Restructuring European Union Trade with Central and Eastern European Countries

JARKO FIDRMUC*

Central and eastern European countries (CEECs) participate in the European economy with trade shares of the European Union (EU) and levels of intraindustry trade comparable to peripheral EU countries. However, the opening of CEECs has induced increased specialization in EU countries, which contrasts with the development in previous decades. This partially explains the cautious approach to the eastward enlargement in the EU. Furthermore, CEECs are more similar to each other than to EU countries. The pattern of the CEECs' trade with the EU resembles that of Turkey. Trade diversion is likely to present a significant burden for countries omitted from the first wave of the enlargement. (JEL F12, F14, F15)

Introduction

The opening of eastern Europe and the proposed full integration of the eastern European countries into the European Union (EU) imposes different effects on EU countries. The current discussion of eastern enlargement of the EU focuses on the question of budgetary, migration, and adjustment costs. Nevertheless, foreign trade represents an important channel for both the adjustment needs and growlh potential in East-West integration in Europe.

This paper analyzes the impact of an EU eastern enlargement by comparing it with previous developments in EU trade with the central and eastern European countries (CEECs) (Bulgaria, Czech Republic, Hungary, Poland, Romania, Slovakia, and Slovenia) in the manufacturing sector, which has been significantly liberalized by the Europe Agreements. The CEECs were relatively successful in regional and structural changes of exports and imports. The restructuring of East-West trade provides a comparably better indication of the convergence of CEECs to EU countries than any other economic development. The growth of intraindustry trade, which is observed in intra-EU trade, also dominates the recent development of East-West trade. This could lower the possible negative impact on EU countries. This paper analyzes the political economy implications of restructuring EU trade with CEECs.

This paper is organized as follows. The second section discusses the political economy of trade liberalization and integration. The third section presents the development of EU

*Institute for Advanced Studies--Austria. The author would like to acknowledge many useful comments provided by Reinhard Neck, Andreas W6rg6tter, Jan Fidrmuc, Katharina Helmstedt, and the participants of the Forty-Seventh International Atlantic Economic Conference, Vienna, Austria, March 16-23,1999.

83

84 AEJ: MARCH 2000, VOL. 28, NO. 1

trade with CEECs in a comparison to intra-EU trade and trade with selected third countries. Conclusions will be presented in the fourth section.

Political Economy of Trade Liberalization and Integration

Since the opening of eastern Europe in the late 1980s, the expectations of the impact of trade liberalization between the EU (and other countries of the Organization for Economic Cooperation and Development) and CEECs are driven by the arguments of the Heckscher-Ohlin model. CEECs are seen to be abundant in qualified and unqualified labor, some raw materials, and energy. This pattern of factor endowments is similar to southern European countries, while northern member states of the EU are abundant in capital and human capital. Therefore, this analysis of the impact of trade liberalization with CEECs focuses, first, on competition with southern European countries and, second, on factor price changes through liberalized trade [Collins and Rodrik, 1991]. These arguments follow the discussion related to the creation of the North American Free Trade Agreement in the U.S. [Leamer, 1992].

Based on the trade structure in 1991 and 1992, Neven [1995] finds that northern European countries seem to have comparative advantages relative to CEECs in technology and human capital intensive products. Southern European countries have a comparative advantage in labor-intensive industries with low capital content. This means that CEECs may specialize in labor-intensive products with high capital content, thus filling a niche in the European division of labor. However, CEECs are also likely to compete with southern European countries in labor-intensive products as well. Not surprisingly, the Europe Agreements limited trade liberalization of some products of the former group and nearly all products of the latter group due to its sensitivity.

Under such conditions, trade liberalization should induce changes of factor prices. The price of the abundant factor should increase while the price of the scarce factor should decline as a result of trade liberalization. Therefore, qualified labor (human capital) is assumed to gain in northern EU countries, while less qualified labor and, to some extent, capital are likely to lose from the opening of eastern Europe. The opposite pattern of development is expected in CEECs.

Neven [1995] notes that human capital is less well organized than labor and capital in northern European countries. Therefore, the representatives of labor and firms are likely to have more influence on domestic policy. The pressure for protection could further increase owing to regional differences. In southern European countries, labor could gain at the expense of capital owners. The different interests of these two relatively well organized interest groups could make the protection pressures rather ineffective in southern Europe.

