wS' L31'l PoLIcy RESEARCH WORKING PAPER 1314 -The importance o6f the The Significance of theEuoenAocaon99 'Agreements-signed in 19 anc 199 has bedn- "Europe Agreements"unecoe -:Ucmo -- . .-a-0 .. for Central European hfntrd , X,=,=s, .n, -yh emergtnc o Eur ope I.-and. .Jio -~hf .ao tradin Be. .t.o-:e-K.m--s- The Slgnlficance or the -s- ~~~~~iiiovzi Tfigndil9mq:- The World Bank -- Indero alEconoics Departmt i . - Inerartioalo Trade Division June 1994 I POLICY RESEARCH WORKING PAPER 1314 Summary fIndings In 1991 and 1992, the European Union (EU) and the exports to the EU, significandly improve those countries' economics in transition of Central and Southern Europe access to EU markets. In 1992, the first year they were - the CEE-5 (Bulgaria, the former Czechoslovakia, in force in Hungary, Poland, and the former Hungary, Poland, and Romania) - signed the European Czechoslovakia, the Agreements freed slightly less than Association Agreements. The Agreements establish a new 50 percent of total exports to the EU from import duties framework for their mutual economic relationship, and nontariff barriers (NTBs). In terms of the 1992 including the transition to a free trade regime for composition of cxports, this "free trade" share in total industrial products. exports increases over five years to about 80 percent for The importance of the "Europc Agreemcnts" has becn the former Czechoslovakia, 60 percent for Hungary, and underscored by the rapidly shifting trade patterns 70 percent for Poland. between the CEE-5 countries and OECD markets, and Although there are significant differences in the by the emergence of the EU as their major trading composition of exports from CEE-5 economies affected partner. by EU trade liberalizing measures, these are the result of Kaminski examines the significance of the trade varying shares of sensitive (especially agricultural) concessions granted by the EU to the CEE-5 countries (1) products across countries, not dissimilar of concessions by analyzing the incidence of EU trade barriers on from the EU. imports from rhe CEE-5 before and after implementation The EUs negotiation approach, as revealed in the of the Agreements and (2) by identifying trade flows of Agreements, was to minimize the adverse effects of groups of industrial products subject to different opening up 'sensitive" sectors: the time and the pace of concessions. He focuses on trade liberaizing measures transition tends to be longer and slower for groups of for industrial products for which a free trade regime products with higher NTB-coverage ratios and higher should be in pLae no later than five years after the average tariffs. Whether by design or no; the variation Agreements are in force. (Excluded are tcexiles and in products identified in various provisions assures a clothing, discussed in the Uruguay Round of Trade more equitable treatment of CEE-S countries, judging Negotiations.) from their industrial exporc patterns in 1990-92. Overall1 the industrial product trade provisions of the Agreements, which affect about 80 percent of CEE-S Thispaper-aproductof the InternationalTradeDivision, International Economics Departmnent-is part of a larger effort in the department to analyze the new trading relations developing between Central and Eastern Europe and the European Union. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Minerva R. Pateraa, room R2-040, extension 37947 (37 pages). June 1994. Tbe Patscy Rejwrch Wormg Faproduced byte Policy Refse rch ssein pagr to Cencogc ter change of ideas about deond su An obiwcfm, of the series is to get l&~ s tis skildy, eve if the prcsc" tiomsare kss th"lbar : polisheA 77c popcrs carrdyJanens of Omc ashors and sbogl be used and cAdedacco;&ngly. Tl7c fiis&& idrreaons, and co.sclosiam ) tc abors' oum and shogld not be acs ibd to the lWorld Bank gts kexcutive Board of Diretctr, orany of its tmob counbim Produced by the Policy Research Dissemination Centcr THE SIGNIFCANCE OF THIE "EUROPE AGREEMENTS" FOR CENTRAL EUROPEAN INDUSTRIAL EXPORTS by Bartlomiej Kaminsld International Trade Division, The World Banik and University of Maryland, College Park Table of Content I. Inroduction .................................................. I II. Data, Methodology and Empirical Procedures .............................. 2 Im. Principal Features of Trade Regime as Envisaged in the iAs .4 IV. The Imporrnce of Industrial Products in CEE-5 Imports and Access to EC Markets ... ... 9 V. The Scope of Preferential Treatm of Industrial Products in the ITAs ............. 11 A-. le "One-Year-Delayed" Duty-Free Group ....... ..................... 12 B. The "Four-Year-Delayed" Duty-Free Group ........................... 14 C. The "QuotatFive-Year-Delayed" Duty-Free Group ....................... 15 E. The MFA Group .......................................... 20 D. TheECSC Group ......................................... 23 E. The "Residual" Free Trade Group . ................................ 26 VI. Time Profile of Attining Free Trade in dustrial Products ........ ............ 29 A. Copenhagen Concessions ....................................... 31 B. Reliability of free trade component estimates: How "Free' is a Free Trade Regime? . . . 32 VII. Conclusion .................................................. 34 References ....................................................... 36 i Tables 1. The Relative Importance of Industrial Products in CEE-5 Exports to the EC, 1988-92 .... 10 2. Pre-Agreement Market Access to the EC: Industrial Products versus Other Goods ...... 11 3. The "One-Year-Delayed" Free Trade Group: Exports to the EC and Pre-ITAs Market Access 13 4. The "Five-Year-Delayed' Free Trade Group: Exports to the EC and Pre-ITAs Market Access 15 5. The "Quota/Five-Year-Delayed" Free Trade Group: Exports to the EC and Pre-ITAs Market Access .16 6. Trade Measures Applied Against Selected Items of the "Quota/Five-Year-Delayed" Free Trade Group .............................................. 19 7. The MFA Group: Exports to the EC and pre-lTA Market Access ................ 22 8. The "Steel" ECSC Sub-group: Exports to the EC and pre-rrAs Market Access ........ 24 9. The "Coal" ECSC Sub-Group: Exports to the EC and pre-ITAs Market Access ... ..... 25 10. The "Immediate Free Trade" Croup: Exports to the EC and pre-ITAs Market Access .... 27 11. Composition of CEE-5 Industrial Exports to the EC, by ITAs group, 1988-92 ... ...... 28 12. Share of "Duty-Free" Goods in CEE-5 Industrial and Total Exports to the EC, 1992-98 ... 30 13. The Significance of the Copenhagen Concessions: Net Change in "Liberalized" Exports, Relative to the ITA Schedule .31 ui THE SIGNIFICANCE OF THE 'EUROPE AGREEMENTS" FOR CENTRAL EUROPEAN INDUSTRIAL EXPORTS' Bartlomiej Kaminski I. INTRODUCTION The European Association Agreements concluded between the EC and CentraltSouthern European economies (hereafter, the CEE-51) set a new framework for their mutual economic relationships. To emphasize the difference from the Agreements on Association signed earlier by the EC with other countries, they are referred to as the 'Europe Agreements." Recognizing the time-consuming process of ratifying the Association Agreements by parliaments of both EC members and Central European countries, it was agreed that the trade component-the so-called Interim Agreement on Trade and Trade Related Matters (hereafter, ITAs)-will be implemented before the treaty is ratified. Former Czechoslovakia (FCSK), Hungary and Poland signed the Association Agreements in December 1991, and the bilateral Interim Agreements on Trade went into force on March 1, 1992. Bulgaria and Romania concluded negotiations with the EC in 1992, and the trade provisions became operational in 1993 (Romania) and 1994 (Bulgaria). While some details in the Europe Agreements differ, they broadly follow the same pattern Their major provisions include: (i) the introduction of free trade in industrial goods within a period of 10 years; (ii) improved access for agricultural products, similar to that stpulatd in the Lom6 Convention and Medierranea association/cooperation agreements; (iii) a comnmitment to harmonize economic legislation with that in the EC; (iv) the EC's financial and technical assistance (albeit no specific amounts have been indicated); and (v) the possible introduction of free trade in services. In addition, the EC committed itself to gradually eliminating tariffs and/or increasing quotas on "sensitive" products-mainly textiles, iron, and steel. The objective of this paper is to dissect the trade provisions of the Agreements and identify the extent of change in market access in terms of tariff and non-tariff barriers as well as in terms of their current significance (using actual trade data through 1992) for CEE-5 exporters of industrial products. An imporant I would like to thmk Paul Armington, Ronald Duncan and Vikram Nderu for iheir suggestions and comments. I am also indebted to Maciej Lesny for his assistance in coilecting and processing data for this project. These include Bulgaria, the former Czechoslovalda, Hungary, Poland and Romama. 2 question whether the EC should have made larger trade concessions goes beyond the scope of this paper.2 The analysis is limited to industrial products simply because access to EC markets for agricultural products is governed by principles very different from the largely market-oriented trade policies covering industrial products. These products cover a broader group than convendonally-defined manufactures and account for around 80 percent of CEE-5 exports to the EC. No- attempt is made in this paper to attribute changes in CEE-5 exports to the EC to various provisions of these Agreements. - Too short a period has elapsed folowing entry into force of the Agreements. Moreover, faictors other than improved market access seem likely to be more important reasons for the improved export performance in OECD markets during the initial stages of the transformation. Thus, a proper causal analysis of CEE-5 export performance requires a more comprehensive ftamework? H. DATA, METHODOLOGY AND EMIRICAL PROCEDURES The improvement in market access involves reductions in restrictions. against imports either ftrough cuts in tariff rates and/or in NTBs (nontariff trade barriers). Since extensive use of NTBs often coincides with high triffs, i.e., product groups subject to nontariff measures are also subject to relatively high tarifs (Yeats, 1979), reduction in tariffs alone may not result in improved market access. With te decline i tarffs as the result of postrar trade negotiations under the GATT, nontariff barriers have become the major insment of protection. Thus, an assessment of the degree to whWch market access has changed must include NTBs. Information on tariffs and nontariff measures affecting industial imports from the CEE-5 was obtained from the SMART data base contining inter ala information on EC trade flows disaggregared to eight-digit Combined Nomenclature (CN) codes as well as tarff and nontariff measures applied by the EC 2 Available evidence shows that much deeper concessions would have'a strong welfae-ininasing effect in the CEE-5 without producing significant cost to the EC. See, among others, Messerlin (1992), Rollo and Smith (1993); and Wang and Wmiers (1993) 3 For an extensive anaysis of factors accountable for the improved export performanoe of CEE-5 economies in OECD markets, see Kaminsld (1993). 