WPS3864 Bulgaria's Institutions and Policies: Integrating into Pan-European Markets·/ Bartlomiej Kaminski University of Maryland, College Park and World Bank Abstract: This paper analyzes the process of institutional transformation in Bulgaria and assesses the extent to which it has established institutions and policies fostering domestic economic activity and integration into global markets. After a brief review of characteristics and achieved progress in first- generation reforms, i.e., removal of central control over prices, liberalization of foreign trade and exchange rate regimes, the paper first assesses in the comparative perspective the progress made in the quality of governance and structural reforms. It then takes a look at the extent to which this has impacted foreign direct investment (FDI) inflows and was translated into improved business environment in its domestic and external dimensions. The external dimension relates to backbone services facilitating trade and Bulgaria's trade policies. As far as the latter are concerned, the discussion highlights tensions that emerge from duality--regional versus multilateral--in Bulgaria's trade policy. Despite significant progress in implementation of structural reforms and converging to the EU acquis communautaire that has led to a significant enhancement in the quality of governance and market supporting institutions, `macro' institutional improvements are yet to be fully transplanted to a micro-level, as three areas appear to remain a binding constraint: First and foremost is the low quality of the judicial system and, by the same token, weaknesses in the enforcement of property rights and contracts. Second, backbone services facilitating trade remain a barrier. Bulgaria ranks low relative to the levels of efficiency achieved on average by both EU-8 and the EU-15 countries in management of ports, IT infrastructure and customs. Third, there are recurrent complaints among businesses of government bureaucracy, poor infrastructure and frequent changes in the legal framework including taxation. As a result, the regulatory burden remains huge. There are still redundant and excessive sector specific regulatory regimes. Bulgaria's markets for industrial goods are fully contestable for pan-Europe (EU-25, EFTA, Romania, and Turkey), exposing local producers to duty-free competition from imports. With relatively high MFN tariff rates, the level of reverse discrimination significantly increased over the last couple of years. While this has not resulted in perceptible trade diversion, organizational arrangements preventing that to happen unnecessarily increase administrative intervention in the economy. Key words: integration, transition, governance, foreign trade, trade facilitation, foreign direct investment, Pan-European markets. World Bank Policy Research Working Paper 3864, March 2006 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. ·/Background report prepared for Bulgaria: The Road to Successful EU Integration--The Policy Agenda: Country Economic Memorandum, The World Bank. The author is grateful to Rosalinda Quintanilla, Stella Ilieva, Gallina Vincelette and Juan Carlos Ginarte for useful comments and suggestions. 1 1. Introduction Participation in the EU accession process, which began with signing with the EU of the European Association Agreement in 1993, has deeply impacted Bulgaria's institutions and policies designed to establish market economy and plug it into regional and global markets. Until the 1996-97 crisis, its impact was limited to conditions in market access. The policies pursued in 1990-96 have created conditions conducive to neither economic growth nor membership in the European Union. With the invitation extended by the EC to five Central European countries (Czech Republic, Estonia, Hungary, Poland, and Slovenia) in 1996 to begin accession negotiations, the prospect of EU membership for Bulgaria has become more real. The combination of a deep economic crisis and plausibility of EU membership has changed the political economy in Bulgaria in favor of reforms. The acquis communautaire has provided a template for structural reforms. The latter took off with certain delay, as the government had to deal with the legacy of depredation that took place before 1997. In recognition of the progress achieved in building market supporting institutions and policies, the European Commission declared Bulgaria a fully functioning market economy in 2003, and accession negotiations were successfully completed on June 15, 2004. This paper analyzes the process of institutional transformation in Bulgaria and assesses the extent to which it has established institutions and policies fostering domestic economic activity and integration into global markets. After a brief review of characteristics and achieved progress in first generation reforms, i.e., removal of central control over prices, liberalization of foreign trade and exchange rate regimes, the paper first assesses in the comparative perspective the progress made in the quality of governance and structural reforms. It then takes a quick look at the extent to which this has impacted foreign direct investment (FDI) inflows and was translated into improved business environment in its both domestic and external dimensions. The external dimension relates to both backbone services facilitating trade and Bulgaria's trade policies. As far as the latter are concerned, the discussion highlights tensions that emerge from duality--regional versus multilateral--in Bulgaria's trade policy. Several observations emerge from this discussion: · Bulgaria has mostly completed the process of implementing first-generation reforms. Prices are not only stable but also almost fully shaped by market forces. Its exchange and foreign trade regimes are liberal and do not obstruct trade and financial flows. · Progress in implementation of structural reforms and converging to the EU acquis communautaire has led to a significant enhancement in the quality of governance and market supporting institutions. This has led to the change in perception of Bulgaria by foreign investors, who have been increasingly responsive to friendlier investment climate. In 2004 FDI per capita in Bulgaria was higher than that for the CEEC-8 economies. · Bulgaria scores high on most institutional and governance dimensions. Thanks to recent structural reforms, it outperforms now other Balkan economies including Romania and Croatia and on some counts scores higher than some new EU member states from Central Europe · Economic benefits from structural and institutional reforms are yet to be fully tapped, as most of them were implemented over the last four years. Until around 2000 there was rather limited progress. · `Macro' institutional improvements are yet to be fully transplanted to a micro-level, as three areas appear to remain a binding constraint. First and foremost is the low quality of judicial system and by the same token weaknesses in enforcement of property rights and contracts. The second relates to backbone services facilitating trade. Bulgaria ranks low relative to the levels of efficiency achieved on average by both CEEC-8 and the EU-15 countries in management of ports, IT infrastructure and customs. In addition, there are recurrent 2 complaints among businesses of government bureaucracy, poor infrastructure and frequent changes in the legal framework including taxation. Despite progress in simplifying and streamlining the regulation of business activities achieved over 2002-04, the regulatory burden remains huge. There are still redundant and excessive sector specific regulatory regimes. Labor regulations allow for limited flexibility in labor markets and cost of business entry remains unnecessary high. Courts operate slowly in enforcement of contracts and mediating commercial disputes. · All of these constraints mentioned above are important. Taken together, they block the entry of new businesses and prevent domestic firms from participation in a rapidly developing new global division of labor based on international outsourcing and fragmentation of production. The increase in FDI inflows is not necessarily a testimony that Bulgarian business climate has been cleaned of weaknesses, as large multinational companies are well endowed with its potential weaknesses. Rather the ultimate test is the extent to which the existing business environment removes obstacles to the emergence of new firms and foreign "small scale" investments. On this account, regulatory regime or rather their multiplicity remain an important obstacle to be addressed by government policies. · Bulgaria's markets for industrial goods are fully contestable for pan-European (EU-25, EFTA, Romania, and Turkey) exposing local producers to duty-free competition from imports. With relatively high MFN tariff rates, the level of reverse discrimination significantly increased over the last couple of years. While this has not resulted in perceptible trade diversion, organizational arrangements preventing that to happen unnecessarily increase administrative intervention into the economy. In all, thanks to structural reforms and liberal regional trading environment, Bulgaria has successfully, albeit only belatedly, begun taking advantage of opportunities offered by participation in the EU-driven Eastern Enlargement regional integration project. The challenge ahead is to maintain and build upon momentum of regulatory and institutional reforms launched in the 2000s and reduce MFN tariff rates on industrial products to those levied by the EU. 2. Investment and Business Environment for Plugging-into Globalization The experience of countries that have successfully taken advantage of opportunities offered by global markets suggests that two elements have to be in place--successful implementation of first-generation reforms (liberalization of prices, foreign trade and exchange regimes) and consistent movement towards a rule-based institutional regime with the capacity of their enforcement. The former is relatively easy to implement, provided absence of political opposition, whereas the latter requires advanced institutional capacity of the state. Bulgaria moved swiftly in implementing first-generation reforms. Its 1991 stabilization- cum-transformation program was successful in restoring macrostability and laying the foundations for a demand- rather than supply-constrained economy, characteristic of central planning. Yet, many reform measures were reversed and others not implemented. Weaknesses in state capacity combined with strong political resistance have stood in the way of moving fast in structural reforms. As a result, inflation turned into hyperinflation in 1996 bringing the economy to the brink of collapse. With the change of government in 1997, Bulgaria launched its transition program to competitive markets for a second time, although under much more adverse circumstances than in 1991. Collapsed banking sector combined with a six-year of economic mismanagement and plundering of public resources dramatically raised the cost of necessary reforms. How quickly the returns from liberalization will occur always depends on initial conditions. Devastated banking sector and public finances heavily extended the payoff period in spite of a rapid progress in restoring macroeconomic stability and government commitment to structural reforms. While progress in first-generation reforms is relatively straightforward to assess (e.g., inflation, convertibility of domestic currency for current account transactions), this is not so with second generation 3 reforms covering as a rule activities and policy areas where progress can be gauged only indirectly. Moreover, they usually tend to be highly intertwined. Progress in one area can be `neutered' by the lack of progress in another area. For instance, large-scale privatization may be completed. Yet, its benefits may not materialize if privatized companies are subsidized by still state owned banks, as it was the case in the Czech Republic in the mid-1990's, or competition policy is unable to contain their abuse of dominant position. The same comment applies to various indicators measuring the quality of business climate. Huge minimum capital requirements or instabilities in tax policies, for instance, may offset simplicity of procedures for business entry. In order to assess Bulgaria's progress in building institutions supporting plugging-into global markets, we adopt an eclectic approach. Proxy variables for assessing the progress in structural reforms in a comparative perspective is the improvement in the quality of governance supplemented by the `cost of doing business' indicators and external assessments of corruption. We shall use the methodology and data developed by the World Bank for the first two, and Transparency International's country rankings for the latter. They are all intertwined with the progress in building institutions supporting competitive markets and improving business climate. Governance takes into account the capacity of state to provide environment conducive to business activity, i.e., transparency, accountability, enforcement of the rule of law, regulations supporting competition, etc. The `cost of doing business' indicators provide measures to assess business- friendliness of regulations, whereas corruption indicators the extent of graft as perceived by international investors. The incidence of corruption supplements an assessment of the quality of governance pointing to its symptoms. Two caveats are important. First, the shared weakness of governance and cost of doing business indicators is that a low score in a single dimension can offset high scores on other dimensions. For instance, friendly conditions for entry of start-ups would matter little if the government preys on existing businesses. In a similar vein, political instability will make superfluous positive scores on other dimensions of governance. Second, governance is not a synonym for corruption. Although better governance implies lower corruption, the two are different. Governance includes supply of public goods and services. Better governance and regulatory environment, low `hassle' cost of conducting business, labor force, etc., all increase the probability of receiving higher FDI inflows and provide strong incentives to the development of the locally-owned private sector. Countries that have made the largest progress in transition to competitive markets have also registered stronger growth and attracted the largest inflows of FDI, whereas countries with weaker business climate have been less successful in attracting FDI. This section examines progress achieved in the post-crisis stage of transition and sets it in a comparative perspective. 2.1. First-generation reforms: prices, foreign trade, foreign exchange and macrostability Bulgaria followed initially similar policies as other radical reformers in the region. It introduced in 1991, i.e., roughly at the same time as the former Czechoslovakia and Romania, the stabilization-cum- transformation program that "... essentially resembled the Polish program of 1990" (Bruno 1994, p. 24). All these countries freed most prices, removed the multiple exchange rate regime, introduced convertibility for current account transactions and dramatically liberalized foreign trade by eliminating all quantitative restrictions and the state monopoly over foreign trade. Some differences, however, emerged early in transition indicating that Bulgaria as well as Romania moved with an initially smaller bang than other Central European early reformers. For instance, not all central controls over prices of tradables were removed and the foreign exchange rate regime was not liberalized in line with IMF Article VIII obligations. In consequence, there were significant restrictions on current account convertibility of Leva. In addition, hard budget constraint on state-owned enterprises was not imposed by state budget policies. Yet, from the perspective of developments in 1995, Bulgaria seemed poised to perform like other radical reformers that appeared to be on the verge of a strong recovery from transformational recession. In 4 terms of the scores of the EBRD, measuring progress in transition, Bulgaria reached levels comparable to that of Poland in 1990, i.e., the year of implementation of stabilization-cum-transformation program (see Table 1 below). However, in 1996, dynamic similarities in macroeconomic performance between Bulgaria and other early reformers disappeared. While in Hungary or Poland inflation continued falling to low double digit-levels, in Bulgaria inflation turned into hyperinflation. The GDP fell in 1996; inflation and government debt spiraled out of control; financial markets and banking sector collapsed; and foreign exchange reserves almost disappeared. Bulgaria became then the only European transition economy "... that has first achieved respectable economic performance, but later seen that performance deteriorate into full-blown hyperinflation and strongly negative growth of domestic product" (Wyzan 1998:6). Faced with hyperinflation, Bulgaria had to implement a stabilization package and considerably expand on measures introduced back in 1991 and subsequently aborted. Bulgaria thus began its second transition to competitive markets, characterized by strong commitment to macroeconomic stability and implementation of structural reforms, under much more demanding conditions than in 1991. Some interesting comparative insights can be gleaned from the examination of EBRD transition indicators of progress in liberalization of prices, small-scale privatization, and foreign trade and exchange rate regimes normalized for the purpose of this analysis.1 Indeed, much lower scope of initial liberalization in Bulgaria following the implementation of the stabilization-cum-transformation program than in Poland is clearly visible. While Poland scored 57 percent in its first year of the program in 1990, the value of Bulgaria's index of aggregate progress in first generation reforms was only 38 percent. Bulgaria's reached the level of dismantling central price controls, state ownership of small firms and foreign trade and foreign exchange rate regime that Poland had already in 1990 only in 1994 and stayed at this level until 1996. Bulgaria reached in 1998 the level of progress in first stage liberalization reforms achieved by former Czechoslovakia in 1992, i.e., during the second year of the transformation program, Poland in 1993 and Estonia in 1994. Bulgaria's performance was in stark contrast to Estonia--both countries began their transition from the same level in terms of state micromanagement (Table 1). While Estonia moved rapidly in dismantling vestiges of central planning, Bulgaria was falling behind fast reformers. Table 1: Progress in first generation reforms as revealed in values of EBRD-based aggregate index in 1990- 2004 in selected transition economies (in percent) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bulgaria 9.5 38.1 38.1 44.5 57.1 57.1 57.1 66.7 76.2 82.5 85.7 88.9 88.9 88.9 88.9 Poland 57.1 57.1 66.7 76.2 76.2 76.2 82.5 82.5 95.1 95.1 95.1 95.1 95.1 95.1 95.1 Estonia 9.5 19.0 38.1 57.1 76.2 76.2 79.3 79.3 92.0 92.0 95.1 95.1 95.1 95.1 95.1 Romania 0.0 9.5 38.1 47.6 60.3 63.5 57.1 69.8 85.6 88.9 88.9 88.9 88.9 88.9 88.9 Memorandum: median value in SEE-4 47.6 47.6 36.5 38.1 33.3 47.6 47.6 52.4 66.7 66.7 68.2 73.0 74.6 76.2 76.2 CEEC-8 9.5 52.4 57.1 71.4 76.2 76.2 80.9 80.9 92.0 92.0 95.1 95.1 95.1 95.1 95.1 Notes: (1) the aggregate is the average of scores ranging between 1 (no liberalization) and 4.5 (liberalization at the levels of highly developed market economies) for prices, foreign trade and exchange rate regime and small privatization. It has been normalized with 1=0 and 4.5=100; (2) SEE-4 include Albania, Bosnia and Herzegovina, Macedonia and Serbia and Montenegro. Source: own calculations based on data from EBRD Annual Transition Reports. The second transition shares more similarities with the initial path of the radical Central European reformers than Bulgaria's first, failed transition. In some respects, it progressed faster, as aggregate output rebounded in the next year after the implementation of the stabilization program, and inflation was brought under control in the second year of the program. In some others, it progressed slower. While it would be tempting to take the value of the EBRD-based first-generation reform index that Poland reached in the first year of its stabilization-cum-transformation program in 1990 as a frame of reference, this would be 1These three policy areas constitute the core of first generation reforms. Scores for each area ranging between 1 for no reforms and 4.5 for reaching the average conditions in highly developed economies have been normalized with 1 equal 0 and 4.5 equal 100. The average of these three indicators amounts to the achieved progress in first generation reforms. 