Exchange rate strategies of EU accession countries: Does exchange rate policy matter?

by Carolin Nerlich1 Prepared for the KOBE Research Seminar on “Regional economic, financial and monetary co-operation: the European and Asian experience”, Frankfurt, 15-16 April 2002 Preliminary version (6 May 2002)

Abstract

Against the background of similar starting and endpoints, the paper focuses on the question why accession countries adopted different exchange rate regimes and whether this led to different macroeconomic results. The main findings of the paper are: (i) Macroeconomic stabilisation, the degree of openness, the level of foreign reserves, capital flows and institutional factors can largely explain the regime choices of accession countries. (ii) Discrepancies between de jure exchange rate regimes and de facto policies were widespread but lost momentum in recent years, and the euro became increasingly important as reference currency. (iii) Regime choices had no discernible impact on real and nominal convergence, while the management of capital flows and external competitiveness tended to be facilitated by more flexible regimes.

1

The author is particularly grateful to Gunnar Jonnson and Christian Thimann for helpful comments and suggestions and to Stefan Wredenborg for excellent research assistance. The views expressed in this paper are those of the author and do not necessarily express those of the ECB.

1

1

Introduction

In view of the monetary and economic integration process in Asia, the experience of the countries currently negotiating accession to the European Union might be of relevance to the Asian countries for a number of reasons. 2 Obviously, the Asian countries and accession countries differ in terms of their respective degree of regional integration and their specific economic conditions. Still, a few general developments and current issues may be of interest. First, the EU integration process foresees similar to the experience of current EU Member countries - that economic integration precedes monetary unification and, in this context, the process is guided by an already existing institutional framework. Second, accession countries provide an illustrative example of integrating rather different economies into a comparatively homogenous economic and monetary union; consequently, convergence towards the EU is an important issue for accession countries. Finally, to better advance real and nominal convergence with the EU, accession countries have been managing their exchange rate policies in a rather proactive manner. The following paper will focus on this latter point. The accession countries display a number of similar features. First, they had a similar starting position, as most of them were former communist countries launching transition in 1989. Second, accession countries were faced with similar challenges during the past decade, which were characterised by the need of macroeconomic stabilisation, transition towards a market economy and adjustment with the EU requirements. Third, accession countries have the same objective, as joining the EU and eventually the euro area are the overriding economic policy objectives in all accession countries. With regard to exchange rate issues, the similarity among accession countries holds particularly true for the starting and endpoints of the process. At the beginning of the transition, most countries opted for a fixed exchange rate regime since an external anchor was regarded as the most effective strategy to combat inflation and reduce disinflation costs. This common approach was supported by the fact that accession 2

The 12 countries currently negotiating EU accession are Bulgaria, Cyprus, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, Slovakia and Slovenia. Turkey, which is also an accession country but has not yet started negotiations, is not considered in this paper.

2

countries were small and open economies and some of them even had introduced new currencies. As for the endpoint of accession, all accession countries aim at achieving greater stability of their currencies vis-à-vis the euro and eventually adopting the euro. In this regard, they will follow the same continuum of three different steps, namely EU accession, ERM II membership and eventually adoption of the euro.3 Against this background, one would expect that all accession countries had maintained rather similar exchange rate strategies for the whole period until the adoption of the euro. This was, however, not the case. Indeed, all types of exchange rate regimes are in place at present. They range from free floats in Poland, managed floats in the Czech Republic, pegs with large fluctuation bands shadowing ERM II in Hungary and Cyprus, hard pegs in Latvia to currency boards in Estonia. Moreover, since the beginning of transition from planned to market economies a number of countries experienced several regime shifts, mainly from fixed pegs towards more flexible regimes. These shifts were, with a few exceptions, not related to currency crises but the result of proactive policy management, which was largely facilitated by the institutional framework for EU integration. Finally, not all accession countries changed their regimes, as some countries maintained their regimes for the whole period since the beginning of the 1990s. What explains the high diversity of exchange rate regimes, what are common elements and have the different choices led to different macroeconomic results? This paper aims to shed some light on these questions. It provides an overview of the monetary and exchange rate strategies in accession countries from the beginning of transition throughout the present. The paper explores the differences between de jure and de facto exchange rate regimes and the role of the euro for the countries’ exchange rate strategies. Finally, the paper analyses whether the various regimes were associated with different achievements with regard to real and nominal convergence, the management of rising capital inflows and external competitiveness and to what extent exchange rate policies were used proactively to advance the transition process. 3

Each of these steps has different implications for the monetary policies and exchange rate strategies in accession countries. Before joining the EU, no single exchange rate strategy is prescribed, and accession countries are free to choose a strategy that best suits their objectives of real and nominal convergence. Upon EU accession, accession countries will join EMU with the status of “countries with a derogation”, as they are committed to eventually adopt the euro, and they should treat their exchange rate policy as a “matter of common interest”. In this context they will be expected to join the ERM II mechanism for a minimum of two

3

The main findings of the paper can be summarised as follows. At the beginning of transition, the choice of the exchange rate was largely driven by the need for macroeconomic stabilisation, the availability of foreign reserves and the degree of openness. Thereafter, the capability of managing capital flows became increasingly important, while more recent exchange rate regime shifts are mainly related to institutional factors taking into account future participation in ERM II. Discrepancies between de jure exchange rate regimes and de facto exchange rate policies were widespread in the mid-1990s but have been less prevalent in recent years, which could be interpreted as a sign of greater credibility of institutions and policy frameworks. Over the period, the role of the euro steadily increased, as it has become the main reference currency for most of the accession countries, including those operating managed floating regimes. With regard to macroeconomic achievements, exchange rate policies indeed mattered, and were of crucial importance in several cases. Exchange rate policies have been conducted in a forward-looking manner, which allowed accession countries to advance the parallel pursuit of real and nominal convergence. Moreover, the choice of different exchange rate strategies at different points in time had an impact on progress in disinflation, the management of capital flows and the countries’ external competitiveness.