3 to these items.4 Industrial imports are broken down to groups identified in the Agreements in terms of the CM. For each CEE-5 country, the appropriate range of imports and of NTBs to be considered is denoted in the SMART data base. Two major indicators, extracted from SMART, are employed. The first is the proportion of imports subject to restraints, i.e., the NTB coverage ratio. The second is a simple arithmeic average of tariffs calculated over a group of tariff lines. There are three problems with using these measures as indicators of market access before and after the entry into force of the liberalizing measures stipulated in the ITAs. The first, relating to indicators of market access, is general-not country- .or region-specific. The problem is that both NTB coverage ratios and tariff averages tend to be downward biased in terms of measuring restrictiveness, albeit for different reasons. The NTB coverage ratio downplays the restrictive impact because imports of products facing NTBs are depressed, lowering their share in total imports.5 Average tariff rates are understated, especially on imports that are subject to GSP preferential rates within l[nits and above them to MFN rates. (SMART does not take account of these limits.) Second, the SMART data base has not been able to keep pace with the dramatic shift in trade paerns of the CEE-5 and the changes in their access to EC markets. Its trade data is for 1988 while its inventory of nontariff measures and tariff rates is for 1990. As a result, the data base incorporates the GSP status granted by the EC to Hungary and Poland in 1990, but not that extended to Bulgaria and the FCSK in 1991. The changes in export composition which have taken place affect more the NTB coverge ratio An the unweighted tariff rate average, as the latter is not sensitive to the size of imports but changes only when new products are added. Thus, the primary utility of these indices is that they point to areas where the reduction of tariffs or NTBs would have an impact on EC imports from the CEE-5. Another reason for treating the results of this analysis with caution is that NTBs are often used in response to increased pressure from imports. Growing levels of CEE-5 exports have already led steel and chemical producers to urge the EC to implement trade restraint. The irony is that some provisions of the 4The list of NTBs, containing 25 measures covers the major instrments used-by the Community to protect its markets. SMART can be used to generate information on average tariff rates (for a range of imports from a selected countay) and the percentage of imports affected by (selected) NTBS. The data base and procedures are described in UNCTAD/World Bank (1989). 5 For a comprehensive discussion, see Yeats (1979). 4 ITAs make it easier to erect extra barriers against CEE-5 exports (Ostry, 1993: 11). For instance, trade in steel had been governed by quantitative quotas and pricing arrangements. With the removal of these restrictions under the ITAs, CEE-5 exporters have become more vulnerable to anti-dumping actions. As a result, both the structure of NTB measures and their coverage may change as rapidly as CEE-5 exports to the EC. The empirical procedure identifies market access before entry into force of the Europe Agreements. The trade provisions of the ITAs distinguish among various products in terms of changes in their market access over the next ten years. These products are identified by CN codes. In order to use the available trade statistics, the CN items have to be converted into SITC (Standard International Trade Classification) equivalents. The SITC. Rev. 3 classification is used, because it is more extensive and quarterly trade data are available only in the Revision 3. Since SITC. Rev. 3 is broader fian other classifications, the loss of accuracy associated with moving from the very detailed CN scheme (around 9.6 thousand items) to a less disaggregated trade classification (3,118 basic headings or items) is least. Because export data for the CEE-5 are less reliable, the study uses trade sWistics of the EC. m. PRINCIPAL FEATURES OF THE ITAs The liberalization of EC-CEE-5 trade began in the late 1980s and climaxed with the signing of the Association Agreements in 1991 and 1992. Access improved significandy following the collapse of communism in Central Europe in 1989- Some trade liberalizing measures were implemented by the EC in anticipation of the successful completion of Europe Agreement negotiations.6 As a result, the level of restrictions on CEE-5 exports into the EC was considerably lower in 1990 and 1991 than in the 1980s and, by the same token, concessions granted in the ITAs improved market access in 1990-91. Except for Romania which enjoyed GSP status, until around 1988 there had been no significant 6 For instance, Hungary and Poland were granted GSP status effective on January 1, 1990, while it was grated to Bulgaria and the FCSK in 1991. The problem with GSP status is that it is at the discretion of the importing country and subject to periodical review. Therefore, an indisputable benefit of the EAs for the CEE-5 (as can be readily seen from comparing MFN and GSP tariff rates in Table 2) is that most tariffs levied on EC imports will be at least lowered to GSP rates, thus removing uncertainty concerning GSP status in the future. 5 differences in access of CEE-5 countries to EC markets (i.e., for the same tariff lines).7 GATT membership was not a differentiating factor because the EC conferred MFN status on all countries. MFN status did not mean most favorable treatment (as is the case, for instance, in the US market where GSP is the main device differentiating in terms of market access among various sources of supply). It only meant that their exports were subject to the same tariffs as EC imports from non-European industrial economies which, in turL, were considerably higher than those applied on imports from developing countries or European developed economies. Moreover, their exports were subject to restrictions imposed only on centrally planned economies. Because of the state monopoly of foieign trade, CMEA countries-including those which were GATT members-were defined as "state trading countries" exempt from GATT's Article 13, abolishing quantitative restrictions (Tovias and Laird, 1991:15). Basically, tariffs applied on EC imports from the CEE-5 were higher than on imports from developing countries and the EC used nontariff barriers against the CEE-5 with higher fiequency and their types were ".. among those generally considered most restrictive (i.e., quotas, variable levies, discretionary licensing schemes, etc.)" (Olechowski and Yeats, 1982). Thus, before the collapse of central planning in the 1980s. the CMEA had b_en at the very bottom of the EC preferential trading arrangement (Schumacher and Moebius, 1992).A The Europe Agreements (EAs) signed bilaterally between the EC and CEE-5 governments are essentially the same in terms of their structure.9 They are composed of a preamble, 122 articles grouped in 9 chapters, and annexes containing lists of goods included in the Agreements as well as separate protocols and declarations. The preamble of the EAs sets a framework for political cooperation and acknowledges that association with the EC should be conducive to full membership of the Community. Recognizing the time-consumig procedures of ratification of the EAs by the respective parliaments, the trade component of 7 Thi analysis excludes the former Sovie Union whose exports were subjet to more restfictive contls thn- applied against imports from other CMEA counties (see Kamiaski and Yeats, 1993). I It should be noted, however, that this position did not have a significantly adverse impact on their access to EC markets for at least two reasons: resort to the quantitative restrictions was limited; and, for products in which the CEE-5 had comparaive advantage there were few restictions (Pohl and Sorsa, 1992:54). 9 As of September 1993, the negotiations between Bulgaria and the EC were not completed. For the purpose of esdmating the ITAs' impact on Bulgaian industrial exports to the EC, it will be assumed that the Agreement will be similar in product coverage to that signed with Romama. 6 the EAs (Tide mI: Free Movement of Goods)-also referred to- Interim Agreement on Trade and Trade Related Matters--is to become effecdve within a specified period of time independently of the ratification of the EAa. Until the EA is ratified, all issues related to economic relations between the EC which are not covered by the TrA would be governed by the Agreements on Trade and Economic Cooperation (signed prior to the EAs).10 From the point of view of conditions relevant to market access to EC markets, the rTAs are almost identical. They provide for the establishment of a free trade area (excluding agriculwure) between the EC and each of the CEE-5 over the period of "... a maximum duration of ten years divided into two successive stages, each in principle lastng five years." Quantitative restrictions on industial products are diminated on the date of the enry irto force of the ITAs except for textiles and clothing and products listed in the Treaty of the European Coal and Steel Community (ECSC). The full hlberalization of access to EC markets would take six years, while barriers to EC exports will be eliminated over a longer transitional period." This means that CEE-5 exports of industrial products, including textiles and steel, will have duty- and NTB- free access to EC markets. In addition, market access of agricultural products would be enhanced, especially if the EC moves ahead with the proposed reform of its Common Agricultural Policy- scheme. Recognizing the problems that CEE-5 counties face in their quest to establish market economies, the EC has accepted asymmetrical treatment as a principle in some areas. The EC has -agreed to improve market access of Central Europe at a faster pace than its CEE-5 partners, and the initial reduction in protection by the EC is much larger than in the case of the CEE-5. Besides a shorter timetable for reducing trade barriers, however, the agreements allow the CEE-5 governments to unilatay apply additional import duties but for a limited period. These duty rates, however, cannot exceed 25 percent ad valorem and cannot apply to more hn 15 percent of the total value of industrial imports from the EC. Further, they can be implemented not later than three years after the estblishment of a free trade area in dustrial products. The ITAs also envisage a transitional period for upgrading economic legislation in the CEE-5 to EC ' Hungary and Poland signed nonpreferential Trade and Cooperation Agreements in 1988. Other CEE-5 countries signed these agreements in 1989 and 1990. . - 'There are difrces in transition periods envisaged for various countries: for FCSK and Poland it wiill last seven years and for Bulgaria, Hungary and Romania ten years. 7 standards. Among the regulatory measures affected are legislation on unfair competition and anti monopolistic regulations now in force in EC countries, to be deferred by three years and legislation regulating state assistance and subsidies, patterned after EC legislation,'2 whose implementation can be deferred by five years with a provision of extension for another five years. Until those regulations are fully implemented, the GATT subsidy code would be used to assess distortions in market competition caused by monopolistic pracdces and stare subsidies. The use of trade restricting measures is based on full symmetry, however. As the main objectdve of the ITAs is to phase out custom duties and NTBs, the ITAs contain clauses securing gradual implementation of free trade in products covered by the ITAs. According to the Agreements, neither new duties nor any other charges with similar effects can be implemented once the ITA is in force. The same rule (the standstill principle) applies to quantitative restrictions with the exception of agricultural products (not included in the liberalization timetable) for which both tariffs and NTBs can be freely changed. Furthermore, in line with GAiT rules, signatories may resort to various import-liniting measures inchlding anti-dumping, safeguard clauses (only if imports cause serious damage to domestic producers or disruptions in the economic siuation of a country or a region), protection against balance of payments disturbances; protective measures against disruptions in markets for agricultural products covered by the ITAs, as well as to introduce bans and restrictions permitted under GAIT rules. Products covered by the ITAs, identfied by their CN code of tariff items,'3 are arranged in three groups: industrial products (listed in Chapters 25 through 97 of the CN, excluding products listed in Annex I of the ITAs); agricultural products (CN Chapters 1 through 24, excluding fisheries); and -fisheries, covered by the EC regulation No. 3796/81 on the common organization of fishery product markets.'4 Quantitative restrictions on industrial products, except those specified in Protocols 1 and 2 of the ITAs (subject to the ECSC and MFA), are to be removed on the date of the entry into force of the Agreement. The transition 12 As specified in Articles 85,86, and 92 of the EC Treay (Pohl and Sorsa, 1992:59). 