These effects could be limited by the role of the intraindustry trade. A high share of the intraindustry trade suggests a lower role of comparative advantage and factor endowments for trade. Therefore, the distributional effects of trade liberalization and integration could be lower with respect to integration of countries with high shares of the intraindustry trade. Indeed, recent studies found increasing importance of the

FIDRMUC: RESTRUCTLq~ING EUROPEAN UNION TRADE 85

intraindustry trade between the EU and CEECs [Djankov and Hoekman, 1997; Aturupane et al., 1997; Fidrmuc et al., 1998, 1999].

Empirical Results

This paper compares the development of EU trade with 7 CEECs and EU trade with 10 member states of the EU (Austria, Germany, Finland, France, Ireland, Italy, the Netherlands, Spain, Sweden, and the United Kingdom)between 1990 and 1996. EU trade with 2 non-European countries (Israel and Turkey) provides another benchmark for these comparisons. These countries were selected because they have intensive trade relations with the EU, similar to that of CEECs. This paper uses trade flows by Standard International Trade Classification (SITC) three-digit commodity groups (in terms of current prices as published by the United Nations) that were converted to European currency units. In general, export and import prices are relatively stable (under 1 percent annually). Therefore, price growth should be relatively less important given the growth of EU trade with CEECs.

The investigated period was characterized by dramatic institutional changes apart from the opening of eastern Europe, which is the focus of this analysis. Germany reunified in 1990. Slovenia became independent in 1991, while the former Czechoslovak federation was divided into the Czech Republic and Slovakia in 1993. Austria, Finland, and Sweden joined the EU in 1995. Moreover, the United Nations introduced a new scheme of trade statistics by detailed commodity groups (SITC Revision 3) at the end of the 1980s.

Export Growth EU trade with CEECs has been rapidly growing since the opening of eastern Europe

and trade liberalization between the EU and CEECs. In current prices converted to European currency units, EU imports of industrial products from CEECs rose by 23.1 percent on annual average between 1991 and 1996. The highest growth was reached by countries that opened later, that is, by the former Czechoslovakia (31.3 percent on annual average) and Bulgaria (23.9 percent). Nevertheless, imports from Hungary, Poland, and Romania also grew by about 20 percent annually. As a result, the share of imports from CEECs (including Slovenia after 1991) increased from 2.0 percent of intra-EU trade in 1990 to nearly triple (5.8 percent) in 1996.

The growth of exports from CEECs to the selected countries of the EU greatly exceeded the growth of intra-EU trade (3.2 percent on average). In the EU, Spain (9.9 percent), Ireland (9.2 percent), the United Kingdom (5.8 percent), and France (4.0 percent) experienced the highest growth of their exports to the internal market. Surprisingly, the lowest growth of manufacturing exports to the EU can be found in Austria (1.5 percent on average), Italy, and Germany (both 1.9 percent) while the exports of the other selected EU countries increased between 2 and 3 percent on annual average between 1991 and 1996. The exports of Turkey and Israel to the EU also grew comparably fast at 9.1 and 9.2 percent on average, respectively.

This suggests that neither exports of southern European countries nor exports of non- European countries to the EU were replaced with imports from CEECs. The relatively

86 AEJ: MARCH 2000, VOL. 28, NO. 1

worse export performance of the crescent countries (Austria, Germany, and Italy) could be explained partially by the stronger position of the CEECs in EU border regions, leading to a decline in trade between these countries. Nevertheless, this trade stagnation between the latter countries could be more than compensated by their increased exports to CEECs.

Trade Pa t t e rn

The growth of central and eastern European exports to the EU was associated with significant trade restructuring. The redirection of goods that were traditionally exported to CEECs and the former Soviet Union did not play an important role. Djankov and Hoekman [1997] find that the export growth concerned either products not exported to eastern European countries or that such exports were substantially upgraded.