2

Regime diversity, factors and phases

Accession countries have adopted rather diverse exchange rate and monetary strategies since the early 1990s, and some countries have also changed their strategies over time. This chapter presents on overview of the adopted exchange rate regimes, and tentatively examines the factors responsible for the choice of regimes at different stages of the transition process. In particular, the focus is on the role of inflation, the level of foreign reserves, the degree of openness, capital flows, and other institutional variables.4 As most of these variables changed over time, their relative importance for the choice of exchange rate regime also changed.

years before adopting the euro in order to prepare for monetary union and to achieve further real and nominal convergence. 4

Following the work by e.g. Poirson (2001).

4

The monetary and exchange rate strategies in the accession countries can be divided in three different phases (Table 1), each of which is characterised by certain challenges the countries were confronted with at that point in time. 5 In the “stabilisation phase” between 1990 and 1994 the main focus of monetary and exchange rate policy was to stabilise the economy. The “transition phase” between 1995 and 2000 was characterised by a greater economic and financial orientation towards the EU, large capital inflows and more autonomous macroeconomic policies. Finally, in the “preparatory phase”, which started in 2001 and will last until the adoption of the euro, accession countries are bringing their exchange rate regimes more in line with the ERM II requirements and prepare for joining the euro area by further fostering real and nominal convergence.

Table 1: Exchange rate regimes Stabilisation phase 1990-1994

Transition phase 1995-2000

Preparatory phase 2001 - ERMII/euro

Fix

Intermediate

Float

Czech Rep. Estonia Hungary Latvia (since 1994) Lithuania (since 1994) Malta Poland Slovakia

Cyprus

Bulgaria Latvia (1992-1994) Lithuania (1992-1994) Slovenia Romania

Bulgaria (since 1997) Estonia Latvia Lithuania Malta

Czech Rep. (1995-1997) Cyprus Hungary Poland (1995-2000) Slovakia (1995-1998)

Bulgaria (1990-1997) Czech Rep. (since 1997) Poland (since 2000) Slovakia (since 1998) Slovenia Romania

Bulgaria Estonia Latvia Lithuania Malta

Cyprus Hungary

Czech Rep. Poland Slovakia Slovenia Romania

De jure classification according to the IMF. Fix: currency board, conventional peg, narrow band; Intermediate : tightly managed, broad band; Float : managed float, free float

Stabilisation phase (1990-1994) Most accession countries entered the transition process in 1989 with a monetary overhang and experienced an initial surge of very high inflation (on average around 5

See e.g. Backé (1999); Kopits (1999); Van der Haegen and Thimann (2001).

5

300%). In part, this was due to comprehensive price and trade liberalisation, substantial devaluation of the currencies and rising fiscal imbalances. Only in the Czech Republic and Hungary were inflation rates relatively moderate at around 25%.6 Moreover, the beginning of transition was characterised by a significant fall in output in virtually all countries.7 Chart 1: Inflation and Exchange Rate Regimes Beginning of Stabilisation Phase

450%

(yoy, average 1990-92)

Estonia Lithuania

400%

Slovenia

Latvia

350% 300% 250%

Poland

200% Slovakia

150%

Bulgaria

100% Hungary Czech Rep. Malta

50% 0%

0.5

1

Romania

Cyprus 1.5

Fix

2

Intermediate

2.5

3

3.5

Float

Source: WEO

Against this background, accession countries needed a credible anchor for monetary policy to stabilise the economies. In several countries, an external anchor in form of a pegged exchange rate was regarded as the most effective strategy to combat inflation and reduce disinflation costs (Chart 1). As can be seen in Chart 1, in particular accession countries with high inflation rates at the beginning of the stabilisation phase, namely the Baltic States and Poland, adopted a fixed regime.8 The argument was that a fixed peg would commit the government to stabilise the economy, anchor inflation expectation for price- and wage-setters and allow a remontisation of the economy. Moreover, it would help to overcome problems related to an underdeveloped foreign exchange market and, in some countries, limited experience in central banking.9 6

Inflation rates in Cyprus and Malta were low, at 5.3% and 2.5%, respectively, as they are not transition economies.

7

See Sachs (1996); Fischer et al. (1996); Fischer and Sahay (2000).

8

Latvia and Lithuania, however, had flexible regimes when they first introduced their new currencies in 1992 but moved to fixed pegs relatively soon.

9

See e.g. van der Haegen and Thimann (2001).

6

While many accession countries opted for an external anchor, the policy rules differed between countries – ranging from currency board arrangements to conventional pegs and narrow bands. In most cases, the regimes were fixed against currency baskets where the US dollar had a large weight; Lithuania even adopted a currency board against the US dollar. The only two countries with a peg against European currencies were Cyprus (in form of a peg against the ECU) and Estonia (in form of a currency board with the D mark). The choice of fixed exchange rate regimes may also have been related to the fact that the accession countries are small open economies, with the exchange rate playing an important role in the disinflation process. Still, while the average degree of openness of all accession countries is around 114%, the degree of openness varies among countries, with Malta being the most open economy and Poland being the least open (Chart 2).10 As can be seen in Chart 2, the smaller more open economies have tended to keep fixed exchange rate regimes for longer periods.