13 The CEE5 countries are obligated by ihe EAs to use the Combined Nomenclature in trade with the EC. '4 Although fishery products arc mentioned in the EAs, the relevant articles contain statements quoting. the EC regulation applicable to these products (listed in two separate annexes) and pronmse further concessions n,. on a harmonious and reciprocal basis." schedule for eliminating duties and quantitative restrictions on imports from the EC varies among CEE-5 countries-but within ten years all quantitative restrictions on imports from the EC are to be abolished. Industrial exports from the CEE-5 will benefit from free trade access to EC markets within five years with the exception of textile and clothing products (tariffs eliminated at the end of the sixth year of the Agreement, while the elimination of quantitative restrictions depend upon the outcome of the Uruguay Round). The rules of origin (laid out in Protocol No. 4) stipulate that 60 percent of the value of goods exported under preferentidal treatment should consist of local or EC content. Thanks to this rule, possibilities for subcontracting have significantly expanded. On the other hand, however, the rule is quite restrictive, especially for non-EC potential investors in manufacturing activity-since in the initial stages new production capacities often have to rely on imports of parts. Some expressed concerns that the rule may keep non-EC investors out of the CEE-5 (Inotai, 1993). The EAs envisage enhanced market access for agricultural products. Some quantitative r_strictions (Regulation 3420/83) are to be abolished immediately and others will be either liberalized gradually or maintained pending the outcome of the Uruguay Round. The EAs affect five main product groups: meat, live animals, fruit, vegetables and processed agricultural products. Trade in grain is not covered by the Agreements. Agricultural exports will be permitted to increase by 10 percent in each of the next five years. Duties on listed food products will be reduced by 69 percent, and CEE-5 countries will grant similar concessions to the EC. It remains to be seen whether these provisions will help agricultural exports from the CEE-5. To sum up, the significance of the EAs goes beyond the fall establishment of a normal GATT-based trade regime. It assures the very fast movement of the CEE-5 economies to the top of the pyramid of EC preferential trading arrangements. Although the issue of EC membership is not explicitly addressed in the EAs, the Agreemenss do recognize that the objective of CEE-5 govermnents is to join the EC. Even more significandy, the way that they have been structured leaves this option open. Furthermore, the adjustment in the institutional and legal framework imposed by the Agreements has been designed to bring CEE-5 economic systems into line with the EC. 9 IV. THE IMPORTANCE OF INDUSTRIAL PRODUCTS IN CEE-5 ACCESS TO EC MARKETS Industrial products, as defined in Chapters 25 through 97 of the CN, consist not only of manufactures but also of some agricultiral materials, mineral fuels, and ores and metals. The equivalent of this group in the SITC is much broader than a standard description of manufactures. Accounting for products excluded from the provisions concerning industrial goods,'5 this group includes the following SITC .Rev. 3 items: (2-21-22-29-24403-24404-2631-2632-2651-2652-26851) + 3 + (5- 59223)+6+7+8+(9610+971). The provisions concerning industrial products are the most significant part of the lTAs for two reasons. First, these products account for more than three quarters of EC imports from the CEE-5 (see Table 1). With 93 percent of their exports to the EC fallig into this group, the FCSK and Romania had -he largest share among CEE-5 countries. Bulgaria and Hungary, with strong specilization in agricultural products, had the lowest shares. The importance of the industrial product group is further underscored by the fact that the share of industriat produc in EC imports increased significantly for all CEE-5 economies in the 1988-92 period, except for Romania. Some portion of this increase in 1992 may be attibuted to a 5-8 percent fall in agricultural imports caused by adverse weather conditions." However, this decrease was more than offset by increased imports of industrial products triggered by the decline in domestic demand following the implementation of stabilization-cum-transformation programs and the collapse of previous intra-CMEA trade.'7 15 Anmex 1 of the ITAs identifies 10 six digit CN items faling into chapters 25-97 but excluded from this group: albumins (CN 3502), natural cork (CN 4501), cotton, not carded or combed (CN 520100), flax, not spun (CN 5301), and hemp, not spun (C:{ 5302). In terms of SITC.Rev.3., these are: 0253, 59223, 24403, 24404, 2631, 2651, 2652. 16 The value of imports fell by 5 percent from Bulgaia, by 7 percent from Hungary, and by 8 percent from Poland. 1' The exception was Romania whose exports comtinued falling (in 1992 they fell by -0.5 percent). Other CEE- 5 economies recorded double digit growth rates. As a resut, between 1988 and 1992 the value of industrial imports from Bulgaria increased by 115 percent, from the FCSK by 183 percent, from Hungary by 127 percent, and from Poland by 152 percent. 10 The link between the transformation programs and the growth in exports of industrial products is apparent in data presented in Tables 1 and 2: the share of industrial products in exports to the EC (see Table 1) increased during the first year of the program (Poland-1990; the FCSK-1991; Bulgaria-1992); and the average growth rate of industrial exports in the 198891 period was higher than that of non-industrial exports for the txoika countries while it was lower for the two Balkan countries (Bulgaria and Romania). Table 1: The Rdative Imporbace of Iudustrial Products in CEE-5 Exports to the EC, 19892 Exports of Industrial Products Sbare in Total Exports to the EC (value in US$ million) (n percent) 1988 1989 1990 1991 1992 1988 1989 1990 1991 199Z Bulgaria 418 430 537 677 899 73 72 72 73 79 FCSK 2367 2515 3053 4643 6687 91 89 90 92 93 Hungary 1777 1961 2795- 3331 4034 69 68 74 73 77 Poland 3092 3230 5089 6247 7793 78 75 77 81 85 Romania 2522 2649 1908 1706 1697 95 95 97 94 93 Since Greece, as of July 1993, has not reported trade data for 1992, its ammal imports estimated on the basis of first the quarters. Source: the United Nations CONTRRADE data base Eaier market access for industrial products than for agncltural products has clearly faciltated this increase in exports to te EC. As can be seen from data compiled in Table 2, indlstrial exports fiLced significantly lower tariffs hn other exports in 1990. This difference was particularly large for countries which had GSP status-Hungary, Poland and Romamia. The proportion of export affected by NTBs was also substantially lower for industrial products; the NTB coverage ratio for "other" exports was around double that for industial exports. Although tariffs have lost much of their protective significance, the margins of preference between CEE-5 exporters and those who did not have GSP status were quite large. On average, MEN rates imposed in 1990 on inports from Bulgaria and the FCSK were around 7 percent as compared with GSP rates of 0.1 percent on imports from the other CEE-5. GSP staus sigificantly improved CEE-5 access to EC markets, and its short-erm impact on exports was larger than that of the ITAs (Inotai, 1993). As can be seen from Table 2, GSP status does not have much significance for non-industial products. The level of vulnerability to nontariff barriers, as measured by the share of imports subject to NTBs, reflected the differences in export baskets among CEE-5 countries. The average NTB ratio was larger for countries with higher shares of agricultural products, iron and steel, and textile and clofting in their exports. For instance, Bulgaria and Hungary had the highest NTB coverage. ratios, .mainly because of the high shares of agricultural products. At the other end, imports from Poland and Romania were least-affected by NTBs because of the less restrictive access for labor-intensive engineering and consumer goods as weil as fuels and ores and metals. Table 2: Pre-Agreemeut Market Access to the EQ Industrial Products versus Other Goods Index, 1991 NTB Coverage Ratio Simple Average Tariff Rate- 1988=100 Indust Otler Indust. Other Indust. Other (in percent) (in percent) - Bulgaria 162 165 22.5 48.3 6.9 - 11.6 FCSK 196 163 24.0 52.5 7.0 11.7 Hungary 188 156 24.2 57.7 0.1 9.4 Poland 202 165 23.6 48.6 0.1 10.5 Romania 68 85 28.4 59.8 0.0 -- 8.6 Sources: Derived from the UN COMTRAfDE and UNCTAD-World Bank SMART data bases. V. THE SCOPE OF PREFERENTIAL TREATMENT OF INDUSTRIAL PRODUCTS IN ITAs- The lTAs distinguish six various groups of idustrial products with different schedules of transition to free trade and different mixes of trade liberalizing measures, i.e, tariff and nonmariff measures. As for the latter, quantitative restictions on imports to tl. EC are abolished on the day of entry into force of the Agreements. The groups of indLstrial products subject to different time schedules are as follows: (i) the "one-year-delayed' free trade group (duty-free access envisaged in the second year of the rrAs); (ii) the "four-year-delayed" free trade group, i.e., at the beginning of the fifh year tarfs are eliminted (iii) the "quotalfive-year-delayed" free trade group (free trade at the beginning of the sixth year of the ITAs); (iv) the ECSC group (tariffs on steel fully are eliminated by the end of the fifth year and tariffs on coal imports .. . . . .. . .~~~~~ 12 into the EC, excluding Germany and Spain, are eliinated at the beginning of the second year for the FCSK and Poland and by the end of 1995 for other CEE-5 coal exporters. Duties on imports into Germany and Spain will be abolished by the end of dte fourth year of the ITAs); (v) the MFA group (quotas bilaterally negotiated and the shift to a free trade regime tied to the Uruguay Round of trade negotations); and (vi) the "immediately" free trade group (a residual group of industrial p:oducts including items not covered by separate provisions). While the detailed provisions are similar, there are some differences in assignment to the above- mentioned groups. For instance, tariffs on aluminum oxide and hydroade (CN. 2818200 and 2818300) on imports from Romania are held to a four year transition to duty-free sus (group ii) whereas on imports from other CEE-5 countries these are subject to a one-year schedule (group i). The only area where there are significant differences in both coverage and triff and quotas concessions is in the "quota/five-year- delayed" free trade group. Hungary was granted a different schedule of tariff reductions (by 10% annually, other CEE-5 countries by 15%) and quota/ceiling increases (15% per year, others 20% per year). The list of discrepancies for this group is the longest. While it is unclear whether it was intended by EC negoatr, the dispersion in shares of this group in 1991 industria imports into the EC is significantly lower tian in shares of a group including products listed in all ITAs. MFA and ECSC groups cover the same CN items. In addition, duties on coal imports into the EC (excluding Germany and Spain) from the FCSK and Poland were eliminated by the end of 1992, while on imports from other CEE-5 countries by the end of 1995 (on January 1, 1994, they will be reduced by 50%)_ The Copenhagen Summit of the EC offered additional concessions. As far as industrial products are concerned, these included acceleration of custom duty eliminatim: (a) on products of the "four-year- delayed" group (duties are to be abolished wihin two years); (b) ECSC-steel products (duties are to abolished after four rather than five years); and (c) the MFA group (after five years rather than six years). A. The "One-Year-Delayed' Free Trade CGroup The "one-year-delayed" free trade group comprises products for which customs duties