The Grubel-Lloyd index of intraindustry trade has crucial importance for the type of restructuring of foreign trade between the EU and CEECs. The index represents the share of absolute value of the intraindustry trade in trade turnover. That is, G L I i =

1 -IX. - M.[t (X. + M/), where X i and M i denote exports and imports by commodity groups, i, respectively. An index value of 0 shows that there is interindustry trade exclusively. That is, for each country, there is complete specialization according to comparative advantage. An index value of i (or 100 percent) shows a full dominance of intraindustry trade corresponding to increasing returns to scale [Dixit and Norman, 1980].

The growth of intraindustry trade is the most important feature in the development of East-West trade (see Table 1). In 1990, the share of intraindustry trade in trade turnover (as computed by Grubel-Lloyd indices for manufacturing products by SITC three-digit commodity groups) was between 25 percent (24.9 percent for Romania and 26.2 percent for Bulgaria) and 50 percent (43.0 percent for the former Czech and Slovak Federal Republic and 47.4 percent for Hungary). These shares corresponded to the importance of intraindustry trade in EU trade with Turkey (22.0 percent) and Israel (46.3 percent) or also Finland (49.0 percent). These countries had relatively liberal trade regimes vis gt vis the EU, as well as a peripheral location, a relatively less-developed manufacturing sector, and specific resource bases. However, these shares of intraindustry trade were far below the levels of EU countries, which were between about 60 percent (58.7 percent for Ireland, 62.1 percent for Italy, and 62.8 percent for Spain) and 80 percent (81.0 percent for France).

Between 1991 and 1996, all CEECs experienced a significant growth of intraindustry trade. As a result, the levels of intraindustry trade in EU trade with the Czech Republic (62.8 percent), Slovenia (60.5 percent), and Hungary (57.7 percent) were comparable to or even larger than in EU trade with Spain (61.3 percent), Italy (61.4 percent), and Sweden (61.8 percent) in 1996. In turn, Poland and Slovakia showed somewhat lower levels of intraindustry trade at 41.0 percent and 50.9 percent, respectively, in 1996. These levels were comparable to Israel (44.4 percent) but also to Finland (48.3 percent) and Ireland (53.6 percent). However, the share of intraindustry trade in EU trade with Romania (29.4 percent) and Bulgaria (32.8 percent) still remained only slightly above the level of intraindustry trade with Turkey (25.5 percent).

FIDRMUC: RESTRUCTURING EUROPEAN UNION TRADE

TABLE 1 Intraindustry Trade in the Manufacturing Sector

of Selected Countries with 10 EU Countries

87

1990 1 9 9 1 1 9 9 2 1 9 9 3 1 9 9 4 1 9 9 5 1996

EU Countries:

Austria 69.34 70.24 70.15 69.13 68.90 67.37 66.92

Germany 73.26 78.25 77.05 78.04 77.73 78.59 78.48

Spain 62.80 59.22 61.89 60.05 60.07 60.92 61.27

Finland 49.04 53.05 52.12 49.91 50.04 48.33 48.26

France 81.00 82.33 83.77 80.55 81.57 80.56 80.57

United Kingdom 74.63 78.42 78.59 76.81 78.41 80.01 80.85

Ireland 58.71 59.27 56.91 56.46 53.88 51.77 53.55

Italy 62.08 61.56 60.26 60.83 59.94 61.57 61.40

Netherlands 75.69 76.45 76.88 73.51 73.35 73.53 73,45

Sweden 66.52 65.79 64.82 61.72 64.29 60.68 61.76

Selected Countries:

Czech and Slovak Federal Republic 43.02 47.13 50.27 54.76 58.55 60.83 62.34

Czech Republic n/a n/a n/a 55.85 59.36 61.34 62.80

Bulgaria 26.16 27.20 29.11 33.63 36.01 30.52 32.82

Hungary 47.39 51.33 52.47 52.56 53.78 56.53 57.73

Poland 36.20 35.11 38.04 38.36 38.87 40.69 40.98

Romania 24.88 26.59 23.62 24.68 27.22 27.67 29.44

Slovakia n/a n/a n/a 40.16 44.55 46.80 50.86

Slovenia n/a n/a 49.92 54.54 57.29 59.98 60.48

Israel 46.31 42.08 42.39 42.69 41.73 44.11 44.39

Turkey 22.02 20.30 21.02 18.69 24.72 27.19 25.49

Notes: Figures are in percentages.