Chart 2: Degree of Openness and Exchange Rate Regimes Over Time (exports and imports of goods and services, % of GDP, average 1990-2001) 200%

Malta

180% 160% Estonia

140%

Slovakia

120%

Latvia Lithuania

100%

Cyprus

Bulgaria

Hungary

80%

euro area average

60% 40%

Slovenia

Czech Rep.

Romania

Poland 0

1

Fix

2

Fixed for > 5 years

3

Intermediate

4

Float for > 5 years

5

6

Float

Source: WEO and IFS

It is interesting to note, however, that a few countries (Bulgaria, Slovenia and Romania) opted for a flexible exchange rate regime initially, despite being plagued by

10

In comparison to the euro area, the degree of openness of the individual member countries of the euro area is on average around 70%, while it is for the euro area as a whole around 35%. Moreover, while trade among the member countries of the euro area is large, the share of interregional trade among accession countries is rather small (15% of total exports and 10% of total imports).

7

high inflation rates and as in the case of Slovenia being highly open economies. The main reason for this choice seems to have been a lack of international reserves, which made it difficult to back a fixed exchange rate regime. Indeed, while the level of foreign reserves was quite low in most accession economies during the initial years of transition, this was in particular a problem in Romania and Slovenia (Chart 3),11 and – as they were also lacking other assets to be used as collateral instead – the only realistic option was to adopt a (managed) floating regime. Chart 3: Foreign Reserves and Exchange Rate Regimes Beginning of Stabilisation Phase

25%

(% of GDP, average 1990-92) Cyprus 20% Estonia 15%

10% Hungary Bulgaria

Poland

5%

Slovenia Romania

Lithuania 0%

0.5

1

Fix

1.5

2

2.5

Intermediate

3

3.5

Float

Source: IFS and WEO Data for Czech Republic, Latvia and Slovakia not available. In Malta the reserves in % of GDP stood at 49.5%.

Transition phase (1995-2000) At the mid-1990s, real growth had resumed in almost all accession countries, although a complete recovery from the initial output decline at the beginning of transition was not yet achieved (with the exception of Poland). At the same time, most accession countries had made large progress in disinflation with the average inflation rate declining to around 25%. Thus, the need to stabilise the economies via an external anchor became less compelling and a number of accession countries gradually adopted more flexible exchange rate regimes (Chart 4). 12 However, as inflation 11

Foreign reserves in Lithuania were also low, which might partly explain why it adopted a fixed regime only in 1994.

12

The only exception was Bulgaria, where macroeconomic imbalances and hyperinflation led to a currency crisis and forced the country to adopt a currency board in 1998.

8

differentials with the main trading partners, namely the EU countries, persisted, further disinflation was needed to maintain the countries’ external competitiveness. Yet, the fixed exchange rate regimes made it difficult for some countries to advance the disinflation process further, in part because large capital inflows (as further discussed below) contributed to high money growth and inflationary pressures, and as inflation expectations among price- and wage-setters remained stubbornly high. Chart 4: Inflation and Exchange Rate Regimes Beginning of Transition Phase

100%

(yoy, average 1995-97) Bulgaria*

90% 80%

Romania 70% 60% 50% 40% 30%

Lithuania Estonia Latvia

20%

Hungary Poland Slovenia Czech Rep. Slovakia

10% 0%

Malta 0.5

1

Fix

Cyprus 1.5

2

Intermediate

2.5

3

3.5

Float

Source: WEO (* Bulgaria average 1995-1996)

Due to the general progress in the transition towards market economies, including robust economic growth, substantial disinflation and the liberalisation of capital accounts, accession countries experienced large capital inflows (including portfolio flows that started in the mid-1990s. Depending on the countries’ absorption capacity, the size of domestic capital markets and the composition of the flows, capital flows started to pose problems in some countries that eventually resulted inter alia in regime adjustments. In some countries with initially fixed exchange rate regimes, large portfolio inflows had to be met by large-scale sterilisation. This was in particular a problem in the Czech Republic, Hungary and Poland. At the same time, the inflows made further progress in disinflation difficult, as interest rate increases to combat inflationary pressures attracted further inflows.

9

As a consequence of these developments, some countries moved pro-actively and gradually from fixed peg regimes towards more flexible regimes (see Chart 5).13 In particular the Czech Republic, Poland and the Slovak Republic continuously modified their regimes to allow for more exchange rate flexibility – first by gradually widening the fluctuation bands, and eventually by adopting managed floating regimes or free floats (in combination with inflation-targeting frameworks).14 Moreover, Hungary and Poland were the countries with the most frequent regime adjustments over time. Overall, the smooth experience of these countries with regard to regime adjustments indicate that exchange rate policies were in most cases pursued in a rather forwardlooking way and supported by the institutional framework of integration with the EU. Chart 5: Net Capital Flows and Exchange Rate Regimes Beginning of Transition Phase (% of GDP, average 1995-97)

14% 12%

Estonia

10%

Latvia Lithuania

Slovakia Czech Rep.

8% 6%

Romania Hungary

4%

0%

Poland

Bulgaria Malta

2%

0.5

1

Fix

Slovenia

1.5

2

Intermediate

2.5

3

3.5

Float

Source: WEO (Data for Cyprus n.a.)