5 misleading for one major reason. The index, as constructed here, captures only the extent of liberalization of prices and foreign trade and exchange rate regimes and progress in small-scale privatization. Bulgaria reached in 1994 Poland's level only thanks to the improved EBRD rating in liberalization of foreign trade and exchange rate regime. It is not clear why the EBRD analysts raised this score, as the extent of convertibility on current account transactions in Bulgaria was lower. It seems that the reason might be the fact that Bulgaria's MFN applied tariff rates were lower. Hence, a better comparator is the level of liberalization achieved by Bulgaria in 1997, i.e., the year of implementation of the second stabilization program. The value of first generation reform index reached this level in 1992 for Poland and in 1997 for Romania. Estonia, which is included here because the values of the index were the same as for Bulgaria in 1990 and in 1992, improved its score from 57.1 percent in 1993 to 76.2 percent in 1994. Since both scores are in the same distance away from Bulgaria's score in 1997, we set 1993 as a comparator data. Two observations can be derived from data presented graphically in Figure 1. First, the progress in first generation reforms was larger in Bulgaria during the first five years, i.e., in 1997- 2001 than that achieved in Estonia and Poland during their respective five-year periods but much smaller than in Romania. The values of the index for both countries were almost identical in 1997. But Romania outperformed Bulgaria in 1998-2000 due to faster progress across all policy areas. Both countries reached parity in 2001. Figure 1: First-generation reform index for Bulgaria and Romania (1997-04), Poland (1992-97) and Estonia (1993-98) 100 Estonia, 1=1993 90 Bulgaria, 80 1=1997 70 Poland, 1=1992 60 Romania, 50 1=1997 1 2 3 4 5 6 7 8 Source: as in Table 1. Second, first-generation reforms appear to have been shelved during the remaining three years, that is, in 2002-2004. While there was little room for improvement in liberalization of prices or foreign trade and exchange arrangements, there was plenty of room to go ahead with small-scale privatization identified by the EBRD as an unfinished business. Neither Bulgaria nor Romania, however, has made any progress. However, because of at least five-year delay in establishing relatively competitive markets, accompanied by gross economic mismanagement, private appropriations of state subsidies, huge contingent liabilities, which led to the collapse of the banking sector, and massive asset stripping, the overall cost of stabilization operation was enormous. The challenge of restoring stability and rebuilding devastated banking sector and international reserves pushed off some first-generation reform measures the government policy agenda. In consequence, some measures that might make Bulgaria more attractive for foreign investors as well as would facilitate trade were introduced with considerable delay. For instance, its outdated and fragmented foreign currency legislation, with significant restrictions on current account transaction, was overhauled and brought in line with IMF Article VIII obligations only in 1999. Similarly, following the liberalization of many prices in 1997-98 there was no further dismantling of price controls until 2001. 6 2.2. Second generation reforms: governance and FDI While macroeconomic performance has been impressive since the introduction of the stabilization- cum-transformation program in 1997, the progress in implementing the second-stage institutional reforms has been less spectacular, albeit not disappointing. With some delays privatization took off and a number of measures improving business climate have been implemented.2 Since 1998, Bulgaria has reformed the regulatory and institutional framework of the banking sector and privatized private banks to strategic investors. As a result, the financial sector has fully recovered from huge financial losses and damage to its credibility suffered in the aftermath of the financial meltdown in 1996-97. This section examines Bulgaria's position in terms of the progress in structural reforms as captured by corruption indices, World Bank's indicators of governance, and EBRD indicators of economic reforms. It assesses progress both over time and vis-à-vis its regional peers. Since the test of pudding is eating, we also look at links between achieved progress and FDI inflows. A. Perception of corruption, governance and structural reforms How does Bulgaria compare to other countries in the region in terms of progress achieved in structural reforms? In order to answer this question, we use three different indicators: Corruption Perception Index (CPI), World Bank-based aggregate index of governance quality, and EBRD-based index of progress in second generation reforms. Admittedly, each of them is very subjective based on experts' opinions rather than `hard' numbers. Yet, taken together they provide basis for assessment of changes over time and Bulgaria's position vis-à-vis its regional peers. Taken together, they point to very significant progress in improving business climate achieved especially over the last three years. We begin with CPI providing information on perception of international investors of corruption in a country in comparative context. International investors' perception of the Bulgarian investment climate has substantially improved since 1998. The verdicts given by participants of Transparency International annual surveys have been increasingly favorable for Bulgaria. While in the mid 1990s Bulgaria together with Romania was ranked somewhere between CEEC-8 and CIS economies, the latter notorious for high incidence of corruption, Bulgaria has been moving up. It has not only the `cleanest' institutional environment among Balkan countries but, by 2000, Bulgaria has also become like other CEEC-8 countries in terms of the value of CPI (Table 2). The decline in perception of corruption has been impressive since 2000, as Bulgaria has been moving towards the median CPI value of 4.4 in 2000-04 for CEEC-8 countries. Bulgaria's current rating stands at 87 percents of the CEEC-8 average and 93 percent the CEEC-8 median value of CPI. Table 2: Values of Corruption Perception Indices for selected countries in 1998-2000 and 2002-2004 1998 1999 2000 2002 2003 2004 Index, 2004 2000=100 Bulgaria 3.3 3.5 3.5 4.0 3.9 4.1 117 Romania 3.3 2.9 2.6 2.8 2.8 2.8 108 Average CEEC-8 4.2 4.6 4.5 4.6 4.6 4.6 104 Best 1999-2004: Slovenia a/ n/a 6.0 5.5 6.0 6.0 6.0 109 Worse 1998:Latvia 2.7 3.4 3.4 3.7 3.7 4.0 118 Worse 2004:Poland 4.6 4.2 4.1 4.0 4.0 3.5 85 Average EU-15 7.7 7.6 7.6 7.6 7.8 7.8 103 Average SEE-5 n/a 2.7 N/a 2.8 2.8 2.8 103 b/ Memorandum: Bulgaria in % of the CEEC-8 average 78.6 76.5 78.2 87.9 85.7 88.4 116 Notes: / no data available for 1998. / base year is 1999. a b Source: derived from data available at the web site of Transparency International at http://www.transparency.org/ 2For instance, the Privatization and Post-privatization Act repealed the 1992 Law on the Transformation and Privatization of State and Municipal-owned Enterprises only in March 13, 2002. 7 The value of CPI still remains below the average for EU-15 and the gap has slightly increased. The fact the average for CEEC-8 countries has also fallen behind does not change an overall assessment of the magnitude of the challenge faced by Bulgaria in improving regulatory environment to minimize the potential for rent seeking. The same conclusions boiling down to "huge improvement but a challenge ahead" can be drawn from the analysis based on diagnostic tools developed by the World Bank. Although better governance implies lower corruption, the two are different as the former also includes the supply of public goods and services and their quality. Tools designed to assess the quality of governance are much more elaborate than Transparency International's rankings covering separately six dimensions of governance and drawing on a much larger pool of expert assessments. Table 3 reports values of three rather than all six governance indicators. Dimensions of governance included are political stability, government effectiveness and regulatory quality.3 A single aggregate index of governance is a simple average of the values of these three governance indicators. Their values range between ­2.5 (the worst case) and +2.5 (the best case). In 1996-98 Bulgaria together with Romania were scored well below CEEC-8 economies and below the average for SEE-4 countries on political stability and government effectiveness dimensions. The regulatory quality was one of the highest among SEE economies in 1998 but offset by low government effectiveness. In terms of aggregate index, Bulgaria was on a par with Romania in 1998, well above SEE-4 economies rather dramatically below Poland the overall worst performer among CEEC-8 countries. Table 3: Selected indicators of the quality of governance in 1998-2002 Political Stability Government Effectiveness Regulatory Quality Aggregate (average) 1998 2000 2002 1998 2000 2002 1998 2000 2002 1998 2000 2002 Bulgaria 0.39 0.30 0.56 -0.97 -0.13 -0.06 0.47 0.21 0.62 -0.04 0.13 0.37 Croatia 0.61 0.37 0.56 0.29 0.16 0.19 0.34 0.30 0.19 0.41 0.28 0.31 Romania 0.20 0.01 0.42 -0.63 -0.58 -0.33 0.30 -0.27 0.04 -0.04 -0.28 0.04 CEEC-8: best and worse performers Estonia 0.84 0.79 0.98 0.42 1.02 0.78 1.06 1.30 1.35 0.77 1.04 1.04 Poland 0.85 0.83 0.71 0.86 0.39 0.61 0.83 0.60 0.67 0.84 0.61 0.67 SEE-4 Average a/ -0.52 -0.47 -0.46 -0.67 -0.77 -0.62 -0.87 -0.51 -0.43 -0.72 -0.63 -0.57 a/ Note: SEE-4 include Albania, Bosnia and Herzegovina, Moldova and Serbia and Montenegro. Source: derived from data available at www.worldbank.org/wbi/governance/gov_data.htm and discussed in Kaufman, Kraay and Mastruzzi (2003). Since 1998 the improvement in the quality of governance in Bulgaria has been very significant in all dimensions. The value of aggregate index of 0.37 in 2002 was much higher than of any other SEE country including Croatia and Romania. It was, however, still well below the quality of governance in CEEC-8 countries in 2002. It is not clear whether the `governance' gap vis-à-vis CEEC-8 has been falling over the last two years, as there a re no data available for 2003 and 2004. Yet, considering the fall in the perceived level of corruption as reported by Transparency International, one would expect at least continuation of trends observed in 1998-2002. An examination of recent progress in structural reforms suggests that developments over 2002-04 could be characterized as acceleration rather than a mere continuation of trends in 1998-02. This is a general conclusion that can be derived from the EBRD transition indicators pertaining to structural, second- 3There are two arguments to focus solely on these indicators: First, these indicators are critical dimensions of business climate. Political stability impacts investment decisions and in extreme situations of its absence may disrupt economic activities. The quality of regulation is of little relevance unless supported by the government capacity to enforce regulations. Second, three other indicators pertinent to such dimensions of governance as the rule of law, control of corruption, and voice and accountability are not taken into account do not yield extra information. For all transition economies, they are strongly correlated with the selected three indicators, with the values of correlation coefficients equal or above 0.9. 8 generation reforms as well as from a more detailed examination of recent reform measures. For the review of the latter, see Box 1. Box 1: Snapshot of structural reform measures implemented in 2000-04 Recent progress in structural reforms is discussed under the following headings: large-scale privatization; reforms in energy sector; reforms of telecommunications; railways' reforms; water sector reforms; and reforms of the financial sector. In all of these areas, significant progress has been made. Large scale privatization has been almost completed, with the privatization of the BTC, the Bulgarian Telecom (2004), seven electricity distribution companies (2004-05), seven district heating companies (2004), two (Russe and Varna) ship-building companies (2003) and more than 1,100 minority packages sold in 2004. Restructuring of the energy sector has been underway with implementation of substantial the institutional, pricing and regulatory reforms. These included: opening of the sector has been open to private entry and the elimination of former monopolies, bringing the prices over 2002-04 to cost recovery levels; and elimination of cross-subsidies. Simultaneously, enactment of the new Energy Act (2003), that is consistent with EU energy directives, has established a new legal and regulatory framework for the energy sector. The sector has been opened to competition and the institutional capacity of the State Energy Regulatory Commission (SERC) has been significantly enhanced. In the telecommunications sector, a regulatory framework has been improved substantially through the establishment of the Communications Regulation Commission (CRC) and its institutional upgrading and enactment of a new Telecommunications Law (September 2003) consistent with best EU practice. Furthermore, the monopoly over the provision of fixed voice telecommunication services, leased lines and trans-border voice transmissions has been abolished. Restructuring of the railway sector has been advanced significantly, with the restructuring program successfully addressing increasing efficiency and financial sustainability, improving staff productivity, decreasing subsidies to the railways sector, and opening the domestic freight market has been opened to the private sector and the first private freight operator was licensed in 2004.. The organizational and institutional changes included the separation of the railway company BDZ into a new railway operating company and a new railway infrastructure state enterprise (2002), and the full accounting separation between the freight and railway operating company (2004). In addition, market rules were established to enable the entry of private firms and the eventual privatization of the railway operating company. The network and services have been significantly rationalized through the closure of non-economic lines and termination of loss making passenger services. In 2004, 500 km of loss making lines was closed. The railway labor force was reduced by 13 percent at the end of 2004 compared to the level at the end of 2001. Track access charges were adjusted to remove all cross subsidization between passenger and freight services effective as of October 1, 2004, and cost recovery was increased in inter-city passenger services by a further 20 percent, bringing it to over 90 percent (2004). In the water sector, the new modernization program envisions stronger public private partnerships to address the problems of inefficiency, water losses, and lack of investment. With the enactment of the Water Regulatory Law in January 2005, a modern institutional and regulatory framework water regulator (water supply and sewage services) was developed. The restructuring of banking sector has been particularly impressive in the aftermath of its collapse in 1996- 97. With the privatization of the largest state-owned bank, DSK Bank in 2003, the dominant state insurance company, DZI, in 2002, and the fourth largest, Biochim, in 2002, the privatization to strategic investors has been completed. In addition, several measures have significantly improved a legal and institutional framework for financial intermediation. These included among others the enactment of the Bank Bankruptcy Law in 2003, which has improved conditions in exit for insolvent and closed banks; strengthening of supervisory powers of the Bulgarian National Bank over direct and indirect owners of banks (2003); the amendment of foreclosure provisions of the Civil Procedure Code (2003); strengthening of laws protecting minority shareholders; and integration of supervision over non-bank financial institutions. Source: 9 EBRD transition indicators, converted for the purpose of this analysis along similar lines as for first generation reforms into a single index of structural reforms,4 provide interesting insights into Bulgaria's pace of institutional transformation since the collapse of central planning in 1990. The values of structural reform index show very little progress until 1997. An almost total absence of government commitment to structural reforms during the initial stages of transition is best illustrated by the fact that the value of this index for Bulgaria in 1994 was the same as for Poland in 1990, that is, when structural reforms barely started. Even then the achieved progress in 1995-96 would put Bulgaria on par with Poland's achieved progress in 1992 (Table 4). Table 4: Progress in structural reforms revealed in values of EBRD-based aggregate index in 1991-2004 in selected transition economies (in percent) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bulgaria 0.0 5.7 13.4 17.1 22.9 28.6 28.6 41.9 41.9 41.9 47.6 47.6 51.4 53.3 57.1 Poland 22.9 28.6 28.6 45.7 51.4 57.1 57.1 60.9 62.8 62.8 64.7 66.6 66.6 66.6 66.6 Estonia 0.0 0.0 11.4 38.1 43.8 49.5 51.4 59.0 59.0 64.8 64.8 66.7 68.6 68.6 70.5 Romania 0.0 3.8 3.8 11.4 22.9 28.6 32.4 38.1 36.2 38.1 40.0 41.9 41.9 41.9 45.7 Memorandum: median value in SEE-4 0.0 0.0 0.0 3.8 8.6 11.4 14.3 14.3 23.8 23.8 26.7 27.6 29.5 30.5 31.5 CEEC-8 0.0 11.4 24.8 41.0 47.6 51.5 52.4 57.1 57.1 60.0 61.9 63.8 63.8 64.8 65.7 Notes: (1) the aggregate is the average of scores ranging between 1 (no liberalization) and 4.5 (liberalization at the levels of highly developed market economies) for government and enterprise restructuring, competition policy, banking reform and interest rate liberalization, security markets and non-bank financial institutions, and large scale privatization. It has been normalized with 1=0 and 4.5=100; (2) SEE-4 includes Albania, Bosnia and Herzegovina, Macedonia and Serbia and Montenegro. Source: own calculations based on data from EBRD Annual Transition Reports. Figure 2: Values of index of structural reforms in 1994-2004 60 50 40 30 20 10 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Source: as in Table 4. Perceptible change began to take place only around 2000.The change came in three stages--two piecemeal increases followed more recently by a continuous growth. The value of the index jumped in 1997 and then stayed at the same level in 1998-99 (Figure 2). Another jump occurred in 2000-01, with the value of the index rising from 42 in 1999 to 48 in 2000. In 2002-04 the value of the index increased significantly every year indicating that various pieces of institutional reforms were finally coming together. Progress in large-scale privatization drove the change in 2000 and 2004, in banking reforms in 2002 and 2004, and in governance and enterprise restructuring in 2003. 4 The index is a simple average of measures of reforms, reflecting policies in different areas directly related to restructuring: government and enterprise restructuring, large scale privatization, banking reform and interest rate liberalization, security markets and non-bank financial institutions, and competition policy. The values ranging from 1 (no market economy) to 4.5 (equivalent of institutions and policies in highly industrialized countries) were normalized with 1=0 and 4.5=100. Hence, the value of index of 50 implies that a country is half-way from highly industrialized countries in terms of institutions and policies. 10 How does Bulgaria compare to other countries in the region? While the index based on EBRD's scores--for obvious reasons--does not allow for precision in capturing the institutional progress, it nonetheless indicated the direction of changes and achieved progress. Bulgaria has made huge strides especially since 2001 and is much more advanced than Romania and SEE-4 economies, yet its level of institutional maturity remains well below that achieved by CEEC-8 economies. The value of structural reform index is roughly the same as achieved by a median for CEEC-8 countries around 1997-98. Hence, Bulgaria has still some way to go in order catch up with recent EU entrants. B. Impact on FDI inflows Better governance and regulatory environment, low `hassle' cost of conducting business, labor force, etc., all increase the probability of receiving higher FDI inflows. Empirical research strongly suggests that progress in second-generation reforms, i.e., with the focus on institutions supporting competitive markets, provides explanation in country variation of FDI inflows. Garibaldi et al. (2002) have shown, that the quality of institutions explains the variation in FDI flows to transition economies. In a similar vein, Broadman et al. (2004, p. 20), plotting the data on FDI per capita and EBRD's governance and enterprise restructuring indices for all Balkan countries also find a very strong positive association between these two variables. Countries that have made the largest progress in transition to competitive markets have also attracted the largest inflows of FDI, whereas countries with weaker business climate have been less successful in attracting FDI. Ineffective protection of property rights and weaknesses in contract enforcement always discourage foreign investors.5 In the analysis that follows, we focus, like in the studies mentioned above, on the institutional environment as critical to explaining variation in FDI inflows. In contrast to Broadman et al. (2004), we look at cumulative FDI inflows per capita and use an aggregate EBRD-based indicator measuring the progress in building market-supporting institutions. We take FDI inflows per capita rather than in terms of GDP simply because the latter is positively correlated with higher quality of governance. As discussed earlier, the aggregate indicator is the average of EBRD scores limited to areas not directly associated with first generation reforms. The progress in these areas no longer strongly differentiates among transition economies, as was the case during the initial stages of transition, whereas institution and governance-related domains do. The index is averaged for the 1996-2002 period. The rationale behind averaging over the eight-year period is straightforward: a good climate has to exist for a sufficiently long time to affect investors' decisions. As a second test, we also plot cumulative net FDI flows per capita and the aggregate indicator of quality of governance. As mentioned earlier, the quality of governance is defined here as an average value of the indicators of political stability, government effectiveness and regulatory quality. These three aspects of the business environment are critical to FDI inflows. The indicators pertain to 1996, 1998, 2000 and 2002 and range from ­2.5 (lowest score) and +2.5 (highest score). For the same reason, we take the average over time (Table 5). The variation in the progress in structural reforms and the quality of governance corresponds very closely to the variation in cumulative FDI net inflows per capita over 1990-2003. Countries whose institutions were below 40 percent of the level in industrialized market economies have failed to exceed US$ 500 dollar worth of cumulative net FDI inflows per capita (Figure 3). So have the countries with the negative values of the aggregate governance indicator (Figure 4). 5There are two exceptions to this general rule: First, investments in natural resource sectors tend to flow as long as the state is firmly in control and can shield an investor from the domestic investment climate. Second, even the best climate may fail to attract investors if a country, for instance, has no viable transportation infrastructure, it suffers from chronic shortages of electricity or is located in the middle of nowhere. As we shall argue later, whereas Bulgaria does not have attractive natural resources (oil or natural gas), its infrastructure remains relatively underdeveloped. 