The growth of shares of EU intraindustry trade with CEECs contrasts sharply with the stagnation or even relative reduction of the levels of the intraindustry trade in the EU. Of the selected 10 EU countries, the share of the intraindustry trade increased only in

88 AFt: MARCH 2000, VOL. 28, NO. 1

Germany and the United Kingdom, but the increase in Germany seems to be driven largely by the reunification in 1991. The level of the intraindustry trade in EU trade with Germany stagnated in the following years. The largest declines of levels of intraindustry trade are observed in Ireland, Sweden, Austria, Spain, and the Netherlands. Except for the Netherlands, these countries' shares of the intraindustry trade were relatively low within the EU. Moreover, these countries were integrated into the EU relatively late. They are located at EU peripheries, and some of them had to catch up to the level of the EU. All these factors are assumed to have a positive effect on the development of the intraindustry trade. In contrast to recent developments, these countries had shown fast convergence to EU levels of intraindustry trade in previous decades.

The nearly uniform downward development of Grubel-Lloyd indices in the EU cannot be explained by business cycles, although the slowdown of EU growth might also play a role at the beginning of the 1990s. The extension of the intraindustry cooperation to CEECs can also explain the performance of intraindustry trade in countries having intensive trade relations with CEECs (Austria and Sweden) but not developments in Spain and Ireland. Rather, it seems that the opening of eastern Europe and its stepwise integration (Europe Agreements) induced increased specialization in the EU.

This shape of East-West trade restructuring helps to understand the cautious approach to trade liberalization with CEECs and the eastward enlargement in those countries, which are expected to gain from the enlargement including Germany and Austria. Although it is generally accepted that these countries gained from the earlier trade liberalization and that they will benefit from the accession of their trade partners to the EU, their producers in all sectors face increased competition and seek increasingly for protection.

Similarity o f Trade Between Selected Countries

Net exports represent a counterweight to the intraindustry trade. The pattern of net exports is explained by the factor proportion theory (Heckscher-Ohlin model). From the political economy point of view, sectoral trade imbalances are assumed to have direct effects on employment. The unemployed from the contracting sectors often lose a significant part of their qualification that makes their reemployment difficult. Moreover, the contracting sectors are often likely to be regionally concentrated. This might give birth to effective political groups fighting against trade liberalization and integration. Furthermore, the sectoral and regional concentration of the adverse effects of trade liberalization and integration also increases the role of interest groups in the country's policy making.

In country-specific analysis and bilateral comparisons, net exports are proportional to the indices of revealed comparative advantage, RCA t = ( X i - Mi ) / ( X i + M i ) . The correlation of the pattern of net exports in 2 years shows the persistence of the pattern of comparative advantage or, respectively, the adjustment capability.

A comparison of similarities reveals several valuable insights. First, CEECs represent a relatively homogeneous group with the highest similarities of net exports in Europe (see Table 2). Moreover, the trade structures of CEECs seem to be becoming more similar.

TA

BL

E 2

Si

mila

rity

of

10 E

urop

ean

Uni

on C

ount

ries

' N

et E

xpor

ts t

o Se

lect

ed C

ount

ries

: 19

96

Cze

ch

Uni

ted

Hun

gary

R

oman

ia

Rep

ubli

c S

love

nia

Ger

man

y*

Finl

and*

K

ingd

om*

Ital

y*

Swed

en*

Pol

and

Bul

gari

a S

lova

lda

Aus

tria

* Sp

ain*

Fr

ance

* Ir

elan

d*

Net

herl

ands

* Is

rael

Pola

nd

Rom

ania

Bul

gari

a

Cze

ch

Rep

ubli

c

Slov

akia

Slov

enia

Aus

tria

*

0.58

0.61

0.

68

0.50

0.

61

0.71

0.48

0.

71

0.47

0.

45

0.48

0.

58

0.67

0.

64

0.58

0.36

0.

55

0.54

0.

28

0.37

0.

59

0.22

0.

08

-0.0

9 0.

09

0.31

-0

.09

-0.3

6

Ger

man

y*

-0.2

7 -0

.36

-0.1

7 -0

.18

-0.3

4 0.

00

0.06

-0

.46

-0.0

8 -0

.01

0.05

-0

.02

0.03

0.