Not all accession countries, however, moved towards more flexible regimes during the transition phase. Some countries maintained the fixed exchange rate regime they had initially chosen, while at least one country moved away from the floating regime. The Baltic countries were flexible enough to cope with transition-related challenges and to preserve external competitiveness despite their fixed regimes. Cyprus and Malta also maintained their relatively fixed regimes, as they were not confronted with the transition-related issues. Meanwhile, Romania and Slovenia officially maintained 13

The degree of exchange rate flexibility still differed to some extent, see Corker et al. (2000).

14

The regime modifications of the Czech Republic, however, ended abruptly with a currency crisis in 1997. Poland adopted a free float regime only in 2000.

10

their floating regime, although in practice these two countries started to implicitly target the exchange rate. Finally, Bulgaria was the main exception as it moved in the opposite direction – from a rather flexible regime to a currency board arrangement – following its currency crisis in 1998.

Preparatory phase (since 2001) Although the transition process is not completed and accession countries still have to catch-up with the euro area, the main driving force for regime changes in the preparatory phase is related to the prospect of joining the EMU as a full member. On their way to the euro, accession countries will have to join the ERM II mechanism for a minimum of two years and thereby bring their monetary and exchange rate strategies in line with the ERM II requirements. Such a move is expected to be in line with developments of the economic environment, as both the economic structures of accession countries and their economic policies will be increasingly oriented towards the euro area. More specifically, the ERM II mechanism foresees a fixed regime with a central rate against the euro, which has to be multilaterally agreed. While the mechanism could help avoid excessive exchange rate fluctuations, it still allows exchange rate adjustments, when necessary, through the large fluctuation bands of +/- 15% and the possibility of realignments. Fixed pegs against the euro with fluctuation bands smaller than +/-15% are in principle in line with the EU Treaty and will, if not supported by a multilateral agreement, be treated as unilateral commitments. While currency board arrangements might be judged compatible with ERM II – subject to an assessment on a case-by-case basis and regarded as unilateral commitments – fixed pegs against currencies other than the euro, free floats and crawling pegs are not compatible with the requirements. Against this background, a number of accession countries brought their regimes already in line with the institutional requirements. In 2001, Cyprus and Hungary introduced regimes which shadow the ERM II, i.e. with a central parity rate vis-à-vis the euro and a +/- 15% fluctuation band. Recently also Lithuania re-pegged its currency board from the US dollar to the euro. Looking ahead, in particular Latvia

11

(peg to the SDR), Malta (peg to a basket of euro, GDP and USD), and Romania (unofficial crawling peg to a basket of US dollar and euro) have to make their regimes compatible with the requirements before joining the ERM II mechanism. While the monetary and exchange rate policy strategies in accession countries are still quite diverse, it is expected that this diversity will decline to some extent during the preparatory phase. Such a decline in diversity would be in line with expected further progress in economic and institutional convergence towards the EU. In fact, the framework of ERM II is expected to foster gradual adjustment of economic fundamentals and structures with respect to the EU setting. Nevertheless, some diversity might remain, since different monetary and exchange rate policy regimes are de facto compatible with ERM II.

3

Some qualification on the regimes’ diversity

The following chapter focuses further on the question of whether the exchange rate regimes in accession countries are indeed as widely different as suggested in the previous chapter. Two aspects are analysed: (i) whether the diversity in regime is still apparent also after taking into account any discrepancies between the officially stated exchange rate regimes (de jure) and those that are observed in practice (de facto), and (ii) to what extent the euro is used as a references currency under the different regimes. 3.1

Discrepancies between de jure and de facto regimes

Several studies have shown that there is a large discrepancy between the de jure and the de facto regime in many emerging market economies.15 This seems to be the case also in many of the accession countries. In the period from the beginning of transition until the year 2000, discrepancies existed in more than 45% of the observations.16

15

See, for example, Calvo and Reinhardt (2000), Levy-Yeyati and Sturzenegger (2002), Poirson (2001), von Hagen and Zhou (2002).

16

The classification of de facto exchange rate regimes is based on a cluster analysis with three variables, namely the volatility of the nominal exchange rate, the volatility of exchange rate changes and the volatility of international reserves, see Levy-Yeyati and Sturzenegger (2000/2).

12

The comparison between de jure and de facto regimes for the accession countries reveals that intermediate regimes played in practice a much larger role than according to the official regime classification (Chart 6). Indeed, intermediate regimes were twice as often in place than officially stated. Thus, the “hollowing-out” hypothesis of intermediate regimes (Eichengreen, 1994) – according to which intermediate regimes are increasingly replaced by corner solutions – can not be confirmed for the accession countries. While this might be true also for other countries, the presence of intermediate regimes in accession countries could be regarded as an indication that the accession process per se serves as a credible anchor for the countries’ economic policies.

Chart 6: De Jure and De Facto FX Regimes in Accession Countries* 50

(in % of total; 1990-2000)

in % of total

fear of pegging

40

de jure

fear of floating

de facto no hollowing-out

30

20

10

0 fix

intermediate

float

Sources: Levy-Yeyati and Sturzenegger (2002); IMF; ECB staff calculation. * Data for Hungary and Malta not available.

Indeed, the discrepancies between de jure and de facto regimes seem to confirm the existence of a “fear of floating” (Calvo and Reinhardt, 2000) and “fear of pegging” behaviour in the accession countries. Fear of floating describes the situation where regimes are de jure more flexible than de facto, as the countries follow unofficial exchange rate targets. This situation has been present in Bulgaria, Romania, Slovakia and Slovenia. In particular Romania and Slovenia adopted a very tightly managed float regime (rather than a “managed float”) at a time when macroeconomic imbalances were large but foreign reserves too low to adopt an official peg.