11 Table 5: Values of governance index, EBRD-based structural reform index and FDI cumulative inflows per capita in 1990-2003 Albania BiH Bulgaria Croatia Czech R. Estonia Macedonia Hungary Governance -0.38 -0.75 0.10 0.25 0.78 0.88 -0.30 0.85 FDI cumulative per capita 343 280 835 2,147 3,771 3,013 476 3,697 EBRD-based Index 26 13 43 48 65 62 33 72 Latvia Lithuania Moldova Poland Romania S&M Slovak R. Slovenia Governance 0.44 0.50 -0.35 0.62 -0.12 -0.91 0.45 0.75 FDI cumulative per capita 1,461 1,091 156 1,431 482 260 1,647 2,028 EBRD-based Index 49 52 36 63 38 3 51 57 Notes: EBRD-based index ranges between 0 and 100 (fully operational market structures; Governance ranges between minimum of ­ 2.5 (dysfunctional) and +2.5 (perfect quality). Source: EBRD transition reports various issues, IMF balance of payments statistics and derived from data available at www.worldbank.org/wbi/governance/gov_data.htm Figure 3: FDI cumulative inflows over 1990-2003 per capita (in US dollars) and EBRD-based index of progress in second-generation reforms 80 x 70 Inde 60 50 40 30 20 EBRD-based 10 0 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 cumulative FDI per capita Source and notes: See Table 5 Figure 4: Values of governance index and FDI inflows cumulative over 1990-2003 per capita (in US dollars) 4,000 y = 1432x2 + 1616x + 668 R2 = 0.82 3,500 at 3,000 capi 2,500 perI FD 2,000 vetialu 1,500 1,000 cum 500 0 -1.00 -0.80 -0.60 -0.40 -0.20 0.00 0.20 0.40 0.60 0.80 1.00 Governance Index Sources and notes: See Table 5. 12 Like all other countries, except for Moldova, Bulgaria was not an outlier. Its governance was in the positive range and its institutional development lagged less than 60 percent behind the level of highly industrialized economies. The net cumulative FDI inflows predictably exceeded US$ 500 per capita putting it well ahead of countries that had negative scores of governance and low EBRD scores. On the other, Moldova attracted less FDI than then the structural index would suggest. The reason is straightforward: the absence of political stability as captured in the value of governance index. Similarities between the results derived from plotting cumulative FDI inflows and levels of governance and structural reforms come as no surprise. Both indexes are highly correlated, with the value of correlation coefficient of 97 percent (Table 6). Both indices are also similarly strongly and positively correlated with cumulative FDI inflows per capita, with correlation coefficients exceeding 80 percent. It is interesting to note that the variation in values of CPI and FDI inflows is also positively correlated, albeit somehow weaker than the two other indices. Table 6: Values of correlation coefficients among corruption, governance, reform progress and FDI CPI Governance EBRD Corruption Perception Index (CPI) 1 0.83 0.77 Governance aggregate index 0.83 1 0.97 EBRD-based index 0.77 0.97 1 Cumulative net FDI 0.75 0.84 0.82 Source: See Table 5. Data for Bulgaria as well as for Romania provide strong support to the link between progress in governance and FDI inflows. The 1990-2003 time profile of FDI flows to Bulgaria per capita also appears to coincide with changes in the values of structural reform index (see Table 4 and Figure 2 above). So does Romania's time profile. Until 1997 annual FDI inflows to Bulgaria and Romania were very low. There had been little improvement in terms of the structural reform index. Surge in FDI inflows since 1997 has accompanied the implementation of structural reforms. In 2003 FDI per capita in Bulgaria exceeded for the first time the population-weighted average FDI per capita flows to CEEC-8.6 If structural reforms including large-scale privatization to strategic investors explain most of FDI inflows, then recent progress in Bulgaria augurs well for the future. Indeed, according to the most recent estimates, FDI flows increased 6 percent last year and amounted to 10 percent of the GDP or US $286 per capita. Privatization accounted for 23 percent of total FDI inflows over 1999-2003. The share was 26 percent in 2003. On the other hand, relative `stagnation' in Romania's structural reforms was also accompanied by stagnant FDI inflows per capita in the period following large privatization in 1998 (Figure 5). Hence, the progress achieved in building market-supporting institutions combined with the higher quality of governance explains most of the variation in flows of FDI to sixteen transition economies, CEEC- 10 and SEE-6. Bulgaria's inward FDI stock, as approximated by cumulative net FDI inflows since the collapse of central planning in 1990, is in line with what one would expect given its record in second generation reforms. This is the case for both indices of governance quality and structural reforms' progress averaged over 1998-2004. How does the emphasis on structural reforms match studies pointing to other factors explaining FDI flows? The conclusion about the primacy of institutions appears to contradict other findings on determinants of FDI inflows. Most of them identify unit labor cost, openness to trade, size of the host economy, the proximity to the EU as major factors explaining differences in FDI inflows. Some argue that country-specific risk does not influence FDI inflows (Bevan and Estrin 2004), whereas others have found credit ratings as an important explanatory variable (Janicki and Wunnava 2003). It is also interesting to note that announcement about timetables for admission to the EU has positively influenced FDI flows (Bevan and Estrin 2004). 6Because of delays in reporting by foreign firms of reinvested earnings, the BCB revised upward in January 2005 its estimate of FDI for 2003 from EUR 1.25bn to EUR 1.85bn. 13 Figure 5: FDI per capita in Bulgaria, Romania and CEEC-8 population weighted average in 1990-2003 (in US dollars) 250 Romania 200 capita) 150 Bulgaria per 100 (FDI 50 CEEC-8 weighted average 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Source: IMF balance of payments statistics, EBRD data for data missing in the IMF statistics, and Bulgarian national statistics. Data for 2003 include the upward revision (48%) of the BCB (see the footnote below). The contradiction is more apparent than real. Consider first that countries bordering the EU-15 (except those bordering Greece) have also moved fast in transition to competitive markets. In consequence, proximity has also somehow influenced the pace of transition, although the presence of potential foreign investors also probably had an influence on reforms. Second, richer countries are, to paraphrase Fisher, Sahay and Vegh (1998), closer to Brussels. Richer countries have also better institutions and are thus more attractive to FDI. Third, openness to trade influences the extent to which domestic markets are competitive. Contestability of domestic markets, in turn, reduces the extent to which distortions might affect the developments in unit labor costs. Both EBRD-based indicator and governance indicator indirectly captures contestability of domestic markets and thereby developments in unit labor costs. While policy makers cannot do much about the size of a market or proximity to the EU, they can do quite a lot in designing institutions and changing policies. The latter also usually influence both unit labor cost and openness to trade and thus boost country's attractiveness to foreign investors. More importantly, there is evidence that institutional quality towers over other determinants of FDI location. Campos and Kinoshita (2003) examined three groups of determinants: The first group comprised cost of labor, the size of domestic market, skills of labor, quality of infrastructure, and proximity to EU markets. The second group included institutions, macroeconomic policy and other policies that facilitate conduct of business. The third group related to agglomeration economies, i.e., new investors following decisions made by earlier investors to take advantage of positive spillovers related to knowledge spillovers, and availability of specialized labor and intermediate inputs. Their main finding was that institutions and agglomeration economies explain directions of FDI flows to transition economies--taken together they outweigh other determinants related to labor costs, proximity to EU markets and size of domestic market. Recently observed increases in FDI inflows into Bulgaria accompanied by perceptible improvements in the quality of governance and market-supporting institutions appear to corroborate this finding. It also reinforces the policy message about the importance of staying the reform course and suggests that recently observed increase in FDI inflows would generate positive spillovers attracting new foreign investors in the future. 14 C. Conclusion Bulgaria has made significant progress, especially since 2001, in raising the overall quality of governance, reducing corruption and introducing institutions and policies supporting competitive markets. These measures have made Bulgaria more attractive for high-quality FDI. Foreign investors appear to have responded well to the improved governance, as FDI inflows significantly increased in 2000-04. Yet, Bulgaria has still to close the gap vis-à-vis CEEC-8. An important question to which we shall now turn is the extent to which reforms have reduced the `hassle' cost of entering, conducting business (both domestically and internationally), and exiting. In other words, the major question addressed is whether the recent changes have trickled down to lower levels of bureaucracy directly dealing with businesses. 2.3. The `hassle' cost of doing business: domestic and external interaction The discussion so far has focused on first- and second-stage reforms and their impact on FDI inflows assessed in the context of CEEC-8 and SEE economies. It has examined general framework within which economic activities take place leaving aside an important intertwined issue of their microeconomic impact. Well-developed market supporting institutions, the extent of contestability of domestic markets, the capacity of government to enforce rule of law or macrostability are all crucial to establishing environment friendly to conducting business operations. But so are regulations and other conditions influencing transaction costs. Regulations are usually necessary for various reasons but they do not have to be unnecessarily burdensome or costly. These can be discussed from two different viewpoints. The first abstracts from international interaction whereas the second focuses on issues related directly to costs related to foreign transactions. High `hassle' cost of conducting business provides a strong disincentive to establishing new firms and encourages existing firms to slip into the informal economy. Along similar lines, burdensome customs procedures, low efficiency of ports, high transportation costs due to regulatory arrangements and high costs of other services raise transaction cost and undermine competitiveness of domestic producers in international markets. A. Business climate: domestic dimension Serious effort to address high regulatory burden on conducting business activity began in Bulgaria in 2002, with the establishment of the high level inter-ministerial group to review, eliminate and simplify regulation of business activities. Its results were twofold: the proposal for streamlining (117 of 361) and elimination of some regulatory regimes (75 of 361) and the draft of a Law on Administrative Regulation and Administrative Control of Economic Activities subsequently adopted in 2004 (Djankov 2005). The latter has invalidated all existing procedures for the creation of new licensing regimes and has established the new principles for the introduction of new regulatory regimes significantly limiting the scope of new regulations by line ministries and other government agencies. As a result, the administrative burden on business has been significantly reduced. Yet, the administrative burden of conducting business appears to remain excessive and many Bulgarian firms perceive it as the major weakness of business climate in the Bulgarian economy. While deregulation should be approached as a continuous process managed from the top levels of administration in cooperation with civil society organizations, this was, however, a one-time effort. No institutional structure for deregulation has been set up to monitor and devise deregulatory schemes. The institutional capacity of various government agencies is not sufficient to implement effectively various provisions of the new Law on Administrative Regulation. No comprehensive database containing information on regulations has been established. Various government agencies have no access to shared information on businesses. There are still 39 licensing regimes in place in Bulgaria, as compared, for instance, to five in Poland. The silent consent applies only to a handful or licenses, and there is no legislation limiting the number and duration of various inspections. Last but not least, a multitude of overlapping regulatory structures still remains in place. How does Bulgaria's business climate compare with those of its regional peers in each of the six areas pertinent to conducting business activity? Regional peers are here new EU member states from Central 15 Europe (CEEC-8) and Romania that, together with Bulgaria, are on the EU accession path as well as ten CIS economies (no data are available for Tajikistan and Turkmenistan). The diagnostic tool is the World Bank's Cost of Doing Business comparative survey (World Bank 2005a). The compiled data shed lights on tangible components shaping the legal cost of conducting business in Bulgaria in comparative perspective. We look into seven areas covered by the survey: starting business, labor market flexibility, registering property, contract enforcement, bankruptcy, and protection of investors. As a first step, we assess the overall formal regulatory ease of doing business relative to transition countries. Data tabulated in Table 7 contains information on overall ranking in terms of ease of doing business overall and in each area. It was derived as follows: Each area has between one (protection of investors--disclosure measure) and five different indicators (access to credit and labor market flexibility). Except for contract enforcement (four indicators), three indicators describe the state of regulatory practices in other areas. Since rigidity of employment indicator is an average of three other indicators, we take into account the average indicator together with firing cost. We normalize each indicator taking best practice among CIS/CEEC states as 100. We calculate the average for each of seven areas covered by `doing business' survey. The average across areas gives an indication of the overall ease/difficulty of doing business in a country as juxtaposed against the best conceivable regulatory regime, that is consisting of most business- friendly arrangements in the region. The higher values ranging between maximum of 100 (best) and 0 (worst) indicate a better relative performance. Table 7: Ranking of CIS/CEEC states in terms of overall `formal' ease of doing business in 2004 Ease of Starting Flexibility of Registering Enforcing Closing Protecting Getting doing business employment property contracts business investors credit business Lithuania 58 37 74 86 68 100 40 66 Latvia 48 30 15 77 100 83 70 61 Slovak Republic 39 100 27 49 31 100 54 57 Czech Republic 33 56 28 72 18 100 89 57 Estonia 40 37 40 83 45 67 63 53 Armenia 47 64 38 65 68 50 37 53 Georgia 39 51 18 54 53 83 56 50 Poland 32 49 19 51 60 67 57 48 Russia 43 69 21 62 77 50 13 48 Hungary 34 38 27 76 36 83 40 48 Kazakhstan 41 69 18 60 24 83 23 45 Kyrgyz Republic 70 54 22 28 53 50 31 44 Bulgaria 35 46 19 48 41 33 69 42 Moldova 33 50 26 51 41 50 36 41 Uzbekistan 33 39 10 45 47 67 39 40 Romania 79 17 17 51 27 33 53 39 Belarus 18 50 48 54 44 17 44 39 Azerbaijan 44 33 29 56 43 33 24 38 Slovenia 28 28 20 45 27 67 48 38 Ukraine 27 17 14 65 32 50 28 33 Memorandum: Best CIS/CEEC performance in percent of the best practice worldwide 52 0 10 77 58 86 54 48 Source: data from Doing Business in 2005. Removing Obstacles to Growth. The World Bank and Oxford University Press, Washington DC, 2005. Although at a first glance, these may not seem to be very demanding benchmarks in particular for flexibility of employment and registering property where the best CEEC/CIS-wide practice diverges widely from the best practice worldwide (last line in Table 7), this is not the case. Consider that Lithuania and Slovakia were among top 20 economies in the world ranked on the overall ease of doing business in 2004 (World Bank 2005a, p. 2). The ranking appears to defy expectations of the edge that CEEC-10 should have over CIS economies thanks to EU accession path. Although top five positions do not come as surprise, rankings of other countries 16 do. Slovenia--the most developed in the CEEC/CIS group--has better regulatory climate than only Ukraine and ranks 19th. Hungary, which attracted huge amounts of FDI, Poland, ranks 10th below Russia and Poland. The latter, the bold reformer during the initial stages of transition in the early 1990's ranks below the two Transcaucasian states--Armenia and Georgia. Last but not least, Bulgaria has more hostile regulatory environment than 12 CEEC/CIS countries including, in addition to the above-mentioned, Kazakhstan and Kyrgyz Republic. The results may be counterintuitive but revealing. Note that rankings take into account simplicity and cost of formal business regulations. They do not address corruption, illicit payments, taxation and other informal interventions that might be pursued by respective administrations. Simple regulations convolutedly executed may be as bad, if not worse, than more complex ones enforced by less corrupt administration. In order to address these other components affecting the cost of doing business, we supplement the formal ease of doing business with information on the size of informal economy. In consequence, a greater weight is given to formal regulations in some areas, directly affecting decisions to stay or exit shadow economy (e.g., access to credit and contract enforcement). It also indirectly captures other aspects of administration/business interplay including corruption.7 Table 8 presents the values of `revealed' ease of doing business calculated as an average of formal ease of doing business and the normalized size of the informal sector taking the lowest size of informal economy as percent of GNI from the CIS/CEEC group as 100. Table 8: The size of informal economy and `revealed' ease of doing business Informal economy "Revealed' ease of doing Ranking in `formal' ease Current ranking Memorandum: % of 2003 GNI business of doing business CPI, 2000-04 Slovak R. 18.9 78.5 3 1 3.8 Czech R. 19.1 77.8 4 2 3.9 Estonia n/a 74.0 5 3 5.7 Lithuania 30.3 64.2 1 4 4.7 Hungary 25.1 61.5 10 5 4.9 Poland 27.6 58.2 8 6 3.8 Latvia 39.9 53.9 2 7 3.8 Slovenia 27.1 53.7 19 8 6.0 Uzbekistan 34.1 47.7 15 9 2.5 Romania 34.4 47.2 16 10 2.7 Armenia 46.3 46.8 6 11 3.0 Bulgaria 36.9 46.4 13 12 4.0 Kyrgyz R. 39.8 45.8 12 13 2.2 Kazakhstan 43.2 44.6 11 14 2.3 Russia 46.1 44.4 9 15 2.7 Moldova 45.1 41.4 14 16 2.3 Georgia 67.3 39.3 7 17 2.1 Belarus 48.1 39.2 17 18 4.1 Ukraine 52.2 34.6 20 19 2.3 Azerbaijan 60.1 34.5 18 20 1.9 Note: CPI--Corruption perception index, the average for 2000-04. It assumes values between 1 (maximum incidence of corruption) and 10 (minimum) Source: Own calculations based on data in WB 2005a and Web site of Transparency International. The change in ranking is rather dramatic and brings it more in line with expectations. Slovenia moves up 11 positions to join other new EU member states that have moved decisively to the top eight positions in terms of real business friendliness of regulatory regime taking into account the extent to which entrepreneurs prefer to stay in the shadow economy. So has Romania by six positions. Bulgaria's ranking, however, has barely changed. 7The correlation between the values of Corruption Perception Indices, as reported by Transparency International, and the size of informal sector in percent of GNI for CEEC/CIS states is 67 percent. For a discussion on why and how this indicator is derived, see Kaminski (2005a). 17 What does it all say about business climate in Bulgaria as compared to other former centrally planned economies? In contrast to Slovenia or Romania, Bulgaria has a simpler formal regulatory regime but overall less business-friendly climate. Complexities of Slovenian formal arrangements do not discourage businesses from going formal. On the other hand, simpler regulations in Bulgaria do. The question to which we shall now turn is what are these other ingredients erecting barriers to conducting business in Bulgaria. We shall take a closer look at areas of formal cost of doing business there. In relation to other CEEC/CIS economies Bulgaria's regulatory environment scores particularly low on labor market flexibility, protecting investors, starting a business and on system of property registration (see Table 7 above).8 All are important: labor market flexibility augurs well for boosting domestic economic activity and attracting FDI,9 and ease of registering property is the necessary condition, albeit not sufficient, to bring assets into formal sector which, in turn, allows using them to obtain financing. Data in Table 9 corroborate these observations. Table 9: Doing business in Bulgaria, Lithuania, and New Zealand and in `best regulatory environment' in 2004 Area Indicator Bulgaria Lithuania New Zealand Best practice Starting a business Number of procedures 10 8 2 2 Time (days) 32 26 12 2 Cost (% of income per capita) 10.3 3.7 0.2 0.0 Minimum capital (% of income per capita) 116.6 62.8 0.0 0.0 Hiring and firing workers Difficulty of hiring index (0-100) 33 33 11 0 Rigidity of hours index (0-100) 40 60 0 0 Difficulty of firing index (0-100) 10 30 10 0 Rigidity of employment index (0-100) 28 41 7 0 Firing costs (weeks) 30 34 0 0 Registering property Number of procedures 9 3 2 1 Time (days) 19 3 2 1 Cost (% of property value) 2.4 0.9 0.2 0.0 Enforcing contracts Number of procedures 34 17 19 11 Time (days) 440 154 50 27 Cost (% of debt) 14.0 14.1 4.8 4.2 Closing a business Time of insolvency (years) 3.3 1.2 2.0 0.41 Cost (% of estate) 8.0 8.0 4.0 1.0 Recovery rate (cents on the US$) 34.2 52.4 71.4 92.0 Protecting investors Disclosure index (0-7) 2 6 5 7 Getting credit Cost to create collateral (% of income per 1.0 4.1 0.0 0.0 capita) Legal rights index (0-10) 6 4 9 10 Credit Information Index (0-6) 4 3 5 6 Public registry coverage (per 1,000 adults) 13 44 0 637 Private Bureau Coverage (per 1,000 0 0 978 1,000 adults) Source: data from Doing Business in 2005. Removing Obstacles to Growth. The World Bank and Oxford University Press, Washington DC, 2005. Considering that Bulgaria has to compensate for its relative longer distance to EU markets than most other new EU member states, the frame of reference for Bulgaria's regulatory reforms should be at least Baltic states. They are even more remotely located than Bulgaria from the center of EU economic gravity 8It is interesting to note, however, that labor regulations are the fifth most problematic regulatory constraint, with almost 18 percent of respondents to the FIAS survey picking it up as a "major" or "very severe" problem. Yet, since in 2002 labor regulations were the second leading specific regulatory constraint, after tax administration, this suggests an improvement (FIAS 2004, p. 24). 