are reduced by 50 percent upon entry into force of the Agreement and eliminated at the beginning of the second year. In terms of eight digit CN items this is a relatively small group consistng of 92 tariff items. It includes 13 mainly industrial raw materials. Imports from the FCSK and Poland increased quite dramaticaily in the 1988-91 period, although they were very low in the base year. Since these increases were probably triggered by contraction in industrial activity during the first year of the transformation program, their future growth potential may be limited. This group accounts for a small share of total CEE-5 industrial exports to the EC (in 1991 their share was below 2 percent). However, the fast move to duty-free access is a more significant concession than the share of these imports might indicate. First, average NTB ratios on EC imports of this group from CEE-5 countries (except for Poland) are significandy higher than on industial products in general (see Table 3). For EC imports from Bulgaria, the rado is by 11.8 percentage points higher, from the FCSK 16 percentage points higher, from Hungary 28.4 percentage points higher, and from Romania 24.4 pernnage points higher (see Table 2 and 3). The average NTB ratio for EC imports of ail industrial products from Poland is 8.2 percentage points lower than for imports falling within the one-year-delayed free trade group. Second, with the exception of the Balkan countries, CEE-5 exports of this group expanded rapidly in the 1988-91 period. The average growth rate was substantially higber than the average for CEE-5 all industrial exports, especially in the case of exports from the FCSK and Poland. Moreover, exports from these countries crowded out other suppliers, as their share of these EC markets increased from 0.1 percent (both for the FCSK and Poland) in 1988 to 0.4 percent for the FCSK and 0-6 percent for Poland in 1991. While it is unlikely that such a rapid expansion is sustanable over the medium-term, the elimination of trade barriers has significantly improved their competitive position. Table 3: The "one-year-delayed' Free Trade Group: Exports to the EC an d Pre-ITAs Market Access Exports Share in Industrial Index, 1991 NTB Coverage Simpb Average Range of 1991 Exporis, 1991 1988=100 Ratio Tariff Rate tariff mres Max-NIin (US$ min) (in percent) Bulgaria 10 1.4 93 33.3 5.1 6.0-0.0 FCSK 53 1.1 400 40.0 4A 18.3-0.9 Hungary 23 0.7 126 52.6 3.0 7.04.0 Poland 27 OA 500 15A 6-2 18.3-0.0 Romania 1 0.0 60 40.0 2.0 3.240 Sources: See Table 2. 14 But the evidence is inconclusive yet. In 1992 the 50 percent margin of tariff preference had some impact on imports from the FCSK and Poland but not on imports from Hungary. The value of imports from the FCSK and Poland increased by 60 and 85 percent respectively while that from Hungary fell by 10 percent. The two Balkan countries-with no preferential access in 1992-recorded contractions by 17 percent (Bulgaria) and 26 percent (Romania). B. The "Four Year-Delayed" Free Trade Group This group contains products for which tariffs are reduced by 20 percent oi "..e day of entry into force of the ITAs, and then lowered by 20 percent annually so that they are fully eliminated at the end of the fourth year. There are significant differences in the list of products specified in Annex Ilb of each ITA. The ITA between the EC and the FCSK contains the smallest number of items (3 eight digit CN items), while the list in the Polish ITA is the most extensive containing 16 CN items. Leaving aside the variation in products appearing in respective Annexes Ilb of the ITAs, they all are primary intermediate goods, i.e., lightly-processed, resource-intensive products such as ferro-manganese (with a carbon content of less than 2 percent), ferro-silicon, unwrought alumium, and zinc and lead alloys. Excluding Romania, the group accounts for a minuscule share of CEE-5 exports to the EC. In the 1988M91 period EC imports of these products from the CEE-5 were highly volatile." A quick perusal of data compiled in Table 4 suggests that duties levied on this group -are relatively low and the NTB coverage ratio varies significantly across CEE-5 counties with imports from the FCSK subject to full coverage and those from Romania enjoying "NTB-free" access. However, this group has limited potential for growth. The provisions of the ITAs are not likely to provide stimulus to eorts with the possible exception of Romania once its economy rebounds. Although the growth rates of EC imports from the FCSK and Poland were impressive, EC imports of these from the region fell in the 1988-92 period. Moreover, neither a one time upswing in Bulgarian exports to the EC in 1989 (tebir value almost tripled) nor steady growth in Hunprian exports uitil 1990 were sufficient to compensatc for the contraction in Is EC imports fiom Bulgaria increased by 164 percen in 1989 and subsequendy contctedin the 1990-92 period. Imports from the FCSK increased by 28 percent in 1989, by 127 percent in 1990 and by monlyn 8 percent in 1991. The value of imports from Hiungary increased in bodL 1989 and 1990 (by 22 and 36 percen respecdvely) and fell by 36 percent in 1991. Imports from Poland fell in 1989 (-6.4%), tripled in 1990 and doubled in 1991. 15 Romania's exports. Among the CEE-5, Romania used to be the most important supplier of dtese products to the EC. However, Romanian exports collapsed in the 1988-91 period, with its share in EC imnports falling from 1.9 to 0.4 percent. The shre of the CEE-5 fell, however, from 2.3 percent to 1.6 percent. Table 4: The "Four-Year-Delayed" Free Trde Group: Exports to the EC and Pre-fTAs Mirket Access Exports Share in Industrial Index, 1991 NTB Coverage Simple Average Range of 1991 Exports. 1991 1988=100 Ratio. Tariff Rate tariff rates Max-Min (in USS million) (i percent) Bulgaria 1 0.1 132 20.0 3.3 6.2-0.0 FCSK 2 0.1 312 100.0 3.1 6.2-0.0 Hungary 17 0.5 107 40.0 3.6 6.2-0.0 Poland 49 0.8 593 13.3 3.3 6.0-0.0 Romania 27 1.6 21 0.0 3.2 6.0-0.0 Sources: See Table 2. During the first year of the lTAs, exports of these products fell (with the exception of those from Poland which continued to expand by 18 percent). Despite a 20 percent reduction in tariffs, the value of exports from the FCSK fell by 38 percent and from Hungary by 35 percent The contraction in exports from Bulgaria (-52%) and Romania (-77%) was even larger. C. The "Quota/Five Year-Delayed" Free Trade Grou_p The trade liberalizing measures in the ITAs for this group of industrial products are a mixtur of cuts in custom duties and increases in tariff quotas and ceilings. Custom duties are suspended within the limits of tariff quotas to be increased annually. The ITAs contain different stipulations concerning the annIua growth rate of tariff quotas: 15 percent for imports from Hungary, and 20 percent for imports from other CEE-5 countries. Simultaneously, custom duties on imports in excess of quotas are to be reduced progressively to zero by the end of the fifth year. The schedule of reduction in these customs duties, beginning on the day of entry into force of ITAs, calls for annual cutbacks of 15 percent on imports from Bulgaria, the FCSK, Poland and Romania and 10 percent on imports from Hungary. ITAs share the same stipulation that all duties be scrapped by the end of the fifth year, however. 16 The group subject to these provisions is the largest in terms of CEE-5 imports into the EC, accounting for between one fourth and one third of their industrial inports (see Table 5). It includes products of most industrial sectors (organic and inorganic chemicals, some leather products, cork and wood products, glass, some steel products not covered by the ECSC, copper and copper products, electric machinery, optical goods, plastics, footwear, clothing accessories, furniture, motor vehicles, toys, etc.). Furthermore, the group is the most diversified in terms of coverage in separate ITAs. In contrast to other groups, there are substantial differences in the lists of items in the individual Agreements. Lists for Hungary and Poland are more extended in terms of the number of CN items.'9 Table 5: The "Quota/Five-Year-Delayed" Free Trade Group: Experts to the EC and Pre-ITAs Market Access Exports Quoas Share in Industrial Index 1991 NTB Coverage Simple Average Range of 1991 for 1992 Exports t the EC 1988=100 Ratio, Tariff Rate, tariff rates percent Max-Min (in million of 118$) - ---(in percent) Bulgaria 111 na 16.3 243 18.8 8.6 25.84.0 FCSK 1230 379 26.5 213 20.6 8.7 25.8-0.0 HungarY 810 484 24.3 175 21.0 0.0 0.0-0.0 Poland 1477 575 23.6 213 21.7 0.0 0.0-0.0 Romania 516 366' 31.4 83 23.7 0.0 0.0-0.0 'for 1993 Sources: See Table 2 and lTAs (Annex 3). Pre-ITAs market access was better for this group tan for all industial products. The NTB coverage ratio was lower (by 24 percentage points) than for all industrial products (see Table 2 and 5). The average tariff rate was higher than that on all industial products for Bulgaria and the FCSK,1 but lower for other CEE-5 countries. For the reasons discussed in Section II, the "real" average tariff rate for the latter '9 The total group, including items- specified in all ITAs, covers 678 eight-digit CN items of which 677 are subjea to GSP rates. 21 Imports from these two countries enjoyed GSP status in 1991. Hence, the average rare reported in SMART would be near zero, as almost all items in this group had GSP teatmet. 17 countries was higher depending on tariff quota utilization.? It is difficult to assess the extent to which the ITAs improve market access for this group of idusal products. First, for some products the EC sets quows and for othes ceilings-imports below ceiigs and witin quotas dre duty free. This is an important disdnction, complicating assessment of a 'liberalized" component in this group. Duties on imports of products exceeding quotas are imposed automatically whereas duties on those exceeding ceilings are levied only if domestic EC producers demand it As a reu, it is impossible to predict the size of "above ceili" imports that wil be subject to tariffs. Second, one should note that this group on average faces hgher MFN tariffs (applied on "above quota" imports) an industrial products in total. Since the base for tff reducto envisaged in the Agreements is about 50 percent higher than the average on industrial products,2 the margins of preference for CEE-5 exporters are substantial. Taking into account that triff rates are the same for large clusters of CN items, one may assme that the simple average auiff rate will be fan in line with the schedule of tariff reductions for this group. Third, the ceiling or quo a utilization' ratio varies across different products. If quotas were set in a fixed relationship to previous export p ance, then the duty-free compoent of this group durng the first year of the ITAs woud be equal to the total value of quotas.2? But this is not the case: neither growth rates of quotasfceilings were set m te ITAs in any relationsbip to import growth rates nor quotas or ceings reflected earlier levels of imports. For instance, average growth rates of iports from both the FCSK and Poland in the 1988-91 period were around 12 percentage points higher than scheduled ammal increases in quotas (20%) and from Hungary were 6.1 percentage points higher ftam the scheduled increase 21 SMART keeps track only of dutics levied upon morts not exeding tariff quotas and clngs. Ims xceding thse hmits wer subjec to MFN rats independy of dtdr GSP status. For imports exceeding quotas or oeiling, the simple aveage MFN tariff rate is awmnd.9 percent, while average taiffs on all industial products are about 6 pernt. 