13

Of greater relevance for accession countries was the phenomenon of “fear of pegging”, which describes the situation when regimes are de jure more fixed than de facto. Fear of pegging was a common phenomenon in the mid-1990s, when several countries with a fixed peg were confronted with challenges which required a higher degree of flexibility but the central banks had not yet build up sufficient credibility to adopt more flexible regimes.17 Thus, this might be another indication that the modifications of exchange rate regimes in accession countries were often pursued gradually and in a forward-looking manner –by first adjusting the de facto regime and only thereafter the de jure regime.

Chart 7: Fear of Floating and Fear of Pegging in Accession Countries 80

(in % of total, 1994-2000)

in % of total

70

fear of floating

60

fear of pegging in line with de jure regime

50 40 30 20 10 0 1994

1995

1996

1997

1998

1999

2000

Sources: Levy-Yeyati and Sturzenegger (2002); IMF; ECB staff calculation. Fear of floating: de facto exchange rate regimes are less flexibel than accortding to official (de jure) regimes. Fear of pegging: de facto exchange rate regimes are less fixed than according to de jure regimes. Data for Hungary and Malta not available.

It is interesting to note that the discrepancies between de jure and de facto regimes has changed over time, as fear of pegging and fear of floating were both mainly a phenomenon during the mid-90s but lost momentum in recent years (Chart 7). This might support the view that the conduct of monetary and exchange rate policy was particularly difficult at this period in time, which was characterised by substantial structural changes while the stabilisation process was not yet fully finalised. In the second half of the 1990s, however, with the transition process advancing, the difference between de jure and de facto regimes gradually declined. In 2000, the de facto strategies were in most countries in line with the de jure regimes, which might 17

Examples include Poland in the years before the adoption of a free float in 2000 and the Czech Republic before their regime change in 1997.

14

indicate a continuously higher credibility of the institutions and capacity building in the accession countries. Thus, with discrepancy declining over time, it seems reasonable to conclude that not only exchange rate regime diversity in accession countries has reflected differences in individual countries’ conditions, but that exchange rate policies have been carefully managed by taking changing economic and institutional circumstances into account.

3.2

Increasing role of the euro

Although the at present low discrepancy between de jure and de facto regimes suggests that the regimes are indeed rather diverse, it should be kept in mind that no accession country disregards developments in the nominal exchange rate as they are small and open economies. Even in the countries that adopted an inflation-targeting framework, the exchange rate still plays a crucial role in the decision making process. Exchange rate developments affect domestic inflation through a pass-through effect, and the exchange rate channel often constitute a more powerful transmission channel of monetary policy than domestic interest rates. In this context, it can be noted that the euro has increased its role as a reference currency in most accession countries (Chart 8). In the initial years of transition, during which the accession countries’ exchange rate strategies were mainly driven by the need of an external anchor, the US dollar was the main reference currency. As trade linkages became more important for defining the composition of currency baskets, the European legacy currencies (the D mark and the French franc) started to play a larger role in the exchange rate regimes of accession countries. This took place through increases of the weight of the European currencies in underlying currency baskets or even by adopting a European currency as the anchor currency.

15

Chart 8: Role of the euro as reference currency and trade linkages (share in % of maximal possible weight; 4Q 92 - 2Q 02)*

100 stabilisation phase

90

transition phase

preparatory phase

80

70 EMU export share in % of total

60

50

40 4Q 92

2Q 93

4Q 93

2Q 94

4Q 94

2Q 95

4Q 95

2Q 96

4Q 96

2Q 97

4Q 97

2Q 98

4Q 98

2Q 99

4Q 99

2Q 00

4Q 00

2Q 01

4Q 01

2Q 02

Source: National authorities, ECB staff calculation * Excluding Poland since 2Q 2000. Use of the euro as reference currency in all accession countries (weigthed average of all countries. 100 would indicate that the euro is the sole reference currency in all accession countries.

Indeed, the weight of the European currencies in the accession countries’ exchange rate regimes has increased from around 40% in 1992 to nearly 90% as of today. This is strongly related with the increased trade and financial linkages between accession countries and the euro area, reflecting accession countries’ greater orientation towards the euro area. Since 1992, accession countries’ exports to the member countries of the euro area increased from 49% to around 60% in 2000, and for the same period imports increased from around 46% to 53%. Moreover, this trend is expected to continue, as EU accession may further stimulate integration with the EU and euro area. With regard to financial integration, some areas of capital markets are already highly integrated and further progress is expected to take place in the coming years; for example, banks amounting for more than two-thirds of total domestic bank assets in accession countries are currently controlled by EU commercial banks. Since the beginning of the preparatory phase, the increasing role of the euro can be also explained by institutional factors, which are related to ERM II and the euro being the sole reference currency. In this context, recently a few countries have modified their regimes in such a way to be compatible with the requirements of the ERM II mechanism and adopted the euro as sole reference currency. Looking ahead, it is likely that the remaining countries will follow this path during the coming years.

16

4

Macroeconomic achievements

The following chapter analyses whether the exchange rate strategies helped in addressing the economic challenges that the accession countries faced during the transition process. These challenges include the parallel pursuit of real and nominal convergence, the capacity to manage capital flows and the need to preserve external competitiveness.