9As a recent study (Javorcik and Spatareanu 2004) shows, the magnitude in labor market flexibility is significant. For instance, with the flexibility of the host country labor market increasing from the level of Slovakia (inflexible prior to recent reforms) to the level of Hungary (flexible), the volume of investment goes up by between 14 and 18 percent, all else being equal. Moreover, FDI in services sectors appear to be more sensitive to labor regulations than investments into manufacturing 18 epitomized by geographical location of Brussels, although they clearly benefit from proximity to Finland. It appears that they have sought to compensate for relative remoteness by assigning priority to regulatory reforms that would increase attractiveness of their countries to foreign and domestic investors. This is a strategy worth emulating. As can be seen from data in Table 9 presenting indicators for the best performer among CIS/CEEC states in terms of formal ease of doing business, the best performer worldwide, and the values of the `best' scores, the road to regulatory environment enhancing microeconomic efficiency and growth is still a long one. Labor regulations, in particular, remain an important barrier to business activity putting Bulgaria at a substantive disadvantage relative to other countries in Central and South Eastern Europe (Djankov 2005). The outdated socialist-era Labor Code and other labor regulations are full of provisions discriminating in favor of currently employed, providing incentives to absenteeism and generally discouraging new employment. Examples abound. Regulations provide for automatic increases due to seniority even if previous experience has no relation to the current job. The minimum wage is set at a high level. An employee on sick leave receives 80 percent of his or her pay, which, however, is then not subject to personal income tax. Furthermore, there is no limit on the number of sick days that an employee can take. Under these circumstances, employees have little incentive to work and employers have no incentive to employ them unless under very special arrangements.10 Bulgaria's regulatory environment is the most burdensome for private firms in there are three other areas--starting a business, contract enforcement, and protection of investors' rights. As for business startups, the major reasons for Bulgaria's unfriendly regulatory arrangements relate to high administrative costs of starting a business. While in many countries business registration amounts to simply entering information into a statistical database,11 Bulgarian regulations call for judicial assessment of the underlying documents. This is redundant, as there are other means (licenses and permits) to control business entry. The registration costs amount to 10.3 percent of Bulgaria's GDP per capita and minimum capital to be deposited amount to 117 percent of the GDP per capita. In other words, it costs almost US$ 300 in various administrative fees to register a business and one has to deposit around US$ 3000 as a minimum capital requirement. This is in relative terms almost three times and two times more than in Lithuania--an economy with the lowest hassle cost of conducting business among CEEC/CIS states (Table 9). Although these significant capital requirements are often justified as protection against damage from failing businesses, countries with most friendly conditions for business startups do impose very small costs, if any at all. In New Zealand, with the lowest hassle cost of doing business worldwide, the total cost amounts to 0.12 percent of the GDP. There is one country--Denmark--with zero cost. The reasoning behind simplicity and low administrative cost is simple: encouragement of new business activity and making sure it takes place in the formal economy. A high capital requirement pushes entrepreneurs into the informal economy. The welfare cost is significant as firms operating in the informal economy cannot reap benefits of scale and deprive the government of tax revenue. The latter might allow lowering tax rates thus encouraging economic activity. One should also add that there is little reason to have 11 administrative procedures. Although this number does not look bad against Lithuania, it is high when compared against best performers worldwide--Australia, Canada, and New Zealand. As their experience demonstrates, two procedures suffice.12 10In the late 1990s, around 40 percent of workforce was employed at the minimum wage level (to minimize social security and other taxes), with additional renumeration paid in cash directly to employees. The 2002 tax reform only weakened incentives to do so, but has not terminated this practice. 11The only four exceptions among EU-25 economies are Estonia, Hungary, Slovenia and Spain. These countries require judicial review prior to registration (Djankov 2005). 12The European Commission recommends no more than four procedures for individual enterprises and general partnerships (up to 11 for a public limited company), two contact-points for entrepreneurs and no minimum capital requirements for individual enterprises and general partnerships. For private limited and public limited companies, the benchmarks for minimum paid-up share capital are E 3,000 and E 15,000 correspondingly (EC 2002). 19 It is interesting to note that a recent FIAS study has found that firms covered by their survey do not regard enterprise registration as the most questionable administrative procedures encountered by businesses (FIAS 2004, p. 10). However, this merely suggests that other regulatory areas impose heavier burden and should not undercut the ongoing reform effort to improve conditions in business administration in Bulgaria. Several different separate registration regimes involving the courts, the tax administration, the statistics institute, and often the local government all contribute to an unnecessarily arcane system. Another area `standing out' in a negative sense is contract enforcement. The process for a firm is lengthy; 440 days as compared with 335 days in Romania, 154 days in Lithuania and 50 days in New Zealand. Other sources corroborate this assessment of ineffective rule of law and point to a systemic problem. A recent World Bank study argues that (a) trust in the courts remains in Bulgaria low relative to other state institutions; (b) firms do not trust in the ability of courts to enforce decisions; and (c) the gap between demand for judicial services and the capacity of courts to deliver has been on the increase (World Bank 2005b). The absence of specialized commercial courts, summary proceedings and default judgments, out of court settlements and heavy reliance of Bulgarian courts on written over oral procedures together with their complexities further embitter contract enforcement. In all, while fewer than half of firms in almost all transition countries surveyed in 2002 believed that courts were able to enforce their decisions, the situation in Bulgaria appears to be much worse. The dominant perception among Bulgarian firms that it is difficult to have judicial decisions enforced has not abated. This is a serious problem, as demand for judicial services tends to grow with progress in structural reforms and transition to competitive markets. The absence of efficient courts suppresses investments and business transactions, as it limits investor confidence in the ability of the courts to enforce contracts, in ownership and shareholders rights (see below) and intellectual property rights. Closely linked to the contract enforcement, the protection of investor rights is the last regulatory area that is well below international standards. The index captures seven different aspects of disclosure of financial information and ownership protecting investors and shareholders against fraud. The absence of regulations compelling disclosure has several negative consequences: investment are lower as potential investors fear of expropriation; the stock market is undercapitalized; economic growth is lower than under full disclosure;13 and returns from investment are lower (WB 2005a, p. 56-57). With the value of the disclosure index of 2 with its value ranging between 0 and 7, Bulgaria together with Romania was at the very bottom of CEEC-11. For other countries the values of this index ranged between 4 (Croatia, Estonia, Poland and Slovenia), 5 (Hungary and Latvia) and 6 (Czech Republic, Lithuania and Slovakia). Bulgaria also lags behind some SEE-countries, for instance, the index for Macedonia was 4. There are also other factors compounding the overall hassle cost of doing business. Foreign investors often complain about government bureaucracy, poor infrastructure and frequent changes in the legal framework. This view seems to be shared by local entrepreneurs. FIAS survey points to firms' complaints about dealings with tax administration, discrepancies in legal interpretations of procedures across municipalities and fear by enterprises of disputing decisions taken by the administration (FIAS 2004). As for complaints about tax administration, uncertainty over tax rates and rules outweigh other issues, with 65 percent of firms surveyed identifying it as a major burden (Figure 6). Needless to add that lack of stability in tax rates and rules make business planning extremely difficult. So do other deficiencies in dealings with the tax administration identified by firms. The rapidly changing tax rules and their differing administrative interpretations create an environment of uncertainty with firms challenging arbitrary decisions. However, the problem is that--according to almost half of respondents--appeals mechanisms appear to be ineffective and penalties, both severe and capricious. 13For instance, it is estimated that if Jordan had the same regulations as highly developed OECD economies, its turnover in equities would quadruple and annual economic growth would be by up to 1.1 percentage higher a year (WB 2005a, p. 57) 20 Hence, in spite of many improvements also reported in the FIAS study, the state administration has not as yet fully transitioned from its previous role of a micromanager to that of a facilitator required in a modern market economy. Figure 6: Firms' responses to questions concerning burdens associated with authorities' implementation of tax regulations in 2004 (percent of firms) Frequency of changes in tax rates rules Effectiveness of appeals mechanism Severity of penalties Availability of information on requirements Completing tax forms/filing Treatment by tax authorities Burden of Audits 0 10 20 30 40 50 60 70 percent of firms Source: FIAS 2004, p. 22. This analysis provides evidence supporting the view that despite the fact that Bulgaria has made significant progress in simplifying regulations, the business regulatory environment remains an obstacle to the private sector development and economic growth. Business regulations hinder competitiveness and productivity of Bulgarian firms. Although the hassle cost of doing business is now overall lower than in Croatia or Romania, it remains in some areas considerably higher than in CEEC-8, particularly in Baltic states. B. Business environment, infrastructure and external interaction While the above section has focused on the domestic dimension of business climate, i.e., regulatory regimes and measures directly affecting entry, conduct and exit for firms, this section will focus on policies and institutions affecting external interaction of domestic firms. These include both procedures for simplification and harmonization of international trade, state of infrastructure and its management together with provision of such backbone services as telecommunications, banking, insurance, transportation, business services, etc. Together with customs, related border clearance regulatory procedures, technical standards regulations and port efficiency, they shape the ease and speed with which goods and services move across national borders and, therefore, are crucial to trade in goods. Weaknesses in provision of these services make impossible for domestic firms to participate in the emerging division of labor based on international outsourcing, just-in-time production and supply-chain management. Increasingly sliced value-chains, with the individual production stages being moved to countries with corresponding comparative advantages have become trademarks of a current global economic landscape. Interaction among "production blocs" of border-spanning production networks is particularly vulnerable to delays and disruptions between individual stages of the supply chain due to weaknesses in service links. Hence, poor quality of backbone services and trade facilitation acts deters foreign firms from incorporating domestic firms into their supply chains. It also acts as a barrier to other types of trade. Liberalization of trade in services may improve the quality and availability of services through competition, economies of scale and, last but not least, incentives to policy makers to improve regulatory 21 environment. Benefits of services liberalization are not limited to services sectors themselves: they impact all other economic activities. Considering that services contribute on average around 10-20 percent to the production cost of a product and account for all trading costs (transport, trade finance, insurance, communications, distribution services), savings from stronger competition from foreign providers indeed can be substantial (Hodge 2002). So do gains in competitiveness in international markets of both services and goods. Before evaluating the quality of business environment, as it shapes external interaction, several caveats are in order. First, Bulgaria has only recently implemented a number of measures in the context of accession into the EU, which has both provided an impetus and facilitated the design of policy reforms spanning across many services sectors. These as a rule included measures related to market opening, unbundling, third party access, public service obligations and regulation. Given their recent gradual implementation and complexity, their impact is yet to be statistically captured. In consequence, available data may significantly understate current performance. Second, their impact will be positive, as convergence to the acquis makes markets contestable for foreign providers of services and harmonized with EU-wide regulatory provisions represents huge progress over the status quo. Both are crucial for accession, since the EU is already a common market striving to move towards free trade in services and investment. Both increase contestability, improve the quality of services provided and reduce their cost. Third, their impact will not only be positive but also economy-wide, as convergence to the acquis communautaire, i.e., specific directives that must be implemented by a member state, stretches across a large number of service sectors including such important backbone service industries as financial services, transport, telecommunications, and energy. The convergence involves the strict application of competition rules and State aid disciplines leading to greater transparency and competition. In order to assess the quality of environment supporting trade in goods, as a point of departure, we use indicators of trade facilitation as developed in Wilson, Mann and Otsuki (2004). They identify four indicators of capacity to facilitate trade. These are port efficiency, customs environment, regulatory environment, and Information Technology infrastructure. Port efficiency measures the quality of infrastructure of maritime ports and airports. Customs environment measures direct customs costs as well as administrative transparency of customs and border crossings. Regulatory environment measures the country's approach to regulations and their quality. IT infrastructure measures the extent to which an economy has the necessary domestic infrastructure (such as telecommunications, financial intermediaries, and logistics firms) and is using networked information to improve efficiency and to transform activities to enhance economic activity.14 The values for these variables are drawn from, among others, Global Competitiveness Report 2003/2004. We take the average efficiency level of the EU-15 to benchmark Bulgaria's performance in comparative perspective. Except for the quality of regulations or regulatory approach Bulgaria exhibits a low level of performance in all areas of trade facilitation, relative to the EU-15 countries (Figure 7). Development levels in port efficiency, custom regimes, regulatory policy, and IT infrastructure are approximately 54 percent, 71 percent, 84 percent, and 68 percent of the EU-15 level, respectively. The new EU-8 member countries on average exhibit a much higher level of development than Bulgaria using these four indicators. But Bulgaria outperforms Romania in all areas except in port efficiency. The weakest areas of trade facilitation in Bulgaria relative to the levels of efficiency achieved on average by both CEEC-8 and the EU- 15 are management of ports, IT infrastructure and customs. They are respectively 26 percentage points and 16 percentage points for the latter two below the average for CEEC-8. 14WMO's indicators rely on three sources--Global Competitiveness Report, World Competitiveness Yearbook, and a dataset compiled in Kaufmann, Kraay and Zoido-Lobaton (1999). Each indicator is constructed as a simple average of two inputs as described in the appendix.14 22 Figure 7: Trade Facilitation Benchmarks for Bulgaria in comparative perspective against the EU-15 average level 1.0 0.9 0.8 Bulgaria 0.7 0.6 Romania 0.5 Estonia 0.4 0.3 Lithuania 0.2 0.1 CEEC-8 average 0.0 Ports Customs Regulation IT infrastructure Source: Wilson, Mann, Otsuki database (2004) and Wilson, Luo and Broadman (2004). The gap between performances in terms of regulatory environment measures in which Bulgaria's score is higher than the average for CEEC-8 than on other three indicators of trade facilitation appears to stem from relatively recent vintage of regulatory changes. Better regulatory arrangements usually lead, albeit with some lead-time, to improvements in performance. Telecommunications is one of these areas where new laws are yet to significantly alter the quality of IT infrastructure. Their availability and pricing of telephone services deteriorated in comparison to CEEC-8 countries in 1999-2002 on all major indicators except in density of telephone mainlines (Table 10). But the latter was exceptionally high in Bulgaria even before the collapse of central planning. It still exceeded the average for CEEC-8 in 2001 and was higher than in Hungary in 2002. Yet, despite a significant fall in the length of a telephone-waiting list between 1999 and 2002, the gap vis-à-vis CEEC-8 had expanded. The gap in terms of pricing local telephone calls had also fallen, although the telephone average cost of a local call was still well below that in CEEC-8 and Romania in 2002. But such a low cost of local telephone calls--more than five-times lower than on average in CEEC-8 or Romania--is not the result of high efficiency but cross-subsidization. In contrast to Romania, which scored higher than Bulgaria on the indicator of IT infrastructure, Bulgaria's local rates were roughly the same as the average in SEE-3 economies. But the difference vis-à-vis the CEEC-8 slightly declined suggesting slow movement towards a regulatory structure that allows for increased competition. Indeed, movement towards a less competition distorting regulatory structure has accelerated over the last two years. Although Bulgarian Telecommunications Company (BTC) still has monopoly on the provision of voice and leased line services until January 1, 2005, two developments augur well for improvement in the quality of IT infrastructure. First, the 2003 Telecommunications Law has introduced greater transparency in the institutional regulatory framework of the Bulgarian telecommunications sector. It also contains a provision allowing rival service providers access to BTC's subscribers to offer cable, satellite or wireless services from the original liberalization date. Second, on February 20, 2004, 65 percent of BTC was sold to Viva Ventures, a wholly owned subsidiary of U.S. Company Advent.15 While in the absence of competition, privatization alone does not guarantee better services, this may, however, trigger long neglected modernization of the network, which has one of the lowest digitalization rates in the region. While recent developments suggest possibilities for a significant progress in IT infrastructure, this does not appear to be the case for port efficiency for two reasons. First, port inefficiencies appear to be the 15See information at http//www.osec.doc.gov/obl/romaniabulgariatrademission/ BulgariaTelecommunications.htm 23 main bottleneck. It is the weakest area of trade facilitation in Bulgaria relative to the levels of efficiency achieved on average by CEEC-8 on four trade facilitation indicators. It is 26 percentage points below the CEEC-8 average level and 46 percentage points below the average for EU-15. It also stands at 65 percent of Romania's level in port efficiency. In all, the gap is larger than in IT infrastructure indicating a longer way to go to achieve