed nominal rate proved to be useful in the early stage of the stabilization program, but theinsistence on a pegged exchange rate eventually became counterproductive A credible pegged exchangerate regime encouraged strong capital inflows, which proved difficult for the monetary authorities tocontrol In addition, the exchange rate regime encouraged inflows of short-term and speculative capitalseeking to profit from interest rate arbitrage (Cihák, 1997) It was only once the bands around the exchangerage parity were expanded that the capital inflows slowed down.22

(iii) Appreciation of Real Effective Exchange Rate While there may be disagreements about thedegree of distortions of the actual exchange rate from equilibrium rate, there is no disagreement about theevolution of the REER All studies that we have been able to review clearly show that REER hassignificantly appreciated irrespective of the method of calculation.23 The only disagreement is about themagnitude The evolution of the REER has been primarily the outcome of official policy In the CzechRepublic, as elsewhere in the region, views about the appropriate level of the nominal exchange rate weredivided One group of economists pushed for an undervalued exchange rate in order to provide sufficientprotection to the tradable sector Another group of economists argued against undervaluation partly on thegrounds of the high costs of such a policy.24 Moreover, they argued in favor of a strong currency in order

to increase the competitive pressures on domestic industry to restructure The official Czech policy initially

supported the soft currency policy, but this policy was subsequently abandoned The result of this decisionand of a relativly slow growth of factor productivity was the appreciation of the REER

The policy of undervaluation was also defended in a number of other countries.25 What made thedifference in the Czech experience was the desire to maintain a stable nominal exchange rate so as toimprove the country’s standing in international capital markets Ironically, this improvement, combinedwith an unwillingness to devalue, caused capital inflows that effectively undermined the policies thatbrought them about

(iv) The Impact on Commercial Policies Czech exchange rate policy has not been neutral for theconduct of commercial policies Given the undervaluation of the exchange rate in 1990-91, profitability inthe tradables sector remained relatively high for several years However, the appreciation of the REEReventually began to be reflected in the performance of the tradable sector Initially, the government avoidedprotectionist measures by adopting measures directed towards liberalization of the capital account (Dedek1997), which it hoped would moderate capital inflows An adjustment of the nominal exchange rate wasalso avoided The relaxation of restrictions on foreign currency transactions was introduced at the timewhen monetary policy was already under severe pressure from the surge in capital inflows It was thesepressures rather than the deteriorating trade and current account imbalance that led to the decision toliberalize capital outflows with the view of offsetting the pressures of capital inflows The measures wereonly marginally effective, and it was not long before protectionist pressures reappeared

The empirical evidence of the impact of the exchange rate policy is limited but instructive One ofthe more comprehensive studies is de Menil (1994) who compared the trade performance of differenttransition economies He found that the insistence on nominal anchors has been one of the major factorsbehind a relative deterioration in trade balances of the countries concerned, largely due to unsatisfactoryexport performance This conclusion has been challenged by Bruinshoofd (1997) on the grounds of animprovement in trade imbalance in 1997 in the Czech Republic and the increase in Czech exports

However, this improvement could be attributed to the devaluation of the koruna in Spring 1997 and to the

slowdown in domestic growth

Conclusions

Inappropriate exchange rate policies have led to protectionist pressures in transition economiessurveyed in this paper The pressures emanating from such exchange rate policies vary from time to timedepending on the changes in domestic fundamentals As a result, protectionist pressures can be cyclical, andthis appears to have been the case in the transition economies under consideration because they experiencedseveral shifts in commercial policy over this relatively short period of time Because the competitiveness ofdomestic firms was adversely affected by a rising real effective exchange rate, appropriate nominalexchange rate adjustments would have gone a long way toward mitigating the calls for protection At thesame time, exchange rate policies cannot be blamed for all the loss of competitiveness of domesticindustries in the Czech Republic or in other transition economies Other factors, including marketimperfections, inflexible management of currency baskets, and, in general, factors that determine thedomestic growth of output and inflation, have also played a role Moreover, the rise of protectionist lobbiesalso has been instrumental in increasing the pressure on governments to raise protection Soundmacroeconomic policies, institutional changes addressing market imperfections, and other pro-growthdomestic policies are also crucial Hungary, which has been more successful in restructuring its industriesthan other countries has demonstrated that a relatively overvalued exchange can be offset by measures toencourage the growth of productivity The pursuit of nominal-anchor exchange-rate policies has been

based, inter alia, on the belief that a stable nominal exchange rate would provide consistent signals to

investors, both domestic and foreign, about the comparative advantages of the country This belief hasclearly turned out to be wrong Whether comparative advantage remains stable is determined not by thelevel of the nominal exchange rate but by the level of real effective exchange rate, and the latter haschanged dramatically in many transition economies over the last few years Moreover, this change has beenquite rapid, indicating that the authorities will find it difficult to stick to their nominal anchor policies orthat they will face increasingly intense protectionist pressures