39

0.40

-0

.59

0.68

-0.1

9 -0

.11

-0.0

2 -0

.07

-0.0

8 0.

01

-0.0

5 0.

32

-0.2

6 -0

.25

0.06

-0

.17

-0.0

5 0.

00

0.16

-0

.11

-0.2

4 0.

30

0.19

-0

.08

-0.1

4

Spai

n*

Finl

and*

Fran

ce*

Uni

ted

Kin

gdom

*

Irel

and*

Ital

y *

0.22

0.

37

0.02

0.

19

0.27

-0

.02

-0.0

6 0.

19

-0.4

6 -0

.32

-0.4

5 -0

.08

0.06

-0

.04

-0.0

8 -0

.07

-0.1

2 -0

.21

-0.1

7 -0

.07

-0.3

5 -0

.35

-0.0

9 -0

.14

0.31

0.21

0.

24

0.25

0.

14

0.27

-0

.06

-0.0

1 0.

33

-0.6

4 -0

.52

0.01

-0

.03

0.14

-0

.01

Net

herl

ands

* 0.

05

0.21

-0

.04

0.16

0.

29

0.02

-0

.27

0.54

-0

.39

-0.4

2 -0

.08

0.12

0.

55

-0.0

4 0.

15

Swed

en*

-0.2

3 -0

.14

-0.0

4 -0

.11

-0.0

7 -0

.01

-0.0

6 0.

32

-0.1

4 -0

.15

0.88

-0

.02

-0.4

7 -0

.27

0.07

-0

.08

Isra

el

0.12

0.

21

0.02

0.

18

0.23

-0

.05

-0.1

5 0.

41

-0.5

0 -0

.50

0.07

0.

03

0.49

0.

20

0.29

0.

46

-0.0

2

Tur

key

0.42

0.

53

0.50

0.

45

0.47

0.

41

0.34

0.

06

-0.3

6 -0

.02

-0.0

6 -0

.06

0.28

-0

.02

0.29

0.

14

-0.0

8 0.

25

¢3

¢3

©

Not

es:

* den

otes

one

of t

he 1

0 E

urop

ean

Uni

on c

ount

ries

.

90 AEJ: MARCH 2000, VOL. 28, NO. 1

In 1996, the highest similarity is observed between EU net trade with Poland and the Czech Republic (0.71), Bulgaria and Romania (0.71), Slovakia and Romania (0.67) as well as Bulgaria (0.64), and Hungary and Romania (0.61).

Trade flows between EU countries show much lower similarities. High similarity reflects similar positions in the international division of labor. In 1996, the highest similarity can be found for EU net exports to Finland and Sweden (0.88) and to Germany and Spain (0.68). While the former can be attributed to similarities in the geographical location, factor endowments, and the earlier membership in the European Free Trade Association of Finland and Sweden, the latter can be explained by the role of the automobile sector. From this point of view, Germany and Spain have a comparative advantage in sectors exploiting the large domestic markets.

Second, net trade flows within the EU and between the EU and CEECs are much less similar. In 1996, EU net trade with Spain and Slovenia shows the largest similarity (0.40) followed by EU net exports to Poland and the United Kingdom (0.37) and Austria and the Czech Republic (0.31) or Austria and Hungary (0.35) in 1995. The United Kingdom (0.28), Italy (0.26), and Austria (0.12) show the highest similarity of their net exports to CEECs. As a result, these countries could face a stronger adjustment need than the other EU countries.

Often, the net exports of CEECs fit the import needs of EU countries relatively well. This is reflected by high negative correlation coefficients of the EU trade structure with Germany and Poland (-0.36), Austria and Slovenia (-0.36), and Germany and the Czech Republic (-0.34) in 1996. These countries are likely to profit particularly from further trade liberalization. In general, Germany, Sweden, and Finland could draw an advantage from further trade liberalization with CEECs.

The pattern of the CEECs' net exports to the EU is relatively similar to Turkey's trade with the EU. Moreover, EU trade with the countries likely to participate in the first wave of the enlargement is slightly more similar to EU trade with Turkey (0.53 for Poland, 0.47 for the Czech Republic, and 0.42 for Hungary) than EU trade with the other CEECs (0.50 for Romania, 0.45 for Bulgaria, and 0.41 for Slovakia). This helps to explain Turkey's fear of exclusion from European integration.