23 Had the fre trade compoet been equal to quota, it would have been te largt for-Hungary and Romania (with 69 percent of imports falling into the free trade component), followed by Poland (38 percent) and the FCSK (30 percent). This meas that only 31 percent of Hugaian and Romanian imports into the EC in 1991 would be subject to MFN rates, while the s ratios for the FCSK and Poland would be 70 percent and 62 pert wespeclvey. However, tis xould not ggest that Hunga and Romania have obained a better deal from the EC haoher CEES govenmenmt 18 (15%).2 Most quotas set in he ITAs with the Visegrid counies are the same in of their past export performance.? For iustance, the Polish quota for polyetylene is the s-me as that for the FCSK and Hungary, while its exports to the EC in 1991-were a small fraction of other CEE-3 exporls (see Table 6). TIis lack of a link between tarff quotas and past export performance is especially manifest in the cas of cars. The tariffquota of US$ 91 million set for motor vehicle imports (CN.87032110 through 8703909) from the FCSK (ECU 80.483 million as compared with ECU 125 million for ie other two ccmmtries) is about one third lower than that for -Hungary, even though in 1990 FCSK inports it the EC were much larger than Hungrian ones (US$ 84 millon as compared with US$ 2 million)?' It is wort nodng hat actual imports from dte FCSK were significanly larger than quotas for all products listed in Table 6 (especially for wire of iron or non-alloy steel whose. share of quota was only 6 percen). However, this did not prevent FCSK producew from increasing exrs of these products in 1992-the only excption hi the sample was polyethylene whose inports i the EC declid during te first year of the [TA. WiftIo a detailed analysis of export capacity for each tariff quota/celing group, which goes beyond the sope of this paper, no firm judgment n be passed on the extent of EC concessions. Somen isgb can be derived from the sample of products presented in Table 6, however. This sample, accounting for 30-35 percent of the w alue of quotas and for 18 t0 29 percent of troik exports of these produts to the BC in 1992, is quite large and, terefore, vepfor this group? The duty-free compone vares 24 This should not suggeast that te rrAs do not contain siifict ocessions. in fact, market access for thes products will move to duty-fee trade over a five-year period. 2s Examples of different tariff quotas or ceilings interaa indclude: the FCSK has lower quotas for acids and salts (CN. 2981400)-ECU 210,000 as compared with ECU 368,000 for Hunar and Poland), for leather clothing and apparel accessories (CN.4203100, 42032100, 42032991, 42032999, 42033000,42034000-ECU 4.3 million as compared with ECU 6.6 miion). Tbe FCSK and Hungay have smaller quota for fibrbords (CN.4411-ECU 4 million vsus ECU 7 million for Poland). The quotas for Bulgaria and Romania are an avere around 5 percent higher than for the CEB-3. 2 On the odter band, it could have been purely acdental that the Polish quota (the same as the Hungarian) was close to the value of its imports into the EC in 1988 (US$ 145 miion) and 1989 (USS 128 million). Latecomers-Bulgaria and Ronmania-=a not included, simply because dir respcive Agrmnt wit the Co is went into force a year lat. 19 Table 6: Trade Measures Appied to Sdected Iteu of tho "QboIlvYer-Delaedroe Trde" group -FCS Hungary PoLand NFNs(lumLOeverage tariff rate In ?srngwr Newr VekWdex (S .3.7J12) lO Tariff Quota CmiLlIon of US$. 1992 91 141 141 Imports (miLlion of USS$, 1992 m 25 Z59 1991 210 3 39 qltiliztiun retito in prmt). 1992 299 la l Duty free cawonent Cin percent), 1M 33 100 5S Polykae (SriC. r.3.57112) Tariff Quota (miLlion of USS), 1992 15 15 15 [Worts mtillion of US$3. 1992 20 30 1 1991 34 31 1 qUtiLizntiun ratio (Sin purmtu. 1992 IN 201 7 Duty free camnet (in percent), 1992 76 50 100 Fksul*n (SITC.Rv.3.8213) LA Tariff Quts (milLiLon of US$3, 1M 7B 0 75 Ieqorts (million of US$3 1992 121 66 242 1991 61 54 192 UUtiLiz.tio ratios Cin perct),1992 156 P/a 311 Duty free coonent (in percent), 1M 64 0 32 Appwd and CloSing mui of monxlu (SI TC.e3.UI1+ IE 477) Tariff Guots CmiLLion of US$3, 1992 5 8 a Imports (million of USS$, 1992 11 29 6 1991 8 24 10 ftiLizatimn ratio Cin pec t) 1992 23Z 3C6 aZ Duty free component (Cn percent), 1992 43 26 100 Glaze Cemmlcs (SlltRet.3SZdS) Teriff Quota (miLlion of US$), 1992 4 4 4 Imports (million of US$), 1992 Z 7 3 1991 13 7 .7 qUtiLizntim ratio" Cin pero nt), 1992 6w3 152 7 Duty free cowponent Cin percent), 1M 16 66 100 VFWi of Zi or nwxn-aey Med (SflYZ.Zev34fh1Z-12) Tariff Quota (million of USS3 , 1992 2.2 2.2 2.2 leports (million of US$3, 1992 35 2 5 1991 33 3 6 UitiLization ratio" (in pucuit), 1992 1644 80 VA8 Duty free couponent Cin percent), 1M 6 10W 40 Mersmundaim (A.) Share of the sample in: - total imports of the grcoup (in percent). 1992 30 i8 27 1991 29 16 16 - total quotas for the group (in percent), 1992 51 35 43 (B3 Share of duty free iWports Cin percnt). 1992 40 34 47 Source: see Table 5. 20 widely for products and countries.2 This suggest that the EC was concerned mosdy with proecting domesdc markets and ignored dte interests of CEE-5 producers. Nonethdeless, CEE-5 coumtries have obtained significant concessions. As can be seen fom Table 5, a subsmandal proportion of imports of products bad duty-free access to EC markets. Its share in exports for this sample was between 34 and 47 percent. Furthermore, the margins of preference provided in the ITAs for some exporters are quite substantial. For instance, MFN tariffs on polyethylene and cars are relatively high (10 and 12.5 percent, respectively): since CEE-3 suppliers of these products had free access to EC makets for a very significant proportion of their exports (between 33 and 100 prcent), this has undoubtedly given them a subsantial compettive edge over MN suppliers. Finally, in terms of 1992 exports to the EC, tff and ceilings/quotas seem to be on average less 'restraining" on cars than on other products. This conclusion can be drawn from comparing shares of passenger car quotas in total quotas with the shares of these exports in eos of tis group. Tariff ceilings and quotas applied against exports of this group focus on passenger motor vehicles, accountig for 47 percent of the FCSK's value of quotas for this group, for 79 percent of Hungarian quotas, and for 55 percent of Polish quotas. Their shares in exports of this group are 16 percent for the FCSK, 3 percent for Hungary and 14 percent for Poland, suggesting that quotasfceilings free trade component of their exports is considerably larger in products which are not listed in Table 6. C. The MPA Grou, The ITAs contain a special protocol addressing market access for texiles and clothing (CN 50 thugh 63 exchluing 520100, 5301, 530?') subject to the MFA restricting "... the volumes of most imported textiles and clotung products into North America and Westn Europe from developing countries" 25 The "duty -free component is equal to the sum of quotas (if the value of imports exceed the tariff quota) or the value of exports (if it is lower than quota). Excluding cars, the duty free shar raises for the FCSK from 40 to 48 pern, and faIls for Hungy from 34 to 22 pect and for Polamd fm 47 to 40 paen. 2' This is the equivalt of SITC. Rev.3. (26-2632)+(65-65911)+(84848144812-8813). 21 (Hamilton and Martin, 1990:2).3Y The MFA group contains 1,296 eight dlgit CN items, of which 188 are subject to GSP rates (O percent).? Textiles and clothing products rank second in tenms of sharcs in industrial product exports. Tariffs for these products are to be gradually eliminated by the end of the sixth year: in each year they will be reduced by one seventh of ihe level prior to entry into force of the ITAs. Market access for these products is governed by quantitative restrictions rather than tariffs. Average tariffs for countries having GSP stams are close to zero. The maxinum rates are higher for Bulgaria and the FCSK, simply because other CE-S countries did not export any items subject to this rate. NTB trade coverage ratios for these products, which are the highest among the groups of industrial products, will not be affected by the ITAs. Protocol 1 of the ITAs, containing provisions applying to trade in textile and clothing products, provides that quotas will be negotiated bilaterally and new arrangements will be implemented *... as soon as the future regime goveniing international trade in textile products has emerged from the multilateral negotiations of the Uruguay round. "(Protocol 1, Article 3.2). Quantitative restrictions will be abolished over half of the period decided in the Uruguay Round negotiations. Since the MFA restricts the volumes of imports through quotas, exports can increase by filling previously underutilized quotas or obtaining increases in quotas. The increase in CEE-5 exports to the EC was the result of a combination of these two factors with the latter having probably a larger influence. The CEE-5 had exported well below their quotas throughout much of the 1980s.32 With a quota utilization rate lower than 90 percent in the 1980s, they were not binding for around 30 percent of MFA quotas (Erzan and Holmes, 1992). In 1990 and 1991 the EC signed bilateral agreements increasing quotas and market access for re-imports into the EC. CEE-5 economies (excluding Romania) took advantage of the improved market M Although technically they were not qualified as developing countries, CEE-5 economies have been subject to the provisions of the MFA. 3' This relatively small share confirms the observation that GSP schemes, unaterally grated by industial countries, as a rule exclude major textile and clothing products (see, e.g., Erzan, Holmes and Safadi, 1992) 32 For instance, the EC quota utilizationrates in 1982 were 65 percent for Bulgaria, 79 percent for the FCSK, 43 percmt for Hungary, 35 percent for Poland, and 73 percet for Romania (Trela and Whaullcy, 1990:19). 22 access, as their exports increased dramatically in this period.? Taking into account that MFA imports from these countries increased in 1990 by at least one third (see footnote 32), this expansion would not have occurred without increases in EC quotas.34 Table 7: The MFA Group: Exporfl to the EC and Pr-IrAs Market Access Exports Share in Industrial Index 1991 NTB Coverage Simple Average Range of 1991 Exports, 1991 1988=100 Ratio, Tariff Rate tariff rames Max-Min (in USS million) (in percent) Bulgaria 142 20.9 227 90.6 10.8 17-0.0 FCSK 630 13.6 202 87.6 10.7 17-0.0 Hungary 712 21.4 175 85.1 0.1 9.3-0.0 Poland 1099 17.6 258 88.8 0.0 8.6-0.0 Romania 480 28.2 96 86.2 0.1 ".3-0.0 Sources: See Table 2. On -the supply side, it is noteworthy that EC's expanding inports of textiles and clothing from the CEE-5 coincided also with the implaton of transformation-cm-stabilization programs, which provided a boost to MEA expo ts. For instance, the value of MFA imports from Bulgaria rose by 87 percent in 1992 (the first full year of the program in place), from the FCSK by 53 percent in 1991 (and 46 percent in 1992), from Hungary increased by 43 percent in 1990 (by 24 percent in 1992), and from Poland by 74 percent in 1990 (35 percent in 1992). Overall, between 1988 and 1992, the value of MFA imports into the EC from Bulgaria increased by 324 percent, from the FCSK by 194 percent, from Humgary by 116 percent, from Poland by 248 percent, and from Romania by 32 percent.' 33 It is apparent when comparing thetanmmal rates of growth of MFA imports into the EC in 1989 and 1990. The rate of growth of imports from Bulgana rose from 8 percent to 52 pemnt, fom the FCSK fram (-)1.2 to 33 percent, from Hungary hom 3.5to43 percen, and from Poland from4. ito 71 percent. The value of imports from Romania fell by 1.6 percent. 3 For instnce, in 1992 the aveage utilization ratio for Polish imports of textiles and clothing product into the EC was 33 percent (Dunin-Wasowicz, 1993:18). 3 As a result, its shr in their industrial imports into the EC increased especially in the case of Balkan countres where it almost doubled-for Bulgaria, the share rose friom 15 percent m 1988 to 29 percent in 1991 and for RomaWia from 20 persent to 39 percent (this was the only group of industrial products discussed here that recorded an increase in 1992);: 23 Whether this expansion is sustinable will depend on domestic developments and continued inproved access to EC markets. As for the former, there is one important factor pushing to expand supply capacities. This sector is particularly attractive to private entrepreneurs, because of the low labor cost combined with the sector's low average capital intensity in a situation of capital shortage and deficient financial markets. The EC may choose not to erect barriers to CEE-5 exports, simply because EC producers bave increased MFA exports to CEE-5 countries. Furthermore, the experience of many developing countries suggest that quotas have not been effective in preventing flexible and innovative firms from expending exports.36 D. The ECSC Group The ECSC product group, accounting for a significant share (albeit rapidly declining) of CEE-5 exports (from Bulgaria a 12 percent of industrial exports in 1991, from the FCSK 14 percent. and from Poland 12 percent), was not treated uniformly in terms of concessions granted by the EC. The new rules of market access, laid out in Protocol 2 of the ITAs, effectively divide the ECSC group into three subgroups: steel products, coal products (including some manganese and iron ores) imported by Germany and Spain and coal products imported by other EC countries. Setu This subgroup includes 554 eight digit CN items of which 522 are subject to GSP zero rates (Its equivalent is SITC. Rev.3. 