CEEC-8 levels of efficiency. Table 10: Indicators of telecommunications services in Bulgaria in comparative perspective in 1999, 2001 and 2002 Telephone mainlines Telephone mainlines, Telephone average Telephone average International (per 1,000 people) waiting list (per cost of local call cost of call to US telecom, outgoing 1,000) (US$ per three (US$ per three traffic (minutes per minutes) minutes) subscriber) 1999 2001 2002 1999 2001 2002 1999 2001 2002 1999 2001 2002 1999 2001 2002 Average CEEC-8 333 339 n/a 19.9 8.2 2.1 0.10 0.10 0.11 2.42 1.37 1.29 89 88 99 of which: Hungary 371 375 361 7.8 2.03 0.79 0.13 0.09 0.13 1.4 0.98 0.79 61 51 66 Bulgaria 342 359 368 42 23 19 0.00 0.00 0.02 n/a 2.37 1.45 39 43 48 Romania 167 184 194 33 26 24 n/a n/a 0.11 2.35 1.96 1.82 41 44 50 Average SEE-3 143 185 195 18 15 n/a 0.02 0.02 0.02 4.12 3.16 2.44 227 155 149 Bulgaria in terms of percent of Average CEEC-8 103 106 n/a 213 285 1844 2 2 18 n/a 173 113 44 49 48 of which: Hungary 92.3 95.8 102 542 1148 2380 2 2 15 n/a 242 185 63 84 73 Romania 205 196 189 127 90 77 n/a n/a 18 n/a 121 80 94 99 95 Average SEE-3 239 194 189 229 155 n/a 13 9 109 n/a 75 59 17 28 32 Notes: SEE-3 includes only Albania, Moldova and Serbia and Montenegro, as no data available for other SEE economies. Source: World Development Indicators, World Bank's SimaWeb database. Second, reforms of port management have barely begun. The state of the art arrangements call for the introduction of the landlord port, with the separation of operational and commercial functions, privatization of port services, concessioning of large container and bulk terminals to private operators. Bulgaria's Ministry of Transport and Communications issued a port operator's concession program notice only in 2004 (see www.mtc.government.bg). It has invited interested parties to participate in the future bidding process for operation and maintenance of ports' terminals, as well as for the investment program for reconstruction and development of the existing and new terminals. The focus so far has been almost exclusively on airports leaving aside maritime ports, even more pertinent to trade in goods. A competitive tender, announced in August 2004, is limited to two airports in Bourgas and Varna. As of the end of January 2005, the concession rights for Burgas and Varna airports, which are expected to attract investments of nearly EUR 130 M over the next couple of years have not been allotted. Yet, Bulgaria will have to address these inefficiencies before accession to the EU. The EU has been undergoing liberalization of rail transport, landing rights and access to airports (allocation of slots). It has abolished the queuing system for inland waterway markets and has been enforcing of rules on work practices in the road haulage sector. Furthermore, new rules call for the separation of public monopoly providers of port, rail and other transport services and reporting on the results of each of their activities in order to identify cross subsidies and end them. They call for the shifting to a system of direct subsidies, as opposed to indirect ones, designed to achieve social objectives such as universal service and, consequently, banning cross-subsidies from freight to passenger traffic or from ports to rail. There is clearly a need to accelerate reform effort in these areas. The desirability of fast convergence to the acquis communautaire underlying the single market for services applies also to air transport services, which has so far remained outside Bulgaria's reform agenda. Despite ill-fated privatization of the national flag carrier, the approach to aviation policy has remained in the old mould of the exclusive route rights. Instead of moving to a new approach emphasizing open skies and unilateral liberalization, the bilateral system of air transport regulations, based on a positive list approach limiting provision of services to those that are explicitly permitted,16 has been retained. Leaving aside safety 16According to this approach, a service can be provided only if it is explicitly permitted. 24 and aviation security issues, which need to be addressed under both old, statist and new, liberal approach, the regulatory philosophy underlying bilateral aviation agreements has been to protect the national carrier from external competition. Anticipating the EU accession, the roadmap for Bulgaria should include actions taken by the European Commission to implement the 1986 Single European Act's provisions on "freedom of services." In the case of transport, they included: Deregulation of road freight in the 1990s through the removal of quotas and other barriers and allowing for cabotage, i.e., provision of transport services within the country by foreign firms; Adoption by EU member-states of common licensing rules, legislation on free market access, and deregulation of rates and fares under the Third Package of Air Transport Liberalization Measures (in force since 1997) combined with the movement towards a European Common Aviation Area (EC 2004). Restructuring of national railway companies through commercialization, separation of regulatory and operational functions, and opening of tracks to competing private operators; It seems that Bulgaria should quickly follow the path of liberalization set by the EU. It should unilaterally deregulate aviation services domestically and open up internationally within the framework of the European Common Aviation Area-model. The argument in favor of this approach is that air transport provides fuel to the expansion of tourism but also to the emerging global division of labor based on production fragmentation. Although Bulgaria is not land-locked, air transportation is a key to its future trade development. Consider the example of the United States. Despite direct access to both the Atlantic and Pacific oceans, more than 40 percent of the value of goods entering and leaving US customs territory moves by air. Airline liberalization has been a contributing factor to economic development. It is often overlooked that among the measures that created a `virtuous circle' leading to the Irish economic boom were not only public expenditure cutbacks that allowed for tax reductions but also airline deregulation. The latter "... facilitated a more than doubling of tourist numbers over the following decade." (Barry 2003, p. 909). Considering Bulgaria's endowments making it a potentially highly tourism intensive country, Bulgaria should follow the Irish path of airline deregulation but in the context of the EU accession process. The EU accession process offers a unique opportunity for Bulgaria to become part of the Single European Sky Initiative and open its skies at least to European carriers. Another area deserving a close attention is management of ports and airports. Ports and airports are crucial to the transaction costs of conducting foreign trade operations. Worldwide, it is estimated that the total economic costs of transport inefficiencies are equivalent to 3 to 5 percent of the total value of trade. Both are tempting to rent seekers, as they offer huge opportunities to put a private hand in the public pocket. The third weak area of trade facilitation relates to customs. Bulgaria's level of efficiency of customs procedures stands at 84 percent of that achieved on average by CEEC-8 and 71 percent for the EU-15. Despite progress in aligning customs procedures with those at the EU, foreign companies continue regarding customs regulations and policies as inconsistent, cumbersome, arbitrary, and unstable. The Bulgarian public seems to share this perception holding the police, judges, and customs agency in low regard, due to the perceived extent of corruption in those areas. U.S. companies complain also about excessive documentation requirements, slow processing of shipments, and corruption. 17 They cite, for instance, the requirement of the 17See US Trade Representative 2004 National Trade Estimate Report on Foreign Trade Barriers available on line at http://www.ustr.gov/assets/Document_Library/Reports_Publications/2004/2004_National_Trade_Estimate/2004_NTE_ Report/asset_upload_file172_4739.pdf 25 Customs Agency to submit invoices even for equipment transfers from corporate offices in other countries to Bulgaria as excessive and counterproductive. There have been, however, some indications of progress in customs efficiency. Customs clearance time has been significantly reduced in pilot sites under the Trade Facilitation in South Europe (TFSE) project. Between June 2004 and the average for 2002 it fell 34 percent and was the shortest among selected SEE-economies (Table 11). Moreover, the scope of random inspection of shipments has been significantly expanded. While in December 2001 all shipments were examined by customs, in June 2004 only 2.3 percent were inspected. Table 11: Reduction of Clearance Time at selected Borders Crossing Points and Inland Clearance Terminals (in minutes and percent) in 2001, 2002 and June 2004 (in minutes and percent) Pilot site 2001 2002 Jun-04 Reduction, June 04 Final target over 2002 Albania Tirana 270 102 73 28% 60 Bulgaria Plovdiv 222 90 59 34% 40 Croatia Jankomir 318 180 148 18% 40 Romania Constanta 258 180 113 37% 120 Source: TFSE Indicator Workbook (June 2004). As for other areas not covered by four indicators of trade facilitation, two observations on transport and technical standards can be made. First, the road transport has been fully privatized and both forwarding and transport services are available on competitive basis. The improvements in customs and port efficiency would further cut in the cost of these services. Second, the EU Regular Report commends Bulgaria for having had made "good progress" in aligning standards with the EU, although the process of fully aligning them to European standards has not been completed as yet. There are still Bulgarian quality standards not conforming to generally accepted international standards, albeit these have been on the decline. C. Concluding observation Both domestic and international dimension of the cost of doing business point, on the one hand, to a very significant improvement and, on the other hand, to continued weaknesses. Bulgaria ranks high in terms of overall ease of doing business. It is above Croatia, Slovenia and Romania. There are three areas where Bulgaria lags behind most of CEEC-11--starting a business, contract enforcement, and protection of investors' rights. The latter is easy to address through appropriate legislative change. Considering, however, the limited confidence of firms in the effectiveness of the rule of law and the quality of judicial services, the contract enforcement strikes one as a barrier to business activity. This has negative implications for entry of de novo firms and for domestic firms' ability to take advantage of international outsourcing. The ability to participate in international outsourcing is further hampered by Bulgaria's low level of performance in terms of efficiency of trade facilitation relative to CEEC-8 and Romania in port efficiency. Bulgaria's exceeds benchmarks only in the quality of regulations. The quality of regulations is important. But as long as the capacity to enforce them is weak, it has rather limited impact on external transaction cost. 2.4. Conclusions Bulgaria has made large strides in implementing second generation, structural reforms considerably improving the quality of governance. Three different indexes--Corruption Perception Index (CPI), World Bank-based aggregate index of governance quality, and EBRD-based index of progress in second-generation reforms--all point to a very significant progress achieved over the last three-four years. But the values of these indices still remain below the average for CEEC-8 indicating the magnitude of challenge faced by Bulgaria in improving regulatory regimes and their enforcement in order to minimize the potential for rent seeking and enhance efficiency. 26 The improvement in governance and institutions has not as yet been translated into similarly impressive progress in business climate as it affects both domestic and external transactions. As far as the domestic dimension of doing business is concerned, Bulgaria's regulatory environment is still highly burdensome for private firms especially in such areas as starting a business, contract enforcement, and protection of investors' rights. Bulgaria lags behind most of CEEC-11 in these areas, with negative impact on private sector development and wealth creation. The most binding among these relates to contract enforcement. While lowering thresholds for starting a business or even improving legal protection of investors' rights can be done almost with the strike of a pen, contract enforcement is a much more complex task. It points to deficiencies in the functioning of courts. This is a serious constraint, as without improvement in this area, such other measures, as, for instance, introduction of laws protecting investors' rights will be of little relevance. Poorly functioning courts do not assure enforcement of the law. Equally worrisome are the low levels in the values of various indicators assessing the capacity to facilitate trade--port efficiency, customs environment, regulatory environment, and Information Technology infrastructure. The new EU-8 member countries on average exhibit a much higher level of efficiency in the development of measures fostering external interaction than Bulgaria using these four indicators. The weakest areas of trade facilitation in Bulgaria relative to the levels of efficiency achieved on average by both CEEC-8 and the EU-15 are management of ports, IT infrastructure and customs. They are respectively 26 percentage points and 16 percentage points for the latter two below the average for CEEC-8. This is so despite relatively high quality of regulations. It points to weak enforcing capacity. While there is evidence pointing to recent improvements in IT infrastructure and customs, the quality of infrastructure of maritime ports and airports is not likely to improve without privatizing their management. Although domestic aspects of business environment taken together with their international dimension related to trade in goods hamper the development of private sector activity and participation of Bulgarian firms in international outsourcing, these barriers have been on the decline. At any rate, they are not perceived as excessively binding by international investors. Furthermore, the prospect of accession and the necessity of convergence to the acquis communautaire both provide insurance of significant improvement in overall business/investment climate. This, combined with already observed gains in the quality of governance, seems to explain the recent surge in FDI inflows. But large foreign owned firms are less vulnerable to regulatory burden, which appears to be at the core of business climate weaknesses in Bulgaria. They have easy resort to higher echelons of administration and are, therefore, overall better endowed to deal with potential administrative harassment. They also have significant resources to allow them navigate through various pitfalls in Bulgarian economic and judicial regime. This, however, applies neither to small foreign investors, who were the driving forces behind recent Romanian export expansion (Kaminski, Ng 2004), nor to potential domestic new firms, who are at the core of innovation and business activities of highly developed countries. Both of these groups faced with prospective regulatory burden and high costs of entry will either choose other country for their business operations or will not start them at all. This cost of foregone opportunities may be quite considerable in terms of gains in employment, competitiveness and productivity. 3. Contestability of domestic markets for goods: the increase in reverse discrimination The EU Eastern Enlargement project has led to the emergence of an economic architecture based on a network of preferential trade arrangements encompassing now two overlapping circles of thirty-six European states. The first circle, organized around the Pan-European Agreement on the Cumulation of the Rules of Origin, is a de facto single free trade area for industrial products based on diagonal cumulation of the rules of origin. Its participants include the EU-25, EFTA, Bulgaria, Romania and Turkey. The latter is tied to the EU-25 through customs union arrangements. Since January 2003 no duties have been imposed on trade in industrial products among `Pan-European economies.' The second circle has emerged as a result of the EU-sponsored Stabilization and Association process for South Eastern European (SEE-5) economies launched in 1999. It has led so far to the network of bilateral free trade agreements among SEE-5 economies as well as the members of the first circle--Bulgaria, Rumania and Turkey. This preferential arrangement is 27 based on bilateral cumulation (for explanation, see Box 2 below). Full liberalization of trade in industrial products among them will be completed in 2008. In addition, SEE-5 economies have preferential access to EU-25 but on asymmetrical base, i.e., the EU-25 does not have similar to access to their markets. Box 2. Pan-European Cumulation of Rules of Origin Rules of origin are used to determine the national/territorial provenance of goods for trade purposes. The EU has two types of such rules: non-preferential and preferential. Non-preferential rules define the origin of a good for trade statistics, import quotas and use of safeguards. Preferential rules are stricter. They are to ensure that only goods originating in one of the member countries enjoy the low tariffs or other benefits laid down in the agreement. Such rules ensure that goods from countries not party to the preferential arrangements cannot be diverted via a preferential partner to circumvent the EU tariff. Under the rules products acquire origin if they are wholly produced or if they are sufficiently processed in the preferential partner. For manufactured goods the basic criterion is a change of tariff heading at four-digit level. However, for many products alternative criteria must be observed. For example, there may be a limit on the value of non-originating materials used, often 40% (effectively a 60% added value rule), or a rule specifying a process. The concept of cumulation is an important element of preferential rules of origin. It allows inputs originating in one party to a preferential agreement to count as originating in another when those goods are used in further processing. There are various types of cumulation. Bilateral cumulation is the simplest form and applies in preferential trade between the EU and a preferential partner. Diagonal cumulation applies in preferential trade between EU and groupings of countries where the parties to an agreement share identical rules of origin. It allows materials originating in the EU or in one or more of the country groupings to count towards fulfillment of the rules of origin. Full cumulation represents an extension of diagonal cumulation. It provides for the aggregation of all processing, whether or not origin conferring, undertaken anywhere in the preferential area. The concept of pan-European cumulation introduces diagonal cumulation across a Europe- wide free trade area encompassing the EU, EEA, EFTA, the Central and Eastern European countries, the Baltic States and Turkey. Source: UK Department of Trade and Industry (http://www.dti.gov.uk/ewt/rules.htm) Bulgaria's foreign trade policy has two dimensions: that of actively pursued regional liberalization in the context of the EU Eastern Enlargement project and rather passively pursued multilateral liberalization under the umbrella of WTO agreements. Since the track of regional liberalization encompasses Bulgaria's major potential trading partners including one of the world's economic superpowers, the EU-25, the multilateral track is of secondary significance to Bulgaria's economic interests. Bulgaria is now part of one of the largest, if not the largest in the world, free trade areas for industrial products. In consequence, Bulgarian producers have duty-free access to most of its markets and, on the other hand, Bulgaria's markets for most goods are fully contestable. The deep tariff reductions under the aegis of regional integration agreements combined with relatively high MFN applied tariff rates imply high reverse discrimination and significant potential for trade diversion. High preference margins provide a strong incentive to home consumers to switch from a low-cost supplier outside the FTA to a high-cost one from FTA partner country. However, there is one factor limiting the scope for trade diversion. The extent of trade diversion hinges not only on preference margins, i.e., the differences between MFN rates and preferential tariff rates, but also on the level of pre-FTA trade with FTA partner. And Bulgaria's pre-FTA imports from the current preferential partners were high. Yet, there are strong reasons for reducing reverse discrimination through lowering MFN applied tariff rates to the levels of the EU CET (common external tariff) on industrial products. High tariff protection is always a bad trade policy. Bulgaria's MFN tariff schedule for manufactured goods is both redundant and harmful. Its redundancy stems from limited, if any at all, protection afforded to domestic producers by high and dispersed MFN tariff rates. Pan-European suppliers already have duty-free access and government resorts to tariff exemptions if these products are not readily available in pan-European free trade area for industrial products. But if they are available, high MFN tariffs merely protect FTA suppliers from MFN 28 competition. This may lead to rents going to FTA suppliers in highly concentrated sectors where major competitors are from `outside' unless, of course, the government exempts them as well from paying duties. The overall result is that domestic producers remain exposed to competition from imports and government is deprived of customs duty revenues. 