4.1

The impact on real and nominal convergence

A difficult task of monetary and exchange rate strategies in accession countries is to support the parallel pursuit of real and nominal convergence to the euro area, i.e. to advance the disinflation process while allowing real income levels and the structures of the economies to catch-up with those in the euro area. Real and nominal convergence are two interdependent processes that mutually affect each other; thus to find the balance between growth and disinflation is indeed a challenge for accession countries. Depending on the pace, a too rapid catching-up process could cause inflationary pressures in the short term, while a reasonable pace of structural reforms would increase the growth potential and the efficiency of labour markets and thereby reduce inflationary pressures. Likewise, lowering inflation could foster growth by triggering more investments, while a too tight disinflation path could temporarily defer catching-up if not sufficiently supported by structural reforms. In this context, also the choice of the exchange rate regime matters. It has been argued that fixed exchange rate regimes are often associated with better inflation performance but lower and more volatile output growth, while more flexible regimes are associated with higher inflation but somewhat stronger growth (Ghosh et al., 1996). According to this view, fixed regimes would focus more on nominal convergence, while more flexible regimes would put more emphasis on real convergence. However, looking at convergence in the accession countries, this view is not validated (Chart 9). In the first half of the 1990s, almost all accession countries were able to reduce inflation substantially, with the average inflation rate declining from nearly 300% in 1991 to around 25% in 1995.18 Grouped by regimes, the countries with fixed 18

Excluding Cyprus and Malta.

17

regimes made only slightly more progress in disinflation (average inflation was at around 26% in 1995), compared to the countries with flexible regimes (average inflation was at around 32% in 1995). As a result of macroeconomic stabilisation, economic growth recovered in all accession countries from around –10% in 1991 to around 5% in 1995. The countries with an external anchor were quite successful in recovering from the initial output decline, but economic growth was relatively high also in countries with flexible regimes, namely Slovenia and Romania (although inflation remained high in Romania). Chart 9: Real GDP Growth and Inflation (average 1994-1996) 100%

Fix

Bulgaria

90%

Intermediate

80%

Float

Inflation (yoy)

70%

Romania

60% 50% Lithuania

40%

Estonia

30%

Latvia

20%

Slovenia Slovakia Czech Rep. Cyprus Malta

10% 0% -5%

Poland

Hungary

-4%

Source: WEO

-3%

-2%

-1%

0%

1%

2%

3%

4%

5%

6%

7%

8%

Real GDP growth (yoy)

Since the mid-1990s, accession countries’ inflation developments are more in line with the general picture, namely that progress in disinflation was larger in the countries with fixed regimes, as inflation declined to 4% in 2001 (Chart 10). However, the countries with more flexible exchange rate regimes made also substantial progress in disinflation (inflation declined to around 9% in 2001), as restrictive monetary policy were adopted. Nominal convergence was further supported by the direct inflation targeting framework that was implemented in some countries, as it helped the central banks to build up credibility. Moreover, as most of these countries experienced an appreciation of their nominal exchange rates due to the catching-up process and capital inflows, this had a supportive impact on the disinflation development.

18

Chart 10: Real GDP Growth and Inflation (average 1998-2001) 50%

Fix

45%

Romania

Intermediate

40%

Float

Inflation (yoy)

35% 30% 25% 20% 15% 10%

Slovakia

5%

Czech Rep.

0% -1%

Lithuania 0%

Source: WEO

1%

2%

3%

Poland

Bulgaria

Hungary

Slovenia Estonia Latvia Malta Cyprus 4%

5%

Real GDP growth (yoy)

Nevertheless, as can be seen in Chart 10, output growth in recent years has actually been somewhat higher in countries with fixed or intermediate regimes (average growth was around 4.7% in 2001) than in countries with flexible regimes (average growth stood at 2.4% in 2001). While this is partly related to the recessions in Romania (1997-1999) and Poland (2001), it is possible that the transparency of policies and the policy discipline under particularly currency boards might have supported, rather than hindered, the transformation from centrally planned to market economies, and promoted the nominal and real convergence of their economies toward those of the EU members (Gulde et al., 2000). To sum up, accession countries advanced with real and nominal convergence irrespective of the underlying exchange rate regimes. Exchange rate regimes and regime adjustments were generally undertaken in line with developments in the economic environment. Thus, achievements in terms of real and nominal convergence can be considered to have been facilitated rather than hindered by different exchange rate strategies pursued in accession countries.

4.2

Managing capital flows

Most accession countries have gradually liberalised their capital accounts and the level of capital inflows in relation to GDP has increased from around 2% in 1994 to 19

around 8% in 2001. As regards the different components of capital flows, foreign direct investments have sharply increased to around 5% of GDP in 2001 (compared to around 2% of GDP in 1996) due to the countries’ high degree of human capital, low labour costs and the prospect of EU accession. Portfolio investments remained on average at a rather low level of around 1% of GDP, but increased recently in some accession countries, as remaining controls on short-term portfolio flows were abolished, the financial sector being further developed and due to the so-called “convergence play” (whereby financial market participants speculate on the convergence of interest rates with the euro area). Other investments such as foreign credits and loans have been rather volatile in recent years. In some countries, the periodical decline in net inflows was partly related to repayments of external debt. Capital inflows are in general expected to improve growth prospects through higher efficiency of production and to contribute to financial market deepening and development.19 Nevertheless, large and volatile capital inflows could undermine the countries’ macroeconomic or financial stability, in particular if the countries’ absorption capacity is rather low due to underdeveloped financial markets.20 The pace and sequencing of liberalisation and its support by consistent macroeconomic policies are also crucial to avoid financial distress or crises. Although generally positive, the experience of accession countries with managing capital flows differed. Exchange rate policy played together with the pace and sequencing of capital account liberalisation, the size of the domestic capital markets and the absorption capacity an important role in managing capital flows. In particular, the increase of capital flows to accession countries from the mid-1990s onwards made the sustainability of fixed or tightly managed exchange rate regimes at times difficult and impaired further progress in disinflation. As mentioned above, a number of accession countries moved therefore to more flexible regimes, which was to some

19

See survey by the World Bank (2001).