2821+28221+67-67151-67682). The average tariff rate is in the mid-range. With NTB coverage ratios ranging between 57 and 75 percent, this group is the second most "Nfl-driven" among industial product groups idendfied in the ITAs (see Table 8); The share of this group in industrial imports into the EC varies across CEE-5 countries: in 1991 it was around 11 percent for Bulgaria and the FCSK, and 5-6 percent for the remaining countries. With the exception of the FCSK, the value of imports from other CEE-5 countries lfel precipitously in 1991, following a veiy substantial increase in 1990, but picked up again in 1992 despite the contraction in EC import demand for steel products. The ECSC protocol grants both nontariff and tariff concessions. As far as nontariff barriers are 3 For an exmensive discussion, see Global Economic Prospects and the DR_elRinA Countries (1992) and Cable (1990). 24 concerned, all quantitative restrictions are eliminated on the date of entry into force of the rrAs. Customs duties levied on steel products are to be eliminated after five years.37 Taking into account that almost all CN items failing in this group are subject to GSP rates, tariff concessions are not significant. Table 8: The "SteeP ECSC Sub-Group: Exports to the EC md Pre-ITAs Market Access Exports Share in Industrial Index 1991 NTB Coverage Simple Average Range of 1991 Exports, 1991 1988=100 Ratio, Tariff Rate tariff rates Max-Min (in US$ million) - n perce t Bulgaria 77 11.4 227 74.6 5.4 lO-0.0 FCSK 489 10.5 202 64A 5.6 10-0.0 Hungry 139 4.2 175 58.2 0.0 3.2-0.0 Poland 280 4.5 258 S7.4 0.1 4.0-0.0 Romania 61 3.6 96 68.2 0.0 0.040.0 Sources: See Table 2. This should not imply that all NTBs will disappear (to the contary, other instruments have gained prominence as quantiatve restrictions had been removed), or that all tariffs will indeed decline accordg to the schedule. Steel industries in both the EC and CEE-5 counties have significant surplu capacity, prouiding slrong pressures to export and to protect domestic makes against foreign imports.3 While the market share has remained low (3.6 percent of EC imports in 1992), penetrafion by CEE-5 steel imports increased dramatically between 1988 and 1992 with Poland and the FCSK each doubling their share in EC imports. This export success triggered calls for protectionist measures. In 1993, responding to complaints from domestic steel producers about the "flood" of cheap imports from Eastem Europe, the EC approached Hunpgran and Polish governments to impose voluntary resais and set quotas on imports of some steel products (steel coils, sheets, wire rod, strip, and cut lengths) originating in the Czech Republic and Slovak 37 At the begiing of the first year of ITAs the duties are reduced to 80 percent of the basic duties and fiher lowered to 60, 40, 20, 10 and 0 percent (of the basic duties) at the begining of the second, third, fourth, fifth and sixth years, respectively. 3s It is estimated that the CEE-5 sted industry works at around SO percent of its former capacity (Peel, 1993). 25 Republic above which puntive tariffs (from 25 to 30 percent) will be imposed." In addition, in November of 1992 temporary anti-dunping duties were imposed on seamless steel and iron tubes imported from the CEE-3 countries.0 These developments under the new ITA regime show that the Agreements have not deprived the SC of trade management instruments. The removal of quantitative restrictions have exposed CEE-5 steel producers to other, equally potent trade-restraining measues. Given the political clout of steel industries in the EC as well as the dramatic increase in imports from the CEE-3--fueled largely by the redirection of sales from domestic markets and the CMEA-the EC's recourse to these measures comes as no surprise. Cetk This group, consisting of 14 eight digit CN items, embraces coal products, iron ore and concentrates, iron and steel wastes (slags and scalings), and manganese ores and concenrates. Except for the FCSK and Poland, other CEE-5 countries are not significant net exporters of these products. There were no EC-wide NTBs affecting CEE imports in 1990, but tariffs were significantly larger than the average tariff on industrial products (see Table 9). Concessions granted by the EC vary among the CEE-5 countries and they have not been granted by all EC member countries. Except for Germany and Spain, EC duties on imports from the FCSK and Poland are to be eliminated within a year, whereas those on inports from Bulgaria, Hungary and Romania will be reduced by 50 percent on Jamnary 1, 1994 and abolished by the end of 1995. Germany and Spain will maintain duties until the end of the fourth year of the ITAs. As can be seen from Table 9, this differentiation in treatment affects mainly imports from the FCSK (the share of Germany is almost 100 percent) and to a lesser extent Poland (55 percent of EC inports goes to Germany and Spain). 39 Te quotas, set for the 1993-95 period, will change in terms of 1991 imports (1991 =100) according to the schedule: 1993=135; 1994=145; 1995=160. Thus, prohibitive tariffs will not apply annual increases in imports of less than 7 percent in 1993, of less than 8 percent in 1994, and of less thanl 10pernt in 1994. In the case of coils, 1991 levels may be exceeded by 100 percent in the 1993-95 period (Intemational Trade Reporter, May 1993, p.831). 40 The investigation was initiated in December 1991. The rates varied between 30.4 percent on these imports from the FCSK, 21.7 percent on imports from Hungary and 10.8 percent on imports from Poland. See Eleventh Annual Report from the Commission to the Euroeam Parliament on the Community's Anti-Dumping and Anti- Subsidy Activities (1992), Commission of the European Communities, Brussels, 28 October 1993. 26 Table 9: The "Coal" ECSC Sub-Group: Exports to the EC and Pre-rTAs Market Access Exports, 1991 Share in Industrial NTB Coverage Simple Average Range of Expons, 1991 Ratio, Tariff Ratc tariffs Max-Min (a)? (bj) (a) (b) (in USS million) (in percent) Bulgaria 1 0 0.1 0 0 4.2 8.3-0.0 FCSK 1 140 neg. 3.0 0 5.9 8.3-0.0 Hungazy 3 1 0.1 neg. 0 2.0 2.0-0.0 Poland 268 285 4.3 4.6 0 4.3 8.3-0.0 Romania 0 0 0 0 0 0.0 0-0.0 'J This subgroup includes SITC.Rev.3. 27862+281+321+(322-3223). 2/ (a) The EC excluding Gernmany and Spain. (b) Germany and Spain only. Sources: See Table 2. Exports of coal products are not likely to increase significantly in the near future. Increaszs in efficiency of energy use will reduce domestic consumption, however, coal production is not likely to expand and may even contract as many coal mims in the region face bankruptcy. In contrast to steel products, there was litle (if any) redirection of sales from tbeCMEA. The FCSK and Poland were established suppliers Of coal to EC markets well before the collapse of the CMEA. They both sought to minimize coal shipments to their former CMEA parmers, simply because coal was more marketable in hard currency markets than low quality man red goods. Thus, the contraction in FSU import demand affected raw materials less than tmanufacures and there was no explosion in exports to OECD markets as in many industrial products. E. The "Residual" Free Trade Gro. The residual group is subject to free trade upon entry into force of the ITAs. This group consists of products falling into CN 25-97 minus products identified in the five groups discussed above includig imports below ceiling or falling within tariff quota. It icludes 5,078 eight digit CN items of which 4,362 (68 percent of "fre-trade" itms) have been subject to GSP rates. As a resut, average triffs for this group are below ite-average for all industrial products. CEE-5 exports of this group were less affected by NTBs than any otha group of industial products specified in the iTAs-the average NTB coverage ratio of between 27 3.4 and 3.8 percent is significantly lower than for any other group except for coal products (see Table 10). Hence, the largest concessions made by the EC pertained to the "least sensitive" markets in terms of protection offered to domestic producers. The residual group accounts for a large sware of CEE-5 industrial exports to the EC. As can be seen from Table 10, in 1991 the country share of this group was between 32 percent (Romania) and 50 percent (Hungary). If exports below tariff ceilings and quotas which are duty free during the first year of the ITAs are included, then the free trade component-in terms of 1991 export baskets-goes up by 13 percentage points for Romania (to 47%), by 11 percentage points for the FCSK (to 55%) and Poland (to 54%), by 10 percentage points for Bulgaria (to 53%), and by 9 percentage points for Hungary (to 58%). Table 10: The "Immediate Free Tradell Group: Exports to the EC nd Pre-ITAs Market Acess Expons Share in Indusrial Index 1991 NTB Coverage Simple Average Range of 1991 Expors. 1991 1988=100 Ratio, Tariff Rate tariff.rates Max-Min (m USS million) (mpercnt) Bulgaria 293 43 132 3.6 5.6, .37.9-0 FCSK 2074 44 188 3.8 5.7 39.7-0 Hungay 1680 50 218 3.7 0.0 14.0-0 Poland 2707 41 190 3;8 - 0.1 - 14J0-0 Romania 602 32 57 3A 0.0 6.2-0 Sources: See Table 2. Imports of the 'residual" products from CEE-5 counties increased more slowly than othr idusti groups. Between 1988 and 1991 the share of this group in CEE-5 idusrial imports declined except for imports from Hungary it increased by almost 7 prcentage points from.42 to 49 percentr(seef Table 11).41 It is noteworthy that better export performance, as evidenced by the declning share of a free trade "residual" group in CEE-5 industrial exports, in "more protected" than in "less protected" markets also indicates that access to sensitive markets did not significandy restrain impors in the 1988-91 period-except for steel "' A possible explanation is that Hungy's strategy of economic developmentundercetral planning had been- less biased in favor of heavy industry, i.e., steel and chemicals, ta other socialist economies. Since steel and chemical products (both accounang for a very sizble share-of CEE-5 wiports into the EC) in the EC as a ruleenjoy a higher level of protection than other products, Hungary's export basket-with a lower share of steel and chemical product-had a higher share of products less vulnerable to EC protectonist measures. 28 Table 11: Composition of CEE-5 Industrial Exports to the EC, by ITA groups, 1988-92 Bulgaria 1988 2.5 0.2 1 0.9 14.9 9.6 0.1 0.0 61.7 lass 2.1 0,4 12.6 15.7 14.2 0.0 0.0 55.0 1s90 1.4 0.3 13.6 19.1 17.3 0.3 0.0 47.9 1991 1.4 0.1 16.3 20.9 11.4 0.1 0.0 49.6 1992 0.9 0.0 14.0 29.5 8.0 0.2 0.0 47.4 FC k ............................................................................................................... ......................................... . . _ FCSK 1989 0.6 0.0 24.4 13.2 12.6 0.1 2.8 46.4 1989 0.7 0.0 25.3 12.3 13.6 0.1 2.7 45.3 1990 0.6 0.1 24.2 13.6 14.2 0.0 3.0 44.4 1991 1.1 0.1 26.5 13.8 10.5 0.0 3.0 45.2 1992 1.3 0.0 25.9 13.7 10.4 0.0 2.3 48.4 .... 9.X............................................................................................................................................... ......