3.1. The emergence of the Pan-European market for industrial products Having signed the European Association Agreement with its interim trade liberalization component in effect as of 31 December 1993. Bulgaria has become part of the EU Eastern Enlargement Project. This has implied policy commitment to bilateral liberalization carried out within a larger framework of the EU-driven regional integration. Bulgaria is part of two overlapping circles of special trading arrangements among thirty- six European states. It is part of the first circle, that is, a de facto single free trade area for industrial products including the EU-25, EFTA, Romania and Turkey. It is also part of the second circle built around the Stability Pact (Box 3). Box 3: Trade Initiative of the Stability Pact Stability Pact for Southeastern Europe, launched by the European Council, commanding body of the EU, has added a new dimension to the EU-driven European integration process. The Pact was formally adopted by western allies, countries bordering on former Yugoslavia (plus Moldova, which became a member of the Stability Pact on 28 June 2001), and international organizations in June 1999 in Cologne. Bulgaria together with Romania, Albania and former Yugoslav republics (excluding Slovenia) have been members of the Stability Pact and participated in its trade liberalization project to promote free trade areas and economic cooperation in the Balkans. In June 2001, it signed the June 2001 Memorandum of Understanding on Trade Liberalization and Facilitation, which have led to a series of bilateral FTAs with SEE-5 countries.18 The process of full liberalization of trade in industrial products among Stability Pact countries will be completed in 2008. Bulgaria has concluded bilateral agreements with SEE-5 countries within the framework set under the Trade Initiative of the Stability Pact. According to the criteria set in the Memorandum of Understanding, tariffs on at least 90 percent of industrial tariff lines and 90 percent of industrial imports have to be zeroed after the FTA goes into effect. FTAs liberalized trade in more than 98 percent of industrial products. In contrast to the Pan-European arrangements, there are exceptions to free trade. These include some industrial agricultural products,19 and motor vehicles. Furthermore, the majority of the agreements--unlike the Pan-European arrangements--resort to tariff rate quotas, which usually trigger trade deflection and lead to corruption in their allocation. Bilateral agreements within the Trade Initiative of the Stability Pact framework have also significantly lowered barriers to trade in agricultural products. The criteria adopted by the MoU call for duty-free access on only 35 percent of the agricultural Harmonized System lines. The abolition of quantitative restrictions should not be an issue in the FTAs implementation process, as their number is limited and the agreements offer various methods to deal with sectorial economic difficulties in compliance with WTO rules. Bulgaria's FTA with Croatia meets a double 90 percent trade coverage criterion, with Bosnia and Herzegovina it meets the criterion of at least 90 percent of import coverage. FTAs with other SEE economies are slightly below both criteria. The EU-driven project of deepening and widening European integration has evolved over time. First, signatories of the Association Agreement would also sign a free trade agreement with EFTA countries.20 This 18SEE-5 group includes Albania, Bosnia and Herzegovina, Croatia, Macedonia, and Serbia and Montenegro. 19These are raw fibers, natural cork, manitol and sorbitol, essential oils, albumins, gelatin, products of animal origin like skins, feathers, bones, etc. 20 Most of them--Austria, Finland and Sweden--became full-fledged EU members in 1995. 29 agreement, modeled after the trade component of the European Association Agreement, has assured the same conditions in trade of industrial goods with both regional groupings tied through European Economic Area arrangements. Bulgaria signed both agreements. Second, Bulgaria was encouraged to join the existing preferential trading arrangements among Central European candidates to the EU. Last but not least, Bulgaria became part of the Stability Pact for the Balkans and concluded a series of bilateral FTAs with its Balkan neighbors (Box 1 above). As a result, Bulgaria is part of a network of FTAs. Not all FTAs are the same and their significance varies. The most important is by far the participation in `Pan-European' single market for industrial products accounting in 2003 for 77 percent of Bulgaria's trade in industrial goods. The major defining feature of Bulgaria's trade policy emerged with the Pan-European Cumulation of Origin Agreement, which went into effect on January 1, 1997.21 The objective of this agreement-- recommended by the EU Council's decision in July 1996--was to encourage Europe-wide industrial cooperation by diagonal cumulation that would allow treating imports from parties of the agreement as local inputs (see Box 2 above). This new arrangement allowing unfettered distribution of production capacities in the territory of each signatory of the Pan-European Agreement, without worrying about meeting the rules of origin requirements, has created a very attractive environment for MNCs (Kaminski 2004).22 Bulgaria has been effectively part of this arrangement since January 1, 1999. The EU has encouraged regional trading agreements among signatories of the European Association Agreements. CEFTA--established in 1992 by former Czechoslovakia, Hungary and Poland--is a good example. It had provided a framework for bilateral trade agreements among EU candidates. In 1999, Bulgaria--which already had FTA with the Czech and Slovak Customs Union (1995) and Slovenia (1996)-- joined CEFTA, which then included aside from founding countries also Romania (since 1997) and Slovenia (1996).23 Croatia was the last country to join now defunct CEFTA in 2003. The timetable for the removal of duties on industrial was coordinated by the Pan-European Cumulation agreement. Trade arrangements among signatories the Pan-European Cumulation of Origin Agreement top all other preferential trade agreements. Firms from thirty-one signatories of this Agreement operate in a de facto single market for products. They have duty-free access to respective markets for industrial products and can use inputs from any source within the Pan-European region and count them as domestic inputs. Other than rules-of-origin considerations shape the spatial distribution of production activities within the Pan-European region. More importantly, Bulgarian producers are exposed to highly, worldwide, competitive European firms. Furthermore, they have to compete on markets of many countries and that of the EU with very low levels of protection. While FTAs of the Pan-European group have mostly focused on liberalization of trade in industrial products, they have also covered significant portions of agricultural trade. For instance, in accordance with a Double-Zero Agreement of Bulgaria with the EU export subsidies in their mutual trade have been eliminated and a number of products (poultry, meat, cheese, wine, fruits, grain) received full or significant tariff reductions combined with a schedule for increasing trade quotas. As a result of participation in the EU Eastern Enlargement Project, Bulgaria's producers have duty- free access to one of the largest markets in the world for industrial products and, on the other hand, are exposed to competition from internationally competitive suppliers of these products. However, in order to 21The Agreement has paved the way for establishment in 2002 of a single European trading bloc for industrial products, encompassing the EU-25, EFTA, Bulgaria, Romania and Turkey. 22 The countries covered by the `cumulation' framework are EU-25, Bulgaria, Iceland, Norway, Romania, Switzerland and Turkey. 23The CEFTA system had both a multilateral component that comprised commonly agreed preferences and a bilateral component not extended to all CEFTA members. Tariffs on 80 percent of all industrial products were zeroed and reduced on the remaining products upon the entry of the CEFTA agreement in force in 1997. Except for Croatia with which duty reductions are subject to a separate schedule, duties on 100% of industrial products were removed as of January 1, 2002. 30 fully capitalize on benefits offered by regional integration, a country must have liberal multilateral foreign trade regime. 3.2. Multilateral liberalization policy track: trade in goods Bulgaria lost a unique opportunity to lock-in a well-designed liberal multilateral trade regime that the WTO accession process usually offers.24 First, Bulgaria's accession negotiations took place when the attention of WTO member-states focused on then ongoing Uruguay Round of Multilateral Trade Negotiation. Furthermore, the increase in pressures to open up on acceding countries placed by WTO members began taking place after the WTO was established in 1995, i.e., around when Bulgaria already finished its accession negotiations. Second, until 1997 Bulgaria's policies were protectionist and highly unstable (World Bank 2000). The accession negotiations lasting almost a decade have failed to remove a strong anti-foreign trade bias of the foreign trade regime. Bulgaria's commitments under both the 1994 GATT and the WTO GATS were rather limited indicating unwillingness to improve conditions in access for foreign goods and services to local markets. The most visible measure of a WTO member-country's commitment to trade liberalization is the scope of bindings and the levels of bound tariff rates. Although Bulgaria bound 100 percent of its tariff rates under its terms of WTO accession, the bound rates--which are the critical commitment in the WTO--were not only set at high levels but also they were generally higher than the rates it then applied, particularly on agricultural products. The simple average bound rate was negotiated at 35.5 percent on agricultural products and 23 percent on industrial products.25 When compared to other Balkan transition economies with respect to the level of bound tariffs, Bulgaria's WTO `bounds' are rather high. For instance, Albania's WTO `bounds' are 9.4 percent on agricultural products and 6.6 percent on industrial products and Croatia's 9.4 percent and 5.5 percent, respectively.26 Bulgaria scores better vis-à-vis countries that were WTO members already before the collapse of central planning--Hungary, Poland and Romania. Romania's WTO bounds are 98.4 percent on agricultural products and 31.6 percent on industrial products,27 Hungary's were 26.9 percent and 6.9 percent, and the Czech and Slovak Customs Union had bounds of 10 percent and 4.2 percent respectively. With accession to the WTO, Hungary as well as other new members adopted the EU common external tariff and its all other WTO commitments. Comparisons with these countries have only historical relevance illustrating the strength of protectionist lobbies. Applied MFN rates are clearly much more pertinent for assessment of current conditions in market access, as bound rates are only maximum rates that a country has agreed no to exceed. Despite the annual reductions in simple average applied tariff rates on both agricultural and industrial products over 1998- 2003,28 Bulgaria has the second highest, after Moldova, simple average MFN applied tariff rate among Balkan countries. All participants of the EU Stabilization and Association process, Bulgaria's preferential trading partners, have significantly lower averages. Besides Romania, only Macedonia comes close to Bulgaria in terms of high tariff protection of both agricultural and industrial products (Table 12). 24For a discussion of the impact of WTO accession on policies of selected transition economies including Bulgaria, see Drabek and Bachetta (2004). 25The simple average bound rate on agricultural products may be higher, as the share of items with a non-ad valorem binding, not taken into account, is significant. 26Based on Protocols of Accession. See Drabek and Bachetta (2004). 27"Developing country" status under the WTO has shielded Romania from external pressures to make stronger commitments to liberalize access to domestic markets. 28A simple applied MFN tariff rate on agricultural products fell from 28.5 percent in 1998 to 24 percent in 2000 and 21.5 percent in 2003, and on industrial products from 15.3 percent in1998 to 11.1 percent in 2000 and 9.6 percent in 2003. 31 A lower level of the weighted average tariff rate than that of the simple average tariff rate indicates that tariffs effectively suppress imports of items subject to high tariff rates, as sectors with high tariff will generally have a lower weight as imports are smaller. For higher level of weighted average rates, the opposite is the case. Agricultural goods fall into the latter: high tariffs do not completely prevent their imports, while they appear to do so in the case of industrial products. There are two caveats, however. First, as indicated earlier, most of Bulgaria's imports are subject to preferential, mostly zero tariff rates. The UNCTAD database does not take into account preferential trade. In consequence, weighted tariff rates are probably much higher for all countries in Table 12. Table 12: Average Applied Tariffs and Bound Rates: Bulgaria's preferential partners as available for most recent year (in percent) Average Applied MFN Tariff Rate (%) Total Goods Agricultural Goods Industrial Goods Bound Rate of All Binding goods Reporter Year Simple Weighted Simple Weighted Simple Weighted Simple Weighted Coverage Average Average Average Average Average Average Average Average in percent Albania 2002 8.5 8.6 9.9 9.5 8.4 8.2 7.0 7.9 100 Bosnia and Herzegovina * 2001 5.4 5.1 4.2 6.8 5.4 4.6 na na Na Macedonia 2001 10.2 9.3 14.4 17.2 9.8 7.5 na na Na Moldova 2001 5.1 2.8 11.2 9.2 4.5 2.0 na na 100 Serbia & Montenegro * 2001 9.0 5.8 13.1 11.9 8.7 5.0 na na na Bulgaria 2003 10.2 9.5 19.4 21.5 9.0 8.6 24.2 20.0 100 Croatia 2002 6.0 4.7 10.9 11.3 5.6 4.2 5.8 4.8 100 Romania 2001 11.6 8.3 24.8 30.7 10.6 6.3 40.4 40.3 100 Memorandum: European Union (15) 2004 4.2 3.0 7.6 4.5 3.7 2.3 3.9 3.0 100 Turkey 2003 5.2 4.0 20.7 12.3 3.7 2.3 28.6 19.8 50 Note: Agricultural goods are based on WTO classification of HS 01-24 and industrial goods are HS 25-96. * indicates non-WTO member as of August 2003. Sources: Based on UNCTAD TRAINS database and WTO bound tariff rates from WTO IDB database. Second, since Pan-European partners have duty-free access for all their industrial products, the fact that the value of the weighted average is above that of a simple average indicates that higher tariffs on industrial products serve no well-defined protectionist purpose. They seem to be redundant. We shall return to this issue in the next section. Despite benefits that a small economy like Bulgaria can gain from opening up, its multilateral liberalization within the WTO framework has neither led to low tariffs nor extended commitments to remove barriers to entry to services sectors. MFN applied tariff rates remain relatively high by regional standards. While Bulgaria has signed WTO agreements extending either duty free treatment to selected products (e.g., Information Technology Agreement) or countries (Generalized System of Preferences for least developed countries), it has kept relatively high MFN tariff rates especially on agricultural products. One should add that Bulgaria's approach to multilateral trade liberalization does not differ from that of several other Central European countries including Hungary, Poland and Romania. Despite low levels of non-preferential trade, they have declined to lower MFN tariff rates (Kaminski 2001). 3.3. Reverse discrimination: level, scope, implications FTA, by definition implies discriminatory liberalization, that is, it artificially changes relative competitiveness of goods from countries entering the FTA. Following Jacob Viner's (1950) seminal work, two static effects of FTAs are usually distinguished--trade creation and trade diversion. The latter leads to purchases from a higher-cost preferential supplier, which results in static losses, whereas the former leads, like in the case of multilateral liberalization, to lower domestic prices thanks to increased competition from imports and closing down of some high-cost firms replaced by lower-cost imports.29 Since trade creation 29The EU's Common Agricultural Policy provides an extreme illustration of a trade diversion effect, as it increases the "regional" trade at the expense of trade with outside countries (Baldwin 1991). 32 results in a partner country's production displacing higher cost domestic production, it does not increase "regional" trade at the expense of outside countries and yields static benefits to FTA participants. The FTA has the potential for increasing welfare of both partners only insofar as trade creation exceeds trade diversion. In other words, a country can lose when it liberalizes on a discriminatory basis, if the cost of trade diversion outweighs the benefits from trade creation. Large theoretical literature on bilateral trade integration offers many rules of thumb as to when FTA is likely to result in welfare loss or gain (see Baldwin 1991, Schiff and Winters 2003, World Bank 2000). The "North-South" integration, as exemplified by the EU Eastern Enlargement project provides incentives to a `South' country to become gradually like a `North' country in terms of institutions and policies. These incentives derive not only from aspirations to accede to the EU, which calls for the convergence of domestic regimes to the acquis communautaire but also from opening of the economy to competition from imports from the EU and better access to EU markets. The latter may contribute to the increase in foreign investment inflows. But there is also potential for welfare loss even in the case of `North-South' integration. The most important and enduring ones concern MFN tariffs and the fraction of imports that come from other participants in FTA. High and dispersed MFN external tariffs, that is, high reverse discrimination and low mutual trade are conducive to trade diversion. High preference margins (or reverse discrimination) provide a strong incentive to home consumers to switch from a low-cost supplier outside the FTA to a high-cost one from a FTA partner country. The level of pre-FTA trade with an FTA partner affects the size of trade diversion effect. If initial pre-FTA imports are low, the likelihood of replacing a low-cost MFN supplier with a higher-cost FTA supplier increases. Last but not least, there is a danger of trade deflection if MFN applied tariff rates vary across countries of a free trade area. Traders then have incentive to seek to enter through a `lower tariff' country, obtain a fraudulent certificate of origin and move a shipment to a `higher tariff' country resulting in losses of customs revenue. The extent of trade deflection depends on tariff differentials and the quality of customs administration--high differentials and corrupt customs lead to high trade deflection. `Pan-European' arrangements denote broadly the trade component of the European Association Agreement effective as of 1994, Bulgaria's participation in CEFTA (Bulgaria joined in 1999 liberalizing market access for then six other CEFTA members) as well as other parties of the Pan-European Cumulation of Origin Agreement, that is, Romania and Turkey. Turkey has had customs union agreement with the EU since 1997 and applies the EU Common External Tariff on most products. Taking into account schedules of tariff reductions on industrial imports, since around the end of the 1990s most tariff rates on industrial products from pan-European countries have been zeroed. As for other FTAs--with SEE-5 countries and Israel--most industrial products under these agreements either already or soon will have duty-free access to Bulgarian markets. Bulgaria's applied MFN tariff rates on industrial products create significant potential for reverse discrimination. Data tabulated in Table, based on six-digit Harmonized System (HS) tariff lines processed for UNCTAD TRAINS database, offer insights as to its potential scope--potential, as the information is limited to imports by six-digit HS items without identifying whether these imports come from preferential partners. In other words, MFN tariff rates are not applied to all imports of six-digit HS items, yet the database provides information on all imports in 2003, from preferential and MFN partners alike. Had Bulgaria had no FTAs, 21 percent of its imports in 2003 would have been subject to zero tariff rates and 20 percent to MFN applied tariff rates exceeding or equal to 15 percent (Table 13). But this was not the case. In fact, the bulk of Bulgarian imports of industrial products were duty-free. Since duties on industrial imports from `pan-European' countries were zeroed in 2003, the MFN applied tariff rates represent preferential margins enjoyed by `pan-European' suppliers in Bulgarian markets vis-à-vis MFN ones. The `pan-European' share in Bulgarian imports of industrial products was 76 percent in 2003, that is, at least 76 percent of industrial imports that entered Bulgaria were not subject to tariffs. But the actual share of duty-free imports may be much larger. 