20

Measures to absorb capital flows generally include the deepening of financial markets and the strengthening of banking systems to avoid bubbles in the securities and banking markets. With a particular view to foreign direct investment, measures also include structural reforms to enhance competition and flexibility in the economy.

20

extent also conducive to further dismantling of capital controls (as for example in Poland).21 Managing capital flows through more flexible exchange rates, however, is no panacea, as accession countries still need to increase their absorption capacity, so that capital flows can contribute to higher domestic productivity and growth (Corker et al., 2000, Szapáry, 2000). As accession countries are small and open economies and in view of rising capital mobility, the countries’ ability to pursue fully independent monetary policy and to effectively use the exchange rate instrument to manage capital flows is limited. For example, a tightening of monetary policy may attract interest-rate sensitive capital flows (as was the case in Poland 2000/2001) and lead to an appreciation of the nominal and real exchange rate and a deterioration of the countries’ external competitiveness and current account position. To dampen upward exchange rate pressures, some countries have also introduced special accounts at the central bank to channel privatisation revenues and to be used later for the repayment of external debt. Finally, in accession countries with flexible regimes more volatile capital flows might increase exchange rate volatility, thereby negatively affecting trade and foreign investment and the effectiveness of monetary policy. Accession countries that maintained fixed regimes managed capital flows either through capital controls, high absorption capacity or a combination of both (Chart 11). After Hungary had liberalised the capital account at a rather early stage, it reintroduced capital controls in the mid-1990s in order to dampen short-term capital flows and to maintain the exchange rate regime. The capital controls were abolished only recently, whereby Hungary also moved towards more exchange rate flexibility. In contrast, the Baltic States were resilient against financial and macroeconomic vulnerabilities, despite fully liberalised capital accounts since the early 1990s. Their hard peg regimes remained sustainable over the whole period, reflecting sound fiscal policy, a limited domestic bond market, and a high degree of real flexibility within the economy, in particular with regard to labour markets. Moreover, the absorption capacity was high as foreign strategic investors owned most of the commercial banks 21

The experience of the Czech Republic, however, was less smooth. Due to an early liberalisation of the capital account, capital inflows amounted to around 16% of GDP in 1995, including short-term flows. As this capital was mainly used for consumption and inefficiently allocated, the economy started to overheat and the external competitiveness deteriorated significantly. In the Slovak Republic, unsound macroeconomic policy and

21

and enterprises (Sutela, 2001). Finally, the experience of Cyprus and Malta was somewhat different, as they had capital controls and a relatively high absorption capacity, enabling them to redirect capital flows in a productive manner.

Chart 11: Net Capital Flows and Capital Account Liberalisation (average 1998-2001 for capital flows and 1998-2000 for capital account) 10%

Slovakia Lithuania

Net capital flows (% of GDP)

9%

Latvia

Bulgaria

Malta

8%

Estonia Czech Rep.

7% 6% Hungary

5%

Poland

Romania

Fix

4%

Intermediate Slovenia

3%

Float

2% 20

30

40

50 60 70 Capital account liberalisation index

80

90

100

Source: WEO (Data for Cyprus n.a.), national authorities, ECB staff calculations. The capital account liberalisation index ranges between 0 and 100, with 100 indicating free movement of capital (based on IMF h d)

4.3

Safeguarding external competitiveness

Maintaining external competitiveness is an important issue for accession countries, as accession countries are small and open economies, in which exports and imports play an important role for their growth and investment developments. Hence, a loss in external competitiveness can rapidly translate into a deterioration of the external current account. While high current account deficits in accession countries in part reflect significant investment needs, concerns about external sustainability might arise depending on the composition of future capital inflows.22 As accession countries are advancing in real convergence and transition and as they receive large capital inflows, the real exchange rate is likely to appreciate. Depending on the underlying exchange rate regime, the appreciation of the real exchange rate can sluggish structural reforms made the fixed regime unsustainable and forced the adoption of a more flexible regime. 22

The financing of the current account deficits might be endangered, as the privatisation process might come to an end in the coming years. This would lower the size of FDI inflows, if greenfield investments would not increase accordingly.

22

be either achieved through nominal appreciation, higher inflation differentials with the anchor country, or a combination of both. To the extent that the real appreciation of the exchange rate is not fully matched by increases in productivity or product quality, it would threaten the countries’ external competitiveness. This would be the case, for example, if real wage growth in the tradable sector exceed productivity increases.

Chart 12: Real Effective Exchange Rate and Current Account (average 1998-2001) 10% Fix

Real effective exchange rate (yearly changes)

Lithuania

Romania

8%

Intermediate

Latvia Float

6%

Bulgaria

4%

Estonia Poland

Czech Rep.

2%

Slovakia

Malta

Hungary

Slovenia

0%

Cyprus

-2% -10%

-9%

-8%

-7%

-6% -5% -4% -3% Current account balance (% of GDP)

-2%

-1%

0%

Source: IFS, Deutsche Bank Research and WEO.