_ ...... Hungary 1958 1.0 0.9 26.1 22.9 6.9 0.1 0.0 42.1 1989 1.5 1.0 26.7 21.5 6.6 0.2 0.0 43.6 1990 1.0 0.9 24.9 21.6 6.7 0.1 0.1 44.7 1991 0.7 0.5 24.3 21.4 4.2 0.1 0.0 48.9 1992 0.5 0.3 23.7 21.8 3.6 0.0 0.0 50.1 Poland 1998 0.2 0.3 22.4 13.8 4.2 9.6 3.5 46.2 1989 0.3 0.2 22.9 13.7 6.0 8.7 3.2 45.0 1990 0.6 0.4 23.7 15.2 6.4 0.0 3.9 44.9 1991 0.4 0.8 23.6 17.6 4.5 4.3 4.6 44.2 1992 0.7 0.7 24.4 19.0 4.3 3.7 3.5 43.6 .......................................... ...........................,..,,,,,.,,.,, , ,,,,, ......................................, , .......................... .. .... Romanla 1988 0.1 6.1 26.7 19.9 4.2 0.0 0.0 45.1 1989 0.2 5.2 23.1 19.3 4.0 0.0 0.0 48.2 1990 0.1 2.7 28.0 26.4 4.4 0.0 0.0 38.5 1991 0.0 1.6 31.4 28.2 3.6 0.0 0.0 35.3 * 1992 0.0 0.4 30.4 39.0 6.7 0.0 0.0 23.5 Sources: Calculated from data in the United Nations COMTRADE data base and the Interim Trade Agreements. 29 products whose rate of growth was below the average rate of growth of EC industrial imports from all CEE- 5 economies. Imports of chemical products, faling mainly in the "quotalfive-year-delayed" group, expanded at higher rates than the avenge for ai industrial products and their share also increased. VI. TIME PROFILE OF ATTAINING FREE TRADE IN INDUSTRIAL PRODUCTS Assuming that the measures liberalizing trade in industrial products are implemented according to the schedule set in the ITAs, one may estimate the effect of the pace of transition on industrial products in terms of 1992 export baskets. Since the ITAs signed with the troika economies in December 1991 went into force on March 1, 1992, some portion of their imports were free of tariff and nontariff restrictions in 1992. The share of freely traded industial products was larger than the residual' group-which accounted for 46 percent of industal imports from the FCSK, 50 percent of industial import from Hungary, and 44 percent of indusial imports from Poland-because some products falling into the 'quota/five-year-delayed' group obtained unrestrined access to EC markets5 Including below quota imports would raise the share of freely trade products quite significantly for three countries: to 54 percent br the FCSK and Poland, and to 57 percent for Hungary. Table 12 presents estimates of the sbares of freely-taded in industial products in terms of 1992 export baskets. These shares were tabulated on the basis of the foilowing assumptions: (i) all relevant provisions of the ITAs will be applied as scheduled; (ii) below-qota/ceiling imports into the EC equal 50 percent of imports of the "quota/five-year-delayed" free trade group as in 1992; and (iii) textiles and clothing trade will be conducted on the basis of bilateral quotas. The time path of ataining duty-free trade in industial products and the share of the free trade component is determined by the shares of exports of different groups, the share of industrial exports in total exports of that country to the EC, and differences in liberalizing provisions concernig the rate of growth in tariff quotas/ceilings on exports. The time path towards free trade has two jumps: the first occurs during the first year of the Agreements when around 50 percent of products obtain free access to EC markets; the 42 For the reasons discussed earlier, free trade shares calculated for the sample presented in Table 6 are used to esfimate the fee trade portion of this group. For the Balkan countries-which signed the ITAs a year later and for which no data are available to match their imports with quotas or ceilings-we used the avenage of the troika coefficients. 30 second in the sixth year of the ITAs when the remaining tariffs are removed (this increases this share to 100 percent if duties on MFA products are removed). The duty-free trade component is the largest in Czechoslovak and Polish exports to the EC. Although industrial products accounted for 93 percent of Romanian imports (see Table 1), the free access for its imports is smaller due to the large share of MFA products which increased from 28.2 percent in 1991 (see Table 7) to 39 percent in 1992. In terms of their share in industrial exports, the VisegrAd countries benefit more from fully liberalized access to EC markets than Romania and to a lesser extent Bulgaria because of the lower proportion of MFN products in their exports. Table 12: Shee of Duty-Free Goods in CEE-5 Industial and Total Expors to the EC, 1992-98 1992 1993 1994 1995 1096 1997 1998 . .- ;- - (shareci indultrial pworts, in percent) Bulgaria 54 54' 57 58 59 60 61 PCSK 59 63 65 67 71 86 86 Hungary 62 64 66 67 '68 -78 78 Poland 56 63 65 66 72 81 81 Romania ' 39 39' 42 44 46 48 61 CEE-5 58 69 70 70 71 74 80 (share in tora morts, in percent) Bulgaria 43 43 45 45 46 47 56 FCSK 55 59 61 62 66 80 80 Hungary 48 50 51 52 53 60 60 Poland 47 53 55 56 61 69 69 Romania 36 36 39 41 43 45 57 CEE-5 44 59 60 60 61 63 69 Note: In computing the duty-free shares, the commodity composition of flows in 1992 is applied to the anticipated customs stais of each item in all subsequent years. Source: Sce Table 6. Another factor responsible for the different traniton paths is related to the different provisions for quota/ceiting increases. Until 1995 the share of liberalized products is the largest for Hungary's exports to the EC. In 1993-94 the difference between the FCSK and Poland on the one band, and Hungary on the other, decres etnks to the fister increase in duty-free quotas/ceilings for the former countries. In 31 addition to the different pace in quota increases, the FCSK and Poland move ahead because of the elimination of duties on their coal imports into Germany and Spain in 1996; A. Copenhagen Summit Concessions The concessions granted by the EC summit in Copenhagen in June 1993 are significant for two reasons. First, the EC explicitly recognized the aspiration of the CEE-5 countries to acquire membership status, although no specific timetable was agreed upon. Second, it granted the CEE-5 further trade concessions for industrial products. The significance of the latter is that they accelerate the transition to free access to EC markets by a year.' Specifically, these new concessions provide for abolishing tariffs: (i) on steel products after four rather than five years; (ii) on the "four-year-delayed' free trade group after two instead of four years; and (iii) on the MFA group in frve years rather than six years.' Other concessions provide for an increase in the growth rate of tariff-free quotas/ceilings by '10 percentage points over the ITAs' rates of 15 percent (Hungary) and 20 percent (others) per annum and accelerates the elimination of triffs by two years on the "five-year-delayed" free trade group. How meaningful are these concessions? Table 13 highlights the differences between the ITA schedule and the schedule as mmofied by the Copenhagen Summit (again using te 1992 actual commodity composition of flows as weights). Within the next two years (1994-95), the net gain for former troika countries, as measured by the increase in share of products obtaining duty-free access to EC markets, amounts to 2-3 percentage points. The gains for Bulgaria and Romania are negligible except for 1997. However, one should also take into account an accelerated increase in margins of preference, esecially for- the "quota/five-year-delayed" group: although the new provisions do not have a large impact in terms of increasing the liberalized share of exports, they increase the competitive position of the countries, particularly of the Visegrid group. The accelerated schedule of reductions in triffs on textles and clotig may have a similar impact, assuming that CEE-5 quotas are not binting. The major benefits of the Copenhagen concessions are realized in the fifth year of the ITAs when the duty-free c increased - In addition, the EC summit offered-concessions on imports of farm products (levies and dities are to be reduced by 60 percent in two and a half years). 4In addition, the EC has pledged to improve rules concerning "outward processing." 32 by 6-13 percentage points. (If one includes MFA products, this share increases substantially for all CEE-5 countries.43) Table 13: The Significance of the Copenhgen Concessions: Net Change In "Liberalized" hnports, Relative to the ITA Schedule. 1993 1994 1995 1996 1997 1998 n milaion of US dollars) Bulgaria n.a. 6 10 11 83 0 FCSK 87 131 148 841 0 0 Hungary 48 87 103 245 0 0 Poland 95 200 218 499 0 0 Romania n.a. 6 0 114 0 0 CEE-5 229 431 479 1710 83 0 (share in indusiaI imports, in percent) Bulgaria n.a. 1 1 1 9 0 FCSK 1 2 2 13 0 0 Hlmgary 1 2 3 6 0 0 Poland 1 3 3 6 0 0 Romania n.a. 0 0 0 7 0 CEE-5 1 2 2 8 1 0 Note: The canges in customs status in each fiutre year are weighted according to the commodity composition of flows in 1992. Source: See Table 6. B. Reliabit of Nwtv-Free Share Esimates: How "Free" is a Free Trade Regime? The esimates of the free trade share in CEE-5 exports to the EC are based on assumptions tbat: (i) once tariff and nontariff measures are removed ihese products will contine to enjoy free access to EC I markets; and (ii) the export baskets will remain essentily unchanged over the next four to five years. bhe threat of managed trade-to borrow an apt phrase from Ostry (1993)-is writ large in the ITAs whose provisions contain a rich array of loosely defined safeguard clauses and, thus, '...open to virtually 4 The reason for not taking account of the concessions on texiles and cloting products (duty-free access to EC markets after five years instead of six years) is that MFA trade remains managed. Its futun is tied to the Uruguay Round. Including MFA products mises the net gain as follows: for Bulgaria (29%), FCSK (14%), Hungary (22%), Poland (19%), and Romania (39%). 33 unconained administrative discretion" (Ostry, 1993:14). It is impossible, however, to predict how these administrative powers will be used. On the one hand, the period following the implementation of the ITAs in March 1992 witnessed numerous attempts at reversing the liberalization but they were mostly limited to imports of steel products." For instance, the EC levied provisional anti-dumping duties on steel tubes from the FCSK and Poland in August 1992 and imposed quantitative restrictions on some types of steel originating in Czech and Slovak Republics in April 1993. On the positive note, expanding ties with the CEE-5 are also likely to mobilize lobbies in favor of further trade liberalization. Indicative of this development is Thomson's successful intervention to increase the EC quota on imports of cathode-ray tubes from Poland: Thomson has invested in a factory producing these goods in Poland (The Economist, May 1, 1993:55). Moreover, EC exports of sensitive products to some CEE-5 countries have increased significanfly, making them vulnerable to retaliatory measures-- measures aleady considered by some CEE-5 governments.47 Thus, although the ITAs leave a lot of room for a retreat to protectionism, its occurrence on a significant scale does nOt seem to be a likely developmert. CEE-5 export baskets will undoubtedly change in response to opportunities offered by the ITAs, but not significantly within the next dtree or four years. Change in export baskets calls for modifications in technologies and distrbution networks. This in turn calls for capital outlays.4' InvestMent activity was down throughout the region in the 1989-92 period. Although there was quite a significant shift in the composition of imports (see Table 11), it did not have a substantial impact on estimates of the free trade component with the exception of imports from the Balkan countries. Comparison of estimates of liberalized trade under the ITAs, in terms of 1991 and 1992 export baskets to the EC, supports this view. The 46 The other product, affected by an anti-dumping investigation, was ferro-silicon imported from Poland. Provisional and-dumping duties, imposed in May 1991, were converted into definitve anti-dumping duties on December 18, 1992 (see Eleventh Annual Report from the Comnission to the European Parliament on the Community's Anti-Dumping and Anti-Subsidy Activities, p.61). 4" Officials have often stated publicly that t... thqe reserve the right to retaliate in the case of any decision hurting (...) their exports." International Trade Reporter, August 18, 1993:1379. o At present, capital is scarce: one may expect that foreign direct investment will increase once uncertainty associated with market access in the EC is reduced. 