33 Table 13: The scope for reverse discrimination in industrial imports implicit in Bulgaria's tariff schedule in 2003 Applied MF tariff rate is Number of tariff Share in total Imports (million of Share in total lines US $) industrial imports above 25 percent 18 0.4% 36 0.4% below 25 and above/equal 20 367 8.9% 633 7.4% below 20 and above/equal 15 422 10.2% 1,290 15.0% below 15 and above/equal 10 577 14.0% 1,148 13.4% below 10 and above/equal 5 1,547 37.5% 2,320 27.0% below 5 and above 0 734 17.8% 1,351 15.7% equal 0 459 11.1% 1,813 21.1% TOTAL 4,124 100% 8,591 100% Memorandum: Imports from Pan-European partners 6,530 77% Source: Derived from Bulgaria's Applied Tariff Rates by HS 6-digit Product in 2003 (average in % from UNCTAD TRAINS)) `Pan-European' suppliers and, to a lesser extent, FTA partners from Stability Pact enjoy a sizable competitive advantage over MFN and other non-preferential exporters. Consider that their preferential margin exceeds or equals 20 percent ad valorem (of the value of a shipment) for 8 percent of Bulgaria's imports of industrial products in 2003, or US$ 670 million. MFN supplier would have to have costs lower at least 20 percent than a preferential supplier in order to be able to win a contract with a Bulgarian importer. Around US$ 3.1 billion worth of Bulgarian industrial imports were protected by MFN applied tariff rates equal or exceeding 10 percent, thus offering extra advantage to often least-cost effective suppliers or allowing them to capture rents by charging higher prices. Has reverse discrimination implicit in Bulgaria's MFN tariff schedule resulted in significant trade diversion, i.e., replacing more efficient non-preferential suppliers? Without an extensive examination of previous trade patterns and costs of imports, one may only sketch an answer to this question. Note first that `Pan-European economies' accounted for three-thirds of Bulgaria's total imports, with the EU-15 contributing about half of the total. During both crisis in 1996-99 and recovery phase in 2000-02 in Bulgarian foreign trade performance imports from `pan-European' economies have been expanding. This has been in stark contrast to all other imports, which fell in 1996-99 (Annex Table 1). Since the increase in presence of `pan-European' suppliers in Bulgarian markets for industrial goods, which occurred mainly over 1997-99 (Table 14), coincided with slashing tariff rates under the `pan- European' preferential agreements, it would be tempting to point to trade diversion. But this might not be necessarily a right conclusion. Although the accompanying increase in reverse discrimination might have contributed to the increase in their share from 43 percent in 1996 to 60 percent in 1999 and 65 percent in 2003, other factors than market access considerations appear to be more relevant. The share of the EU increased even before tariff rates on industrial products were significantly reduced. So did the share of former CEFTA suppliers from 4.7 percent in 1996 to 7.1 percent in 1998, that is, even before Bulgaria joined CEFTA. Table 14: Pan-European, other FTA partners and ROW in Bulgaria's industrial imports in 1996-2003 (in million of US dollars and percent) Industrial imports (million of US dollars) Share in total industrial imports (in percent) 1996 1997 1998 1999 2000 2001 2002 2003 1996 1997 1998 1999 2000 2001 2002 2003 A. Pan-European 1,992 2,029 2,493 2,995 3,441 4,183 4,648 6,486 46 48 58 62 58 76 78 77 of which EU-15 1,624 1,672 2,050 2,467 2,676 3,369 3,704 5,004 37 39 48 51 45 61 62 59 B. Other FTA 43 41 40 32 35 35 36 44 1.0 1.0 0.9 0.7 0.6 0.6 0.6 0.5 TOTAL (A+B) 2,035 2,070 2,534 3,027 3,476 4,218 4,683 6,530 47 49 59 63 59 77 78 77 Non-preferential partners 2,335 2,192 1,766 1,810 2,413 1,274 1,286 1,936 53 51 41 37 41 23 22 23 TOTAL IMPOORTS 4,370 4,262 4,299 4,836 5,889 5,492 5,969 8,466 100 100 100 100 100 100 100 100 Note: Pan-European partners include EU-25, Romania, Turkey, and EFTA. Other FTA partners include SEE-5 and Israel. Industrial products are defined as SITC 3+5+6+7+8+27+28 in Revision 2 or Harmonized System 25 and above. Source: Derived from UN COMTRADE database as reported by Bulgaria Second and more important, there is no relationship between the shares of preferential imports in total imports of pan-European transition economies and their respective average MFN rates on manufactured 34 products. Countries with very low tariff rates had both relatively high shares of preferential imports (e.g., Latvia, Slovakia) and low shares of these imports (e.g., Estonia). So did the countries with high simple average MFN tariff rates--Hungary with the share of 71 percent in 2003 and high tariff rates and Slovenia with the share of 82 percent (Table 15). Table 15: Share of preferential partners in imports of manufactures of EU Eastern Enlargement economies in 1990- 2003 (in percent) Industrial products Reporting Index,1992 Index, 1996 Index, 1999 Index, 2003 Share Average Peak share Reporter Average MFN period 1990=100 1993=100 1994=100 2000=100 2003 1996-03 Year share Bulgaria 9.0 1996-2003 n/a n/a 64 100 82 81 2001 84 Czech R. 4.7 1993-2003 n/a n/a 97 95 79 82 1995 87 Estonia 0.1 1995-2003 n/a n/a n/a 93 68 73 1995 80 Hungary 9.5 1990-2003 125 102 81 96 71 76 1995 84 Latvia 0.8 1994-2003 n/a n/a 126 100 86 86 2001 87 Lithuania 3.1 1994-2003 n/a n/a 122 89 75 80 2001 86 Poland 9.6 1990-2003 119 97 78 103 82 81 1993 84 Romania 15.9 1990-2003 139 102 76 104 84 82 2003 84 Slovakia 4.6 1993-2003 n/a n/a 101 98 85 86 1999 87 Slovenia 10.9 1992-2003 n/a 104 101 98 82 84 2001 84 Notes: Manufactures defined as SITC 5+6+7+8-68 in SITC. Revision 2. Source: Data reported to UN COMTRADE database. Third, note also that geographical proximity to the EU-15, accounting for most of preferential imports of these economies, does not offer any clues as the share of preferential imports. In fact, dispersion in the shares of preferential partners in manufactured imports of pan-European transition economies is very low, with the coefficient of variation of 0.08 for 2003 and 0.05 for the average shares in 1996-2003. Last but not least, over 2000-03, despite zeroing of the remaining preferential rates on industrial products, the share of pan-European imports in manufactured imports of CEEC-10 mostly contracted. This share only slightly increased in Poland's imports and Romania's imports. Except for Romania, the share of pan-European preferential partners in manufactured imports was in 2003 below the levels recorded in earlier years (Table 16). Table 16: Imports of manufactured goods by major trading partners in 1996 and 1999-2003 (in million of US dollars and percent) Import Value ($ million) Average Index Import Share (%) Growth 2003 Partner 1996 1999 2000 2001 2002 2003 1996- 2000 2000= 1996 1999 2000 2001 2002 2003 99 -03 100 World 2,429 3,497 3,837 4,714 5,331 7,440 9.5 18.0 194 100 100 100 100 100 100 European Union (15) 1,573 2,368 2,565 3,257 3,573 4,835 10.8 17.2 188 64.7 67.7 66.8 69.1 67.0 65.0 CEEC (8) 141 237 287 312 351 509 13.9 15.4 177 5.8 6.8 7.5 6.6 6.6 6.8 EU-25 1,714 2,605 2,852 3,568 3,925 5,343 11.0 17.0 187 70.6 74.5 74.3 75.7 73.6 71.8 Romania 33 21 40 50 65 85 -10.9 20.8 213 1.3 0.6 1.0 1.1 1.2 1.1 Turkey 75 131 170 229 343 557 15.0 34.5 328 3.1 3.7 4.4 4.9 6.4 7.5 EFTA (3) 79 75 81 84 88 111 -1.1 8.4 137 3.2 2.1 2.1 1.8 1.7 1.5 TOTAL PAN-EURO 1,900 2,832 3,143 3,933 4,420 6,097 10.5 18.0 194 78.2 81.0 81.9 83.4 82.9 81.9 Israel 7 8 6 16 16 14 4.3 23.0 233 0.3 0.2 0.2 0.3 0.3 0.2 SEE (5) 17 9 12 14 16 24 -14.3 20.3 200 0.7 0.3 0.3 0.3 0.3 0.3 OTHER FTA 23 17 18 29 32 39 -7.8 21.3 217 1.0 0.5 0.5 0.6 0.6 0.5 TOTAL PREF. 1,940 2,856 3,171 3,970 4,465 6,153 10.2 18.0 194 79.9 81.7 82.6 84.2 83.8 82.7 ROW (NON-PREF) 489 641 667 743 866 1,287 7.0 17.9 193 20.1 18.3 17.4 15.8 16.2 17.3 Note: Manufacturing products are defined as SITC 5+6+7+8-68 in Revision 2. Source: Derived from UN COMTRADE database as reported by Bulgaria. Developments in geographical patterns of trade of other CEEC-9 economies provide an extra support to the conclusion that the growing reverse discrimination in imports of manufactures has not resulted in significant diversion in Bulgaria's trade. They all point to the size of the pan-European free trade for industrial products rather than the level of MFN tariff rates on manufactured goods as explaining the absence of any significant trade diversion effect (Kaminski 2005). Consider the following: First, geographical 35 reorientation took place during the initial stages of transition from central planning rather than during the period of accelerated cuts in preferential tariff rates on manufactured goods in the second half of the 1990s and early 2000's. While no data are available for all CEEC-10 economies, it is rather unlikely that they diverged from the pattern observed in Hungarian, Polish and Romanian imports where the largest increases occurred over 1990-92 (Table 16 above). Why were the shares of manufactured imports from preferential sources in total imports of manufactures immune to the degree of reverse discrimination? There are two explanations. First, MFN suppliers may lower export price to stay competitive in Bulgarian markets.30 Second, authorities may resort to tariff exemptions. This practice appears to be quite widespread. As a result, not all importers have to pay high duties on their MFN supplies. For instance, goods imported by firms operating in Bulgarian Free Trade Zones and economically depressed areas, were exempt from duties. So were some imports of capital equipment. This has weakened the trade diversion effect usually triggered by high preferential margins. Both practices are distortionary, however, in spite of their apparent little impact on direction of foreign trade. They have negative impact on Bulgaria's competitiveness in global markets, as it affects their ability to access the best inputs available worldwide and unnecessarily increased administrative micromanagement of trade flows. High MFN tariff rates are not only redundant but are also a bad trade policy. Shortcomings of regional liberalization unless accompanied by simultaneous liberalization of MFN tariff rates are twofold: First, reverse discrimination often suppresses FDI inflows from firms located in non-preferential countries unless they are able to outsource to suppliers from Bulgaria's preferential patterns. Second, high MFN tariff rate raise prices of imports and thus may undermine export competitiveness of Bulgarian exporters. While in many cases indispensable imports are available from preferential partners, this is not always the case, as highly specialized machinery may be only available in, for instance, Japan or the United States. High tariffs put Bulgarian firms in competitive disadvantage vis-à-vis firms operating in countries with lower tariff rates. In fact, there is no reason to maintain high MFN applied tariff rates on industrial imports, as MFN tariff rates offer neither protection to domestic producers nor generate significant customs revenues. High tariff protection in Bulgaria is redundant, as more than 80 percent of manufactures and slightly less than 80 percent of industrial products originate in `pan-European' countries and therefore do not face any tariff barriers in access to Bulgarian markets. In the absence of extensive regional trade arrangements, tariff protectionism can be defended (or explained) in terms of infant protection argument, government revenue needs or political economy reasons related to political pressures from powerful interest groups. If, however, the same sectors are open to duty-free competition from firms from such highly developed economies as the EU, not to mention, the internationally competitive imports from other recent EU entrants, none of the above applies unless a protected product is not manufactured in any of FTA countries. This is hardly a plausible occurrence, as a `pan-European' free trade area for industrial products is an economic superpower. Under these circumstances, maintaining high MFN tariffs amounts to protecting FTA suppliers from MFN competition and has a negligible effect on competition from imports faced by domestic producers. 3.4. "Shadow" customs union with the EU on industrial products Faced with the raising welfare costs associated with high preferential margins, the best policy option for a small country is usually free trade. Considering, however, Bulgaria's involvement in the EU accession process, this option is not acceptable, as it might trigger demands for compensation from Bulgaria's MFN trading partners once Bulgaria becomes a member of the EU. Under these circumstances, the best policy option is that of adopting the EU's MFN CET on industrial products.31 30Chang and Winters (2002) provide empirical evidence showing that firms lowered their export prices offered to consumers from MERCOSUR countries in order to compete with firms from this regional trading bloc. 31Although upon accession Bulgaria will adopt the CET, there remains still around two years before this happens. 36 While lower MFN applied tariff rates would not shift the overall geographic pattern of Bulgarian imports toward MFN suppliers, they would facilitate trade and increase competitive pressures on Bulgaria's preferential suppliers. Since Bulgaria is on the accession path to the EU, MFN tariff rates on industrial products should be harmonized with those in the EU CET. This would amount to a `shadow' customs union with the EU. This will be turned into a customs union arrangement once Bulgaria accedes to the EU. The cost of such a move is negligible, if any. This is clearly a win-win case. How does the EU CET on industrial products compare with Bulgaria's MFN tariff schedule? Bulgaria's tariff schedule affords overall significantly higher levels of tariff protection to both agricultural and industrial producers. Simple averages for both agricultural and industrial products are more than twice as high as in the EU (Table 17). Although an overall lower dispersion characterizes Bulgaria's tariff schedule for both agricultural and industrial products, the coefficient of variation is lower for Bulgaria's MFN tariff schedule. Second, Bulgaria has 809 six-digit HS tariff lines within which there are rates exceeding the international peak of 15 percent, as compared to the EU's 23 tariff lines. Table 17: Overview of Bulgaria's and European Union's MFN Applied Tariff Schedules in 2003 Simple Coefficient Minimum Maximum Number of Number Imports Value Bulgaria Average of variation Rate Rate International Peaks of Lines (million of US$) Industrial products 9.0 0.75 0 26.8 809 4,126 8,592 EU-15 Industrial products 3.7 0.93 0 22.0 23 4,464 750,172 Note: International peaks are defined as tariff rates exceeding 15 percent ad valorem. The format of the UNCTAD TRAINS file identifies maximum and minimum tariff rates within six-digit lines that represent averages for HS nine-digit original tariff schedules. Source: derived from the EU"s and Bulgaria's Applied Tariff Rates by HS 6-digit products in 2003 (average for 6-digit tariff line in % from UNCTAD TRAINS) Bulgaria's tariffs are not only higher but also offer protection to different products than in the EU (Annex Table 4). Yet, although at the six-digit HS level there are some differences reflecting dissimilar `protectionist' preferences, they are not extreme. Both Bulgaria's and the EU's MFN applied tariff schedules offer larger tariff protection to unskilled labor intensive products. Paradoxically, Bulgarian successful exports to EU markets--clothing and footwear--are also highly protected domestically. So are they in the EU, albeit levels of tariff protection are lower. For instance, clothing is protected in both, but tariff rates on clothing imports in Bulgaria (24 percent) are twice as high as in the EU (12 percent).32 On the other hand, there is one unequivocal difference. Bulgaria provides high tariff protection to low processed industrial goods such as glass and glassware, paper products, fertilizers, modified starches, carpets and other textile floor coverings, whereas the EU protects relatively more such products as trucks, bicycles, road tractors and video cameras. In all, however, the two schedules for industrial products are not dissimilar. The value of correlation coefficient for maximum tariff rates in six-digit HS aggregates is 69 percent and for simple average rates (average also for six-digit HS lines) it raises to 72 percent. Bulgarian MFN applied tariff rates on industrial products are lower for only 42 six-digit HS items out of 4,124 items and the same for 857 items (simple average for each aggregated group from a nine-digit original breakdown) or 939 items (maximum for each group). Furthermore, for 2,277 HS 6-digit items, or 55 percent of all industrial lines, the difference between respective tariff rates does not exceed five percent and for 85 percent is smaller than 10 percent (Table 18). The amounts of imports subject to MFN tariff rates exceeding those in the EU also do not strike one as huge. Consider the following: First, 27 percent of Bulgaria's imports in 2003 fell into tariff lines that would be subject to the same MFN tariff rates; Second, almost 40 percent of imports of industrial products 32Until the termination of the WTO Agreement on Textiles and Clothing as of January 1, 2005, quotas gave an extra protection to EU producers of these products. 37 enter Bulgarian markets either at the same or lower rates than in the EU. Altogether 86 percent of Bulgaria's imports are subject to tariffs that are up to 10 percent higher than respective EU duties on these products. Table 18: Breakdown of Bulgaria's tariff protection on industrial products: differences in applied MFN tariff rates between Bulgaria and the EU The difference between BG and EU Number of Share in Imports Share in applied tariff rates tariff lines total (in million of US dollars) total above 25 percent 1 0.0% 1 0.0% above 20 below/equal 25 16 0.4% 12 0.1% above 15 below/equal 20 110 2.7% 274 3.2% above 10 below/equal 15 412 10.0% 955 11.1% above 5 below/equal 10 1,304 31.6% 2,075 24.2% above 0 below/equal 5 1,382 33.5% 2,192 25.5% equal zero 857 20.8% 2,344 27.3% below zero (EU tariffs higher) 42 1.0% 740 8.6% TOTAL 4,124 100% 8,591 100.0% Source: derived from the EU"s and Bulgaria's Applied Tariff Rates by HS 6-digit products in 2003 (average for 6-digit tariff line in % from UNCTAD TRAINS). Yet, although 14 percent of Bulgaria's industrial imports are subject to MFN applied tariff rates significantly higher than in the EU, the crux of the matter remains that not all of these imports are actually subject to higher rates. Industrial products originating in `pan-European' countries and, increasingly, in SEE- 5 countries are not subject to duties. These have two important intertwined implications: First, preferential `pan-European' and SEE-5 suppliers face lower levels of competition from MFN imports in EU markets than in Bulgarian markets for industrial products subject to rates higher than in the EU. Depending on the structure of industry, this may raise the price of imports and create rents for preferential exporters. For instance, if three firms, one from the EU and other two from non-preferential countries dominate the world market, a fifteen-percent tariff-related preferential margin gives a EU firm an opportunity to charge up to 15 percent more for the same product in Bulgaria than in the EU. Second, wide tariff margins may suppress or make investment projects non-competitive. For instance, if specialized products or machinery available in `pan-European' free trade for industrial products are of inferior quality than those available outside and are subject to higher MFN tariff rates than in other partner-countries, the restructuring or other capital project may never take off. While these foregone opportunities are only subject to speculations, their cost--as anecdotal evidence suggests--may be considerable. Harmonization of Bulgaria's applied MFN tariff rates with the EU CET on industrial products entails several benefits and few, if any costs. Major benefits stem from rationalization of tariff structure and, by the same, token minimization of distortions inevitably produced by diversity in tariffs. EU tariffs on industrial products are significantly lower, less diversified and tariff peaks are lower. Bulgarian users of imports would benefit from lower tariffs and their bargaining position vis-à-vis preferential suppliers would increase once preferential margins are reduced. Furthermore, with the fall in tariffs, transaction costs associated with exports and corresponding tariff rebates on imported inputs would also considerably decrease, as lower amounts of capital would be tied up. In a similar vein, importers of MFN products would be less inclined to actively seek duty exemptions on MFN imports. But would it lead to a significant deterioration in Bulgaria's government revenues, as a result of the declined in custom revenues from duties? Although based on `back-envelop' estimates, the answer is unambiguously negative. Consider first that the loss in customs revenue due to FTAs has already taken place and the contribution of revenues from customs duties is rather small. Had MFN rates been applied to all imports, tariff revenue would have been around US$ 800 million higher in 2003, that is, US$ 1 billion rather than US$ 143 million. Their share in total revenue and grants varied between 1.59 percent and 1.95 percent over 2001-04 (IMF 2004a). In terms of the GDP, it was well below 1 percent. The duty collection rate, i.e., 38 the ratio of customs duty revenue (US$ 143 million) to the value of imports (US$ 10.9 billion), was already very low at 1.3 percent in 2003. Second, the bulk of customs duty revenue appears to have come from duties levied on agricultural rather than industrial imports, as 77 percent of the latter were industrial imports from Pan-European partners with a duty-free access. In other words, only around US$ 2 billion worth of industrial imports (out of the total of US$ 8.5 billion) were subject to MFN tariff treatment. Furthermore, not all MFN imports of industrial products were subject to duties exceeding zero. Around 27 percent of industrial imports in terms of tariff lines in 2003 were subject to zero duty rates no matter their preferential or non-preferential origin (see Table 18 above). Third, the scope of adjustment in tariff treatment is lower than the differences in MFN rates would suggest. Note that around half (53 percent in 2003) of Bulgaria's imports are subject to the same or maximum five percent higher MFN tariff rate as in the EU. The adoption of the EU's CET on industrial products would lower Bulgaria's weighted-average tariff on industrial products from 8.6 percent to 4.2 percent in 2003.33 Box 4: Harmonization of tariffs and WTO rules From the point of view of the WTO rules, accession is simply the shift from a free trade area (FTA already fully operational under Europe Agreements) to customs union (CU). The only new (in relation to FTA already in place) relevant article that will apply when Romania either joins the EU or sign a CU agreement is Article XXIV:6. Article XXIV:6 requires WTO members joining a CU and consequently seeking to increase bound rates of duty to enter into negotiations--under Article XXVIII (Modification of Schedules)--on compensatory adjustment, netting out reductions in duties on the same tariff lines made by other members of the customs union. If such reductions are regarded as insufficient compensation, the CU will either reduce duties on other tariff lines, or provide compensation. If no agreement is reached on compensatory adjustment within a reasonable period of time, the CU may modify or withdraw the concessions and affected WTO members are free to retaliate. A reasonable period of time, as agreed during the Uruguay Round, is to be no more than 10 years. This has the following implication: the legal situation vis-à-vis the WTO will remain the same as long as Romania will not insist upon accession that the EU raise its bound rates to higher Romanian ones. Other issues are unlikely to appear as legal impediments. The reference is occasionally made to Article XXIV. 5 (a), which requires that duties and other regulations of commerce after the establishment be either less restrictive or no more restrictive than prior to the establishment of a CU. If the applied rates are after accession on the whole higher than those applied before accession, the enlarged EU would be in breach of its WTO obligations and any WTO member will be entitled to ask for compensations. This is a misinterpretation. First, as noted above, the trigger for compensation adjustment is the increase in bound rates. Second, an assessment under this Article is based on an overall assessment of weighted average tariff rates and customs duties collected. The WTO Secretariat is responsible for conducting an assessment, which has little practical relevance. Note that since 1948, out of more than 60 FTAs and preferential trade agreements that have been reviewed under Article XXIV provisions, only four agreements (the list includes neither EU CU nor NAFTA) were declared fully compatible with Article XXIV requirements. However, a working group has not censured a single agreement as being incompatible with GATT-WTO rules. This suggests that the costs are insignificant. Although the budget will loose customs revenue from duties, the amounts involved are relatively small as already most potentially tariff revenue-generating imports come from preferential rather than MFN suppliers. In spite of significant differences in their respective structures of tariff protection as captured among others by the value of a correlation coefficient of 0.51, the cost of adjustment will be low, simply because most producers have already been exposed to 33EU tariffs weighted by Bulgaria's industrial imports. 