Conclusions

As of 1996, CEECs had already been successfully participating in the European division of labor. The EU is the most important trade partner for all CEECs. The regional reorientafion of central and eastern European trade was associated with successful restructuring. The rise of the intraindustry trade was the most important feature of the recent development of East-West trade in Europe. The most advanced CEECs (Hungary, Slovenia, and the Czech Republic) already show shares of intraindustry trade comparable to Italy, Spain, and Sweden. Poland and Slovakia reached somewhat lower levels of intraindustry trade. Nevertheless, these levels are comparable to those of Finland and Ireland. On the other hand, the share of the intraindustry trade in EU trade with

FIDRMUC: RESTRUCTURING EUROPEAN UNION TRADE 91

Romania and Bulgaria remained still only slightly above the level of the intraindustry trade with Turkey.

The integration of CEECs will be associated with significant adjustment needs in many countries. The implementation of CEECs into the European trading system already induced increased specialization within the EU, which contrasts with the development in previous decades.

Nevertheless, the effects on EU countries are negligible. As opposed to the initial expectations, the adjustment pressure will probably be concentrated on countries that benefit overall from trade liberalization (Germany, Italy, and Austria). Germany seems to gain from trade with CEECs through the Europe Agreements far more than other EU countries. Unlike Austria and Italy, Germany's comparative advantage is in those sectors where CEECs do not have a comparative advantage and vice versa. Nevertheless, the concentration of positive and negative effects on Germany and, to a lesser extent, Austria and Italy may explain their ambiguous relation toward the eastward enlargement of the EU in recent years. Trade effects on Spain and other South European countries could be lower than expected at the beginning of trade liberalization.

Central and eastern Europe still consists of countries that are more similar to each other than to other countries in the EU. Moreover, the trade structures of these countries have converged to some extent (with the exception of Bulgaria) as a result of similar reform and trade policies. In such an environment, trade diversion is likely to present a significant burden for countries omitted from the first wave of the enlargement (such as Slovakia and Romania).

The trade pattern of CEECs with the EU is very similar to that of Turkey with the EU. This explains Turkey's fear of not participating in European integration along with the CEECs already in the first wave of enlargement. However, Turkey's trade structure is much more different from the intra-EU trade than those of the other membership candidates. In addition to difficult political relations between Turkey and the EU, the unfavorable structure of Turkey's trade with the EU will represent a significant barrier for its integration into the EU.

References

Aturupane, Chonira; Djankov, Simeon; Hoekman, Bernard. "Determinants of Intra-Industry Trade Between East and West Europe," Discussion Paper 172, Centre for Economic Policy Research, 1997.

Collins, Susan M.; Rodrik, Dani. Eastern Europe and the Soviet Union in the World Economy, Washington, DC: Institute for International Economics, 1991.

Dixit, Avinash K.; Norman, Victor. Theory of International Trade, Cambridge, MA: Cambridge University Press, 1st ed., 1980.

Djankov, Simeon; Hoekman, Bernard. "Determinants of the Export Structure of Countries in Central and Eastern Europe," The World Bank Economic Review, 11, 3, 1997, pp. 471-87.

Fidrmuc, Jarko; Helmenstein, Christian; Huber, Peter. "East-West Trade in Transition," in Gary Cook, ed., The Economics and Politics of International Trade, Freedom and Trade: Volume II, London, United Kingdom: Routledge, 1998, pp. 217-35.

92 AEJ: MARCH 2000, VOL. 28, NO. 1

Fidrmuc, Jarko; Grozea-Helmenstein, Daniela; W6rg6tter, Andreas. "East-West Intra-Industry Trade Dynamics," Weltwirtschafiliches Archiv, 135, 2, 1999, pp. 332-46.

Learner, Edward E. "Wage Effects of a U.S.-Mexican Free Trade Agreement," Working Paper 3991, National Bureau of Economic Research, t992.

Neven, Damien. "Trade Liberalization with Eastern Nations: Some Distribution Issues,"European Economic Review, 39, 3-4, 1995, pp. 622-32.