Concerns about competitiveness would apply in particular to countries with a fixed regime, if product and labour market flexibility were insufficient to accommodate shifts in international comparative advantages. As can be seen in Chart 12, the appreciation of the real effective exchange rate and the current account deficits were on average higher in the countries with fixed exchange rate regimes than in countries with free floats.23 Even if the current account as a whole is not endangered in fixed regime countries, individual sectors could be confronted with competitiveness problems that would have a knock-on effect on the rest of the economy.

23

However, accession countries with hard pegs entered their regimes amidst strong devaluation of their currencies in order to support their external competitiveness.

23

5

Conclusions

Exchange rate strategies in accession countries are characterised by similar starting and endpoints: most countries had fixed pegs at the beginning of transition and they all aim at eventually adopting the euro. Furthermore, accession countries are small and open catching-up economies, and thereby confronted with similar challenges, such as real convergence, capital inflows and external competitiveness. Nevertheless, despite these similarities, exchange rate strategies in accession countries are at present rather diverse, ranging from free floats to currency boards. The aim of this paper was to identify the main reasons why accession countries decided to adopt different regimes and how the macroeconomic achievements differed among countries due to the diverse regimes. The main findings are: (i) In the initial years of transition, most accession countries adopted an external anchor to stabilise the economy, while only those countries with a lack of foreign reserves opted for flexible regimes. (ii) Due to large capital inflows, some countries gradually moved towards more flexible regimes, while institutional factors (ERM II requirements) were the main driving force for the more recent regime shifts. These shifts were, with a few exceptions, not related to currency crises but the result of proactive policy management, which was also facilitated by the institutional framework of EU integration. (iii) Large, although declining, discrepancies between de jure and de facto exchange rate regimes also indicate that the exchange rate policy in accession countries was rather carefully managed, by first adjusting the de facto regime and only thereafter the de jure regime. At the same time, the role of the euro as reference currency increased sharply during the 1990s, mainly reflecting the increased trade and financial orientation of accession countries to the EU. (iv) Exchange rate regimes had no discernible impact on progress in real and nominal convergence, although progress in disinflation seemed to be stronger in countries with fixed regimes, while the capacity to cope with large and volatile capital flows and to maintain the countries’ external competitiveness tended to be facilitated by a more flexible regime. Similar progress in real and nominal convergence should lead to the conclusion that these achievements were facilitated rather than hindered by the pursuits of different exchange rate strategies and proactive policy management in accession countries.

24

References Backé, Peter (1999), “Exchange Rate Regimes in Central and Eastern Europe: A Brief Review of Recent Changes, Current Issues and Future Challenges”, Focus in Transition, No. 2, 47-67. Calvo Guillermo and Carmen Reinhart (2000): “Fear of Floating”, University of Maryland, mimeo. Corker, Richard, Craig Beaumont, Rachel van Elkan and Dora Iakova (2000), “Exchange Rate Regimes in Selected Advanced Transition Economies Coping with Transition, Capital Inflows, and EU Accession”, IMF Policy Discussion Papers PDP/00/3. Gulde, Anne-Marie, Juha Kähkönen, and Peter Keller (2000), “Pros and Cons of Currency Board Arrangements in the Lead-Up to EU Accession and Participation in the Euro Zone”, IMF Policy Discussion Papers PDP/00/1. Ghosh, Atish, Ann-Marie Gulde, Jonathan Ostry and Holger Wolf (1996): “Does the Exchange Rate Regime Matter for Inflation and Growth?”, IMF, Economic Issues, No. 2. Eichengreen, Barry (1994), “International Monetary Arrangements for the 21st Century”, Brooking Institution, Washington D.C. Fischer, Stanley; Ratna Sahay and Carlos A. Végh (1996), “Stabilisation and Growth in Transition Economies: The Early Experience”, Journal of Economic Perspectives, Vol. 10, No. 2, 45-66. Fischer, Stanley and Ratna Sahay (2000), “The Transition Economics After Ten Years”, IMF Working Paper WP/00/30. Kopits, George (1999), “Implications of EMU for Exchange Rate Policy in Central and Eastern Europe”, IMF Working Paper WP/99/9. Levy-Yeyati, Eduardo and Federico Sturzenegger (2000/2002): “Classifying Exchange Rate Regimes: Deeds versus Words”, Universidad Torcuato Di Tella, Buenos Aires, mimeo. Poirson, Hélène (2001): “How Do Countries Choose Their Exchange Rate Regime?”, IMF Working Paper, WP/01/46. Sachs, Jeffrey D. (1996): “Economic Transition and the Exchange-Rate Regime”, The American Economic Review, Vol. 86, No. 2, 147-152. Sutela, Pekka (2001): “Managing Capital Flows in Estonia and Latvia”, BOFIT Discussion Paper, No. 17. Szapáry, György (2000), “Maastricht and the Choice of Exchange Rate Regime in Transition Countries during the Run-up to EMU”, NBH Working Paper 2000/7, National Bank of Hungary.

25

Van der Haegen and Christian Thimann (2001): “Monetary Policy Challenges in Transition and Toward Accession”, in: G. Tumpel-Gugerell, L. Wolfe, P. Mooslechner (eds.): Completing Transition: The Main Challenges, Heidelberg, 161-184. Von Hagen, Jürgen and Jizhong Zhou (2002): “De Facto and Official Exchange Rate Regimes in Transition Economies”, Center for European Studies, Bonn, mimeo. World Bank (2001): Global Development Finance, Washington D.C..

26