39 competition. Furthermore, this move should not complicate Bulgaria's entry into the EU in terms of possible problems with the WTO rules (Box 4 above). This diagnosis leads to the following recommendation: Bulgaria should align its MFN applied tariff rates to those in the European Union's CET. This would reflect the reality of its participation in the Pan- European free trade area for industrial products, trade arrangements under the Stability Parts and pending accession to the European Union. While this would neither result in a surge of imports from MFN suppliers or in a significant fall in customs revenue, this would increase competitive pressure on preferential suppliers and thereby might lower prices of imported products. Furthermore, it would reduce shopping and lobbying administrative officials by importers to obtain tariff rebates or duty exemptions. 4. Conclusions and policy implications Bulgaria has made large strides, especially since 2001, in raising the overall quality of governance, reducing corruption and introducing institutions and policies supporting competitive markets. It already has most institutions and policies in place for successful plugging-in and taking advantage of opportunities offered by integration into pan-European and global markets. The prospect of accession and the necessity of convergence to the acquis communautaire provide insurance of significant improvement in overall business/investment climate. This, combined with already observed gains in the quality of governance, seems to explain the recent surge in FDI inflows. But there are still weaknesses related to regulatory environment and the quality of backbone services. Both prevent domestic firms from participation in a new global division of labor based on outsourcing and production fragmentation. They also raise the cost of conducting business domestically and internationally discouraging entry of new firms and small foreign investors. They explain why Bulgarian firms remain largely outside of EU-based networks of production and distribution. The improvement in governance and institutions has not as yet been translated into similarly impressive progress in business climate as it affects both domestic and external transactions. As far as the domestic dimension of doing business is concerned, Bulgaria's regulatory environment is still highly burdensome for private firms especially in such areas as starting a business, contract enforcement, and protection of investors' rights. These are major areas where Bulgaria lags behind most of CEEC-11. The most binding among these relates to contract enforcement. While protection of investors' rights and starting a business appear to be easy to address through appropriate legislative and regulatory change, this is not the cast of contract enforcement. Lowering thresholds for starting a business or even improving legal protection of investors' rights can be done almost with the strike of a pen, contract enforcement is a much more complex task. It points to deficiencies in the functioning of courts. This is a serious constraint, as without improvement in this area, such other measures, as, for instance, introduction of laws protecting investors' rights will be of little relevance. Poorly functioning courts do not assure enforcement of the law. While improvements in the functioning of courts take time, this should not suggest that nothing could be done in the short run. To the contrary, simpler and more transparent regulations usually go a long way to reduce the use of courts. Hence, deregulation should top the government's reform effort. The continued review of regulations and their simplifications should be the first step in ongoing effort to improve the rule of law. But this should be a component of a continuous process rather than a series of ad hoc regulatory actions involving the government, business, and the civil society at large. International good practice suggests that the best way is to establish central regulatory coordination and management capacities (with credible mechanisms inside the government for managing and coordinating complex reforms), allot responsibility to higher levels of government and untie the units responsible for deregulatory management from regulatory missions and linking them to political authorities or existing centers of oversight authority (such as centers of government and finance and trade ministries). The ability to participate in international outsourcing and production fragmentation is further hampered by Bulgaria's low level of performance in terms of efficiency of trade facilitation relative to both CEEC-8 and Romania in port efficiency. Bulgaria's exceeds benchmarks only in the quality of regulations. The quality of regulations is important. But as long as the capacity to enforce them is weak, it has rather limited impact on external transaction cost. While there is evidence pointing to recent improvements in IT 40 infrastructure and customs, the quality of infrastructure of maritime ports and airports is not likely to improve without privatizing their management. The government will have to address inefficiencies related to ports/airports management and transport before accession to the EU. But, since the `EU-template' in these areas is clearly business friendly and efficiency enhancing, this effort should be accelerated. The development of a strategy and its quick implementation should top the current reform efforts. The strategy should address all aspects of transportation even going beyond minimum requirements set by the acquis communautaire following the 1986 Single European Act's provisions on "freedom of services." Among others, this would call for · Taking steps to separate public providers of port, rail and other transport services and reporting on the results of each of their activities in order to identify cross subsidies and end them. · Separation of operational and commercial functions (the landlord port), privatization of port services, concessioning of large container and bulk terminals to private operators. · Unilateral deregulation of aviation services domestically and open up internationally at least to European carriers even before becoming part of a Single European Sky, which now includes EU-members as well as Iceland, Norway and Switzerland. This would include liberalization of rail transport, landing rights and access to airports (allocation of slots). · Opening should include granting on unilateral basis the right to cabotage to EU-based carriers. It would follow suite of tourism intensive countries, e.g., Dubai and some Caribbean countries albeit in its European dimension anticipating the measures that Bulgaria will have to take upon accession to the EU. Another area that hampers Bulgaria's participation in new forms of global division of labor relates to its MFN tariff policy and implicit high levels of reverse discrimination. The redundancy of keeping high MFN tariff rates on industrial products stems from the following: high MFN tariff rates serve no protectionist purpose; they yield limited customs revenues; in some circumstances, they offer preferential suppliers opportunity for rents at the expense of Bulgarian consumers of imports and limit bargaining position of Bulgarian importers; and create demand for duty-drawback or tariff rebate schemes for MFN imports. For these reasons, although Bulgaria will have to adopt the EU common external tariff upon accession, the government should enter into a `customs union' with the EU limited to MFN tariff rates for industrial products. The government should change the level of applied MFN tariff rates to bring them in line with the EU CET observing the following rules. This would involve the following: · If a Bulgarian MFN applied tariff rate is higher than in the EU, lower it to the EU rates; · If it is lower, leave as is. (This would assure conforming with Bulgaria's WTO commitments, i.e., in an unlikely case that the Bulgarian WTO tariff binding is below the current EU MFN applied tariff, raising the MFN tariff rate might trigger disputes with WTO members). The combination of measures focusing on both domestic and external dimensions of Bulgaria's business environment as well as on its multilateral foreign trade policy would contribute to the establishment of an economic regime opening new opportunities to Bulgarian businesses to take advantage of enormous opportunities offered by Bulgaria's participation in the European integration project. 41 References: Baldwin, Richard E. 1991. Towards an Integrated Europe, Centre for Economic Policy Research, London. Barry, Frank. 2003. "Economic Integration and Convergence Processes in the EU Cohesion Countries." Journal of Common Market Studies, Vol. 41, No. 5 (897-921). Bevan, Alan. A., Saul Estrin. 2004. "The determinants of foreign direct investment into European Transition Economies." Journal of Comparative Economics, 32 (775-787). Djankov, Simeon. 2005. "Doing Business in Bulgaria." Background note to Bulgaria's CEM, March. Campos, Nauro F. and Yuko Kinoshita. 2003. "Why Does FDI Go Where It Goes? New Evidence from the Transition Economies." IMF Working Paper WP/03/228, International Monetary Fund, Washington DC, November. Chang, W., Winters L.A. (2002). "How Regional Blocs Affect Excluded Countries: The Price Effects of MERCOSUR." American Economic Review, Vol. 92. No. 4 (617-48). Drabek, Zdenek and Marc Bacchetta. 2004. "Tracing the Effects of WTO Accession on Policy-making in Sovereign States: Preliminary Experience from the Recent Experience of Transition Countries." The World Economy, Vol. 27, Issue 7, July (pp. 1083--1125). EU 2004: Communication from the Commission of the European Communities, A Community aviation policy towards its neighbors, Com (2004) 74 Final, Brussels, February 9. EC 2002: Benchmarking the Administration of Business Start-ups, Final Report. Prepared by the Centre for Strategy and Evaluation Services (CSES), European Commission, Enterprise Directorate General, Brussels, January. FIAS 2004: Bulgaria: Assessment of Administrative Procedures for Doing Business in Bulgaria, Foreign Investment Advisory Service, IFC and the World Bank, Washington DC, June. Fisher, Stanley, Ratna Sahay and Carlos A. Vegh. 1998. "How far is Eastern Europe from Brussels?" IMF Working Paper, WP/98/53, Washington DC. Global Competitiveness Report 2003/2004 Hodge, James. 2002. "Liberalization of Trade in Services in Developing Countries," in B. Hoekman, A. Mattoo, and Ph. English, eds., Development, Trade, and the WTO. A Handbook, The World Bank, Washington, D.C. Hoekman, Bernard and Aaditya Mattoo. 2000. "Services, Economic Development and the Next Round of Negotiations on Services." Journal of International Development. 12 (283--296), p. 292. Hunya, Gabor. 2002. FDI in South-Eastern Europe in the early 2000s. The Vienna Institute for International Economic Studies, Vienna, July. IMF 2004a: "Bulgaria: Request for Stand-By Arrangement--Staff Report." IMF Country Report No. 04/267, International Monetary Fund, Washington DC, August. IMF 2004: "Bulgaria: Briefing Paper for the First Review Under the Stand-By Agreement," International Monetary Fund, Washington DC, November. Janicki, Hubert P., Phandra V. Wunnava. 2004."Determinants of foreign direct investment: empirical evidence from EU accession candidates," Applied Economics, 36 (505-509). Kaminski, Bartlomiej. 2005a. "Ranking Transition Countries in Terms of Real Ease of Doing Business." Mimeo, World Bank. Kaminski, Bartlomiej. 2005b. "EU Enlargement and Trade Diversion: The Case of Romania." Unpublished manuscript. 42 Kaminski, Bartlomiej. 2004. "How to Minimize Distortions Due to Bilateral Tariff Liberalizations within Framework of EU-Driven Integration? The Case of Romania." Background paper for Romania: Country Economic Memorandum, manuscript, ECA, The World Bank, January. Kaminski, Bartlomiej. 2001. "How Accession to the European Union Has Affected External Trade and Foreign Direct Investment in Central European Economies," Policy Research Working Paper 2578. The World Bank, Development Research Group-Trade, Washington DC, April. Kaufmann, Daniel, Aart Kraay and Massimo Mastruzzi. 2003. "Governance Matters III: Governance Indicators for 1996-2002". World Bank Policy Research Department Working Paper, Washington DC. Schiff, Maurice and L. Alan Winters. 2003. Regional Integration and Development. Oxford University Press and The World Bank, Washington DC. TTFSE 2003: Progress Report 2003. http://www.seerecon.org/ttfse/ Viener, Jacob. 1950. The Customs Union Issue. Carnegie Endowment for International Peace, New York. Wilson, John S., Catherine Mann, and Tsunehiro Otsuki (2004) "Assessing the Potential Benefit of Trade Facilitation: A Global Perspective" World Bank Working Paper 3224. The World Bank, Washington DC. Wilson, John S. Xubei Luo, and Harry G. Broadman (2004). "Trade and Transport Facilitation: European Accession and Capacity Building Priorities." Draft paper prepared for the workshop on "Transport Costs and their Impact on International Trade," European Conference of Ministers of Transport and OECD Research Center, Paris, October 21-22. WB 2005a: Doing Business in 2005. Removing Obstacles to Growth. The World Bank and Oxford University Press, Washington DC, 2005. WB 2005b: Judicial Systems in Transition Economies. Assessing the Past, Looking to the Future. The World Bank, Washington DC, February 16, 2005. WB 2004: Kyrgyz Republic: Country Economic Memorandum; An Integrated Strategy for Growth and Trade, The World Bank, Washington DC, June. WB 2001: Bulgaria: The Dual Challenge of Transition and Accession, The World Bank, Washington DC. WB 2000: Trading Blocs, The World Bank, Washington D.C. World Economic Forum 2001: Global Competitiveness Report 2001-2002. World Economic Forum: Geneva. World Economic Forum 2003: Global Competitiveness Report 2003- 2004. World Economic Forum: Geneva. 43 Annex Tables: Annex Table 1: Developments in `preferential' and total imports in 1996 and 1999-2003 (in million of US dollars and percent) Import Value ($ million) Average growth rate Import Share (%) Partner 1996 1999 2000 2001 2002 2003 1996-99 1999-03 1996 1999 2000 2001 2002 2003 EU-25 1,946 2,907 3,205 3,974 4,440 6,026 10.5 17.1 38 54 49 55 56 55 of which: EU-15 1,780 2,629 2,864 3,592 4,015 5,406 10.2 17.2 35 49 44 49 50 50 CEEC-8 166 278 341 383 424 620 13.8 16.1 3.3 5.1 5.2 5.3 5.3 5.7 Romania 70 71 232 173 164 262 12.2 15.0 1.4 1.3 3.6 2.4 2.0 2.4 Turkey 96 164 215 274 394 667 28.5 25.0 1.9 3.0 3.3 3.8 4.9 6.1 EFTA 86 79 89 90 109 135 -2.0 11.2 1.7 1.5 1.4 1.2 1.4 1.2 TOTAL PAN-EURO 2,199 3,221 3,741 4,511 5,106 7,091 10.0 17.3 43 60 58 62 64 65 Israel 9 18 11 19 19 20 -2.0 11.2 0.2 0.3 0.2 0.3 0.2 0.2 SEE-5 105 46 55 46 61 97 -1.4 18.2 2.1 0.8 0.8 0.6 0.8 0.9 TOTAL OTHER FTA 114 64 66 64 80 117 -13.5 15.4 2.2 1.2 1.0 0.9 1.0 1.1 ROW 2,762 2,124 2,698 2,702 2,801 3,693 -6.3 8.2 54 39 41 37 35 34 TOTAL IMPORTS 5,074 5,410 6,505 7,278 7,987 10,901 1.6 13.8 100 100 100 100 100 100 Note: Pan-European partners include EU-25, Romania, Turkey, and EFTA. Other FTA partners include SEE-5 and Israel. Source: Derived from UN COMTRADE database as reported by Bulgaria. Annex Table 2: Pan-European, other FTA partners and ROW in Bulgaria's industrial imports in 1996-2003 (in million of US dollars and percent) Industrial imports (million of US dollars) Share in total industrial imports (in percent) 1996 1997 1998 1999 2000 2001 2002 2003 1996 1997 1998 1999 2000 2001 2002 2003 A. Pan-European 1,992 2,029 2,493 2,995 3,441 4,183 4,648 6,486 46 48 58 62 58 76 78 77 of which EU-15 1,624 1,672 2,050 2,467 2,676 3,369 3,704 5,004 37 39 48 51 45 61 62 59 B. Other FTA 43 41 40 32 35 35 36 44 1.0 1.0 0.9 0.7 0.6 0.6 0.6 0.5 TOTAL (A+B) 2,035 2,070 2,534 3,027 3,476 4,218 4,683 6,530 47 49 59 63 59 77 78 77 Non-preferential partners 2,335 2,192 1,766 1,810 2,413 1,274 1,286 1,936 53 51 41 37 41 23 22 23 TOTAL IMPOORTS 4,370 4,262 4,299 4,836 5,889 5,492 5,969 8,466 100 100 100 100 100 100 100 100 Note: Pan-European partners include EU-25, Romania, Turkey, and EFTA. Other FTA partners include SEE-5 and Israel. Industrial products are defined as SITC 3+5+6+7+8+27+28 in Revision 2 or Harmonized System 25 and above. Source: Derived from UN COMTRADE database as reported by Bulgaria Annex Table 3: Share of industrial products in imports and share of manufactures in industrial imports from main trading areas in 1996-2003 (in percent) Share of industrial imports in imports from 1996 1997 1998 1999 2000 2001 2002 2003 A. Pan-European partners 91 88 91 93 92 93 91 91 non EU-15 93 75 90 91 91 89 87 86 B. Other FTA partners 38 42 41 50 54 55 44 38 TOTAL PRERENTIAL (A+B) 88 86 89 92 91 92 90 91 NON-PREFERENTIAL PARTNERS 19 25 25 26 23 27 29 35 WORLD 86 86 86 89 91 75 75 78 Share of manufactures in industrial imports from A. Pan-European partners 95 94 93 95 91 94 95 94 non EU-15 89 86 83 88 76 83 90 85 B. Other FTA partners 54 60 54 53 51 83 90 87 TOTAL PRERENTIAL (A+B) 95 94 92 94 91 94 95 94 NON-PREFERENTIAL PARTNERS 22 24 37 36 28 59 68 67 WORLD 56 58 69 72 65 86 89 88 Note: Manufacturing products are defined as SITC 5+6+7+8-68 in Revision 2. Source: Derived from UN COMTRADE database as reported by Bulgaria 44 Annex Table 4: Tariff peaks in the EU CET and Bulgaria's tariff schedule: industrial products European Union (15) Simple Min. Max Bulgaria Simple Min. Max. HS-6 Tariff schedule 2004 Aver. Rate Rate HS-6 Tariff schedule 2003 Aver. Rate Rate 870421 g.v.w. not exceeding 5 tonnes 13.5 3.5 22 640219 Other 26.8 26.8 26.8 g.v.w. exceeding 5 tonnes but 870422 not exceeding 20 ton 15.83 3.5 22 640291 Covering the ankle 26.8 26.8 26.8 870423 g.v.w. exceeding 20 tonnes 15.83 3.5 22 640299 Other 26.8 26.8 26.8 Sports footwear; tennis shoes, 870431 g.v.w. not exceeding 5 tonnes 13.5 3.5 22 640411 basketball shoes, 26.8 26.8 26.8 870432 g.v.w. exceeding 5 tonnes 15.83 3.5 22 640419 Other 26.8 26.8 26.8 Of synthetic fibres, Chassis fitted with engines, for measuring per single 870600 the motor vehicle 9.88 4.5 19 611511 yarn les 25.5 25.5 25.5 Of synthetic fibres, Footwear incorporating a measuring per single 640110 protective metal toecap 17 17 17 611512 yarn 67á 25.5 25.5 25.5 Of other textile 640191 Covering the knee 17 17 17 611519 materials 25.5 25.5 25.5 Covering the ankle but not Tableware and 640192 covering the knee 17 17 17 691110 kitchenware 25.5 25.5 25.5 640199 Other 17 17 17 691190 Other 25.5 25.5 25.5 (1996-) Ski boots, cross 640212 country ski footwear 17 17 17 701310 Of glass ceramics 25.3 25.3 25.3 640219 Other footwear 17 17 17 701321 Of lead crystal 25.3 25.3 25.3 Footwear with upper straps or 640220 thongs assembled to 17 17 17 701329 Other 25.3 25.3 25.3 Other footwear, incorporating a 640230 protective metal t 17 17 17 701331 Of lead crystal 25.3 25.3 25.3 Of glass having a linear coefficient of 640291 Covering the ankle 17 17 17 701332 expansion 25.3 25.3 25.3 640299 Other 17 17 17 701339 Other 25.3 25.3 25.3 Sports footwear; tennis shoes, 640411 basketball shoes, g 17 17 17 701391 Of lead crystal 25.3 25.3 25.3 640419 Other 17 17 17 701399 Other 25.3 25.3 25.3 - Footwear with outer soles of Urea, whether or not 640420 leather or composit 17 17 17 310210 in aqueous solution 25.0 25.0 25 Ammonium nitrate, whether or not in 640590 - Other 10.5 4 17 310230 aqueous soluti 25.0 25.0 25 Dextrins and other 870120 Road tractors for semi-trailers 16 16 16 350510 modified starches 23.6 20.7 25 With compression-ignition 870210 internal combustion pist 13 10 16 350520 Glues 25.0 25.0 25 Of wool or fine animal 870290 Other 12.4 10 16 570110 hair 24.9 24.5 25 Bicycles and other cycles Of wool or fine animal 871200 (including delivery tric 15 15 15 570310 hair 25.0 25.0 25 Of nylon or other 852110 Magnetic tape-type 7.33 0 14 570320 polyamides 25.0 25.0 25 Of other manmade 852190 Other 14 14 14 570330 textile materials 25.0 25.0 25 (1996-) Still image video Of other textile 852540 cameras and other video 5.95 0 14 570390 materials 25.0 25.0 25 Tiles, cubes and (1996-) Pocketsize radio similar articles, 852712 cassetteplayers 12 10 14 690710 whether or not 25.0 25.0 25 (1996-) Other apparatus 852713 combined with sound record 12 10 14 690790 Other 25.0 25.0 25 Combined with sound Ceramic tableware, recording or reproducing kitchenware, other 852721 apparatus 12.67 10 14 691200 household ar 25.0 25.0 25 Source: derived from the EU"s and Bulgaria's Applied Tariff Rates by HS 6-digit products in 2003 (average for 6-digit tariff line in % from UNCTAD TRAINS)