34 difference is negligible for the FCSK, Hungary and Poland. The estimates for Bulgaria and Romania seem to be much less reliable, as the composition of their exports to the EC has been undergoing a significant change. In 1992 it changed rather significantly towards sensitive products and, as a result, estimates in terms of 1991 export baskets yield a much higher proportion of products with free accessO VII. CONCLUSION The ITAs substantially improve access to EC mar;ets for CEE-5 exporters by immediately eliminating tariffs on some industrial products and gradually reducing tariffs on others, although its short- term impact is significantly lower than that of granig GSP status. The largest increase in the share of liberalized imports occurs upon entry into force of the Agreements. In 1992, the first year of the ITAs for the Visegr&d countries, around 60 percent of their industrial exports obtained duty-free access to EC markets. The equivalent shares for Bulgarian and Romanian exports in 1993 were 54 and 39 percent, respectively. In the subsequent five years the share of duty-free exports registers steady growth (in terms of 1992 exports) and jumps sharply in the sixth year, i.e., once all trade liberalizing measures are in effect. These estimates tend to underrate the extent of improvement in market access for industrial products ofiginating in CEE-5 countries. In estimatig the liberalized component, no account is explicidy taken of tariff reductions which increase margins of preference for CEE-5 products. These are quite significant in spite of the EC's insistence on only gradually improving access to markets for sensitive products. Note that anmual tariff reductions granted in the fTAs range between 14.3 percent (MFA products) and 50 percent (the "one-year delayed' group) of the basic MEN rate. Furthermore, the decisions taken at the Copenhagen summit cut by one fifth the time it will take to reach the top of the EC preferential trade pyramid, now occupied by EFTA couraies. These reductions translate into a competitive edge over other sulppiers. Tbese concessions will assure the CEE-5 of a significant advantage over potential competitors from other former CMEA countries with comparative advantage in many similar products because of similarities of investment 49 Depending on the year, it varies between (-)1 and (+)1 percentage point for the Visegr*d countnies. economies. T The difference between estimates in terms of 1992 and 1991 export baskets is enormous for Romania, ringing between (-)22 percentage points in 1992 and (-)17 percentage points in 1998. For Bulgaria the estmates in terms of a 1992 export basket are by 4 perutage points lower in 1993-97 and 8 percentage points lower in 1998. 35 patterns under central planning. It will also give them an advantage over exporters from Mediterranean countries which benefit from preferential arrangements with the EC. While it is impossible to predict the extent to which bflaterally negotiated quotas on Imports of textile and clothing product will be binding for CEE-5 producers, they were not limiting in the 1990D92 period. Annual reductions in tariffs on products covered by the MFA by one seventh of initial rates will increase the attractiveness of imports from CEE-5 countries. This situation, combined with recently expanding EC exports of textiles and clothing to some CEE-5 countries (especially Poland), may assure that quotas will not become binding. If one assumes that they are not binding, then the liberalized share of industrial products in the seventh year of the ITAs increases significandy. The threat of the EC's retreat into protectionism is writ large in the Agreements. There are no provisions that would prevent an increase in managed trade through informal agreements and and-dumpig threats. This danger is particularly present in "traditional" industrial sectors (mainly steel). It should be reduced to some extent once institutional provisions conceming the rules of competition and subsidies in CEE-5 countries are in force. In all, this research does not give support to often-expressed views, both in Central/Eastern Europe and in the West, that trade liberalizing concessions offered in the ITAs do not benefit CEE-5 economies.51 It does not support the opinion tbat the ITAs '..have turned out to be disappointingly limited" Mg Economist, May 1-7, 1993:56). No doubt an immediate abolition of all tariff and nontariff barriers on imports from CEE-5 economies would yield higher benefits than those offering duty-free access to around 50 percent of their exports. But given the political underpinnings of the EC trading regime, more relevant standards of reference are their market access before the collapse of central planning and in comparison to other countries. Measured against these frames of reference, the ITAs provide for a rapid extension of liberalized market access for industrial imports from this region. 51 See for instce Marsh and Barber (1993) and Pomfret (1993). 36 REFERENCES Cable, V.(1990), "Adjusting to Textile and Clothing Quotas: A Summary of Some Commonwealth Countries' Experiences as a Pointer to the Future," in C.B. Hamnilton, ed., Textiles Trade and the DeveloRing Countries. Eliminatinig the Multi-Fibre Arrangement in the 1990s, The World Bank, Washington D.C. Dunin-Wasowicz, M. (1993), "Hamowanie w duecie," Gazeta Barncowa, Warsaw, July 3-10. Eleventh Annual Report from the Commission to the European Parliament on the Community's Anti- Dumping and Anti-Subsidy Activities (1992), Commission of the European Communities, Brussels, 28 October 1993. Erzan, R. and C. Holmes (1992), "The Restrictiveness of the Multi-Fibre Arrangement on East European Trade," Policy Research Working Paper 860, February, The World Bank, Washington D.C. Erzan, R., C. Holmes and R. Safadi (1992), "How Changes in the Former CMEA Area May Affect International Trade in Manufactures," Policy Research Working Paper 973, September, The World Bank, Washington D.C. Global Economic Prects and the Developirg Countries (1992), World Bank, Washington D.C. Hamilton, C.B. and W. Martin (1990), "Introduction," in C.B. Hamilton, ed., Textiles Trade and the DevelgRiln Countries. Eliminating the Multi-Fibre Arrangement in tne 1990s, The World Bank, Washington D.C. Inotai, A.(1993), Transformation in Progress: A Preliminay Assessment of the External Economic Factor, Forschungsebrichte, No. 200, The Vienna Institute for Comparative Economic Studies, Vienna, July. International Trade Reporter, various issues, The Bureau of National Affairs, Inc., Washington D.C. Kaminski, B. and A. Yeats (1993), "OECD Trade Barriers Faced by the Successor States of the Soviet Union," Policy Research WorkinZ Papers. WPS 1176, The World Bank, Washington, D.C. Marsh, D. and L. Barber (1993), "Morsels from a Groaning Table," Financial Times, June 7 Messerlin, P. (1992), "The Trade Relations between the EC and the Central and East European Countries, " minmeo, The World Bank and Institut d'Etudes Politiques de Paris. Messerlin, P. (1993), -The EC and Central Europe: The missed rendez-vous of 1992?," Economics of Transition, vol. 1, no. 1. Olechowski, A. and A. Yeats (1982a). "]Implications of the Tokyo Round for East West Trade Relations,* Oxford Bulletin of Economics and Statistics, February. 37 Olechowski, A. and A. Yeats (1982b). "The Influence of Nontariff Barriers on Socialist Countries' Exports," Economia Internazionale, Fall. Olechowski, A. and A. Yeats (1982b). "The Influence of Nontariff Barriers on Exports from Socialist Countries of Eastern Europe," Discussion Paer No. 6., UNCTAD, Geneva. Ostry, S. (1993), The Threat of Managed Trade to Transforming Economies, Group of Thirty, Washington D.C. Peel, Q. (1993), "Aid Urged for East European Steel Cuts," Financial Times, 18 June Pohl, G. and P. Sorsa (1992), "European Integration and Trade with the Developing World, " Policy and Research Series, No. 21., The World Bank, Washington D.C. Pomfret, J. (1993), "Is Another Curtain Descending on Europe?," The Washington Post, April 27. Rollo, J. and A. Smith (1993), "EC Trade with Eastern Europe," Economic Policy, April. Schurmcher, D. and U. Moebius (1992), "Analysis of Current Community Trade Barriers to Central and East European Countries," Deutches Institut fur Wirtschaftsforschung, Berlin. Trela, I. and J. Whalley (1990), "Unaveling the Threads of the MFA," in C.B. Hamilton, ed., Textiles Trade and the Developin2 Countries. Eliminating the Multi-Fibre Arrangement in the 1990s, The World Bank, Washington, D.C. Tovias, A. and S. Laird (1991), "Whither Hungary and the European Communities?," Policy. Research. and External Affairs. Working Papers. International Trade, WPS 584, The World Bank, Washington, D.C. UNCTAD/World Bank (1989), A User's Manual for SMART, The World Bank, Washington D.C. Wang, Z.K. and L.A. Winters (1993), "EC Imports from Eastern Europe: Iron and Steel," CEPR Discussion Paper no.825, Centre for Economic Policy Research, London. Yeats, A. (1979), Trade Barriers Facing Developing Countries, Macmillan, London. Policy Research Working Paper Serles Contact Title Author Dale for paper WPS1292 Services as a Malor Source of Growth William Easlerly Apil 1994 C. Roilison in Russia and Other Former Soviet Martha de Melo 84768 States Gur Ofer WPS1293 Product Standards, Imperfect Glenn Hanison April 1994 N. Artis Competition, and Completion of the Thomas Rutherford 38010 Market In the European Union David Tarr WPS1294 Regulations, Institutions, and Hadi Salehi Esfahani Apnl 1994 B. Moore Economic Performanrce: The Political 35261 Economy of the Philippines' Telecommunications Sector WPS1295 Why Higher Fiscal Spendig Persists Bruno Boccara Apnl 1994 M. Pfefffenberger When a Boom In Primary Commodites Ends 34963 WPS1296 Eamings-Related Mandatory Salvador Valdds-Prieto April1994 H. Rizkaiia Pensions: Concepts for Design 84766 WPS1297 How Relative Prices Affect Fuel Chaues C. Guo May 1994 C. Jones Use Pattems in Manufacturing: James R. Tybout 37699 Plant-Level Evidence from Chile WPS1298 Capital Goods Imports, the Real Luis Serven May 1994 E. Khine Exchange Rate, and the Current Account 37471 WPS1299 Fiscal Policy in Classical and Klaus Schmidt-Hebbel May 1994 E. Khine Keynesian Open Economies Luis Serven 37471 WPS1300 Dynamic Response to External Klaus Schmidt-Hebbel May 1994 E. Khine Shocks in Classical and Keynesian - Luis Serven 37471 Economies WPS1301 Estimating the Health Effects of Bart Ostro May 1994 C. Jones Air Pollutants: A Method with an 37699 Application to Jakarta WPS1302 Sustainabiliiyt Ethical Foundations Geir B. Asheim May 1994 C. Jones and Economic Properties 37699 WPS1303 Conflict and Cooperation in Managing Scott Barrett May 1994 C. Jones Intemational Water Resources 37699 WPS1304 Infornal Gold Mining and Mercury Dan Biller May 1994 D. Biller Poilution in Brazil 37568 WPS1305 Information and Price Deternination Nemat Shafik June 1994 A. Yideru Under Mass Privatizaton 36067 Policy Research Working Paper Series Contact Title Author Date for paper WPS1306 Capital Flows and Long-Term Ibrahim A. Elbadawl June 1994 R. Martin Equilibrium Real Exchange Rates Raimundo Soto 39065 In Chile WPS1307 How Taxaton Affects Foreign Direct Joosung Jun June 1994 S. King-Watson Investment (Country Specific Evidence) 31047 WPS1308 Ownership and Corporate Control in Brian Pinto June 1994 M. Kam-Cheong Poland: Why State Firms Defled the Sweder van Wijnbergen 39618 Odds WPS1309 Is Demand for Polluting Goods Gunnar S. EsKeland June 1994 C. Jones Manageable? An Econometric Study Tarhan N. Feyzioglu 37699 of Car Ownership and Use in Mexico WPS1310 China's Economic Reforms: Pointers Justin Vifu Lin June 1994 C. Spooner for Other Economies in Transition Fang Cai 30464 Zhou U WPS1311 The Supply Response lo Exchange Mustapha fouis June 1994 J. Schwartz Rate Reforn in Sub-Saharan Africa Weshah Razzak 32250 (Empirical Evidence) Cajios Mollinedo WPS1312 The New Wave of Private Capital Eduardo Fernandez-Arias June 1994 R. Tutt Inflows: Push or Pull? 31047 WPS1313 New Estimates of Total Factor Vikram Nehru June 1994 M. Coleridge- Productivity Growth for Developing Ashok Dhareshwar Taylor and Industrial Countries 33704 WPS1314 The Significance of the 'Europe Bartbomiej Kaminski June 1994 M. Patefia Agreements for Central European 37947 Industrial Exports