Full Integration Versus Partial Trade Liberalization: Comparing The Economic Performance of The NMS and MPC’s

FEM31-15 | December 2007

Title

« Full Integration Versus Partial Trade Liberalization: Comparing The Economic Performance of The NMS and MPC’s »

By

Jan Michalek, Warsaw University, Poland

Contributeurs

Alfred Tovias (Hebrew University) Andrzej Cieslik (Warsaw University) Lukasz Goczek (Warsaw University) Jan Hagemejer (Warsaw University and National Bank of Poland) Agnieszka Pugacewicz (Warsaw University) Victoria Roshal (Hebrew University)

Note :

This document has been produced with the financial assistance of the European Union within the context of the FEMISE program. The contents of this document are the sole responsibility of the authors and can under no circumstances be regarded as reflecting the position of the European Union.

Summary :


In the project we compared the depth of trade liberalization and integration with the EU between the MPCs (Mediterranean Partner Countries)[1] and NMS (New Member States)[2]. We analyzed trade implications of preferential arrangements between MPCs and EU in the framework of EUROMED and between Central and East European countries in the frameworks of BAFTA (Baltic Free Trade Area) and CEFTA (Central European Free Trade Area) and so called Europe Agreements. This analysis demonstrated that Europe Agreements and two sub-regional agreements BAFTA and CEFTA have been efficient in promoting bilateral trade among European states. We find that the Europe Agreements as well as BAFTA and CEFTA significantly contributed to increase in both bilateral exports and imports of the CEE countries. In contrast to the NMS the impact of New Association Agreements concluded with the MPCs had been much more limited. We find that while the new Agreements increased significantly imports of the MPCs countries from the EU, they had no impact on their exports to the EU which can be attributed to the asymmetry in trade liberalization between the EU and the MPCs. In particular, liberalization of the EU imports from the MPCs was a gradual process that extended over the last thirty years and there was not much to liberalize in the 1990s while liberalization of the MPCs imports from the EU took place much faster and its scope was much bigger.

Differences in growth rates may stem from differences in trading arrangements of these two regions with the EU-15 and institutional factors within these regions. Despite the similar openness ratio, MPCs maintained (with the exception of Israel) considerably higher levels of trade protection in comparison to NMS. Higher levels of protection, causing production and consumption distortions, can slow down the rate of growth of economies.

The degree of trade openness, given the widespread pressures of lobbies in a majority of countries, depends mainly on the governments’ commitment to act in the interest of all consumers and on the scope external pressure for liberalization. NMS, by concluding Europe Agreements, were forced to remove tariffs and non-tariff barriers in its trade with the EU. The scope of trade liberalization foreseen in preferential agreements between MPC and the EU is much lower. Thus, without strong external pressure MPC could and did maintain considerably higher levels of trade protection, in comparison to NMS.

In order to verify this hypothesis an attempt has been made to apply the Grossman-Helpman (G-H) model, explaining the interplay between sectoral interest groups and a government to the Polish economy as a representative of NMS. On the other hand Israel was treated as an example of MPC, being able to liberalize significantly its trade policy. We demonstrated that rent seeking of the lobbies had some, although very limited, impact on the scope of non-preferential trade liberalization in Poland. On the other hand, the performance of the G-H model for Israel has not been satisfactory, because of lack of the relevant statistical data allowing to strictly follow the original modeling framework. The model estimated in a simplified way did not show a significant impact of lobbies on tariff schedules of Israel, however, this conclusion is not robust due to poor statistical properties of the model. The verification of the model for other MPCs economies appeared to be even more difficult, due to non existence of relevant sectoral data.

Other important conditions, which are crucial for reaching real trade liberalization and convergence to the advanced EU members, are the business climate and the level of corruption. Our results suggest that corruption increases with a higher number of procedures involved in license granting, enforcing contracts and starting new businesses. According to our study, the recent accession to the EU required from the NMS to put in place certain measures that enforce transparency and competition. The gradual law approximation to the EU acquis communautaire boosts performance of the poorer NMS through the facilitation of trade and general economic activity. In this study we compared corruption levels in the MPCs and NMS, that acceded the EU in 2004. The results suggest that if the MPCs reduced their level of business regulation to that of the NMS, we would observe a predicted fall in corruption (with the exception of Israel and Jordan). This fall is quite large, especially for Syria and Algeria. This in turn, could lead to real further trade liberalization, reducing production and consumption distortions in MPCs.

1. Assessing the impact of Europe Agreements and EUROMED FTAs

1.1. Europe Agreements, BAFTA and CEFTA

In the 1990s the European Union played an active role in sponsoring trade liberalization in Central and Eastern European (CEEC) countries. The CEECs during the communist rule remained isolated from the rest of the world for almost fifty years. The so-called Europe Agreements concluded with the CEE countries were intended to support their economic reforms and prepare them for eventual membership in the EU. These efforts culminated in two subsequent waves of enlargement to the East that took place in 2004 and 2007.

The ultimate goal of joining the EU has been the major factor shaping foreign trade policies in the CEE countries throughout the 1990s. The EU concluded bilateral association agreements with the majority of the CEECs in the first half of the 1990s. These agreements aimed at establishing a hub-and-spoke free trade area covering industrial products and granting some preferences to agricultural goods between the CEECs and the EU over a maximum period of ten years.

In contrast to a typical FTA the Europe Agreements implied asymmetric trade liberalization between the EU and the CEECs with more rapid liberalization by the EU. The trade components of the Europe Agreements overshadowed and extended the Generalized System of Preference status granted by the EU to most CEECs in the early 1990s. By January 1, 1997 the EU eliminated practically all tariffs on imports from the CEECs with the exception of agricultural and “sensitive” products.

Although trade parts of the Europe Agreements with some CEE countries entered into force on different dates ranging from 1992 (former Czechoslovakia, Hungary and Poland) to 1997 (Slovenia), schedules of elimination of tariffs and non-tariff barriers on industrial products had one important element in common. They all had to be completed by the target date of January 1, 2002. The liberalization of trade in agricultural goods between the EU and the CEECs, however, did not take place until the two waves of enlargement of the EU to the East in 2004 and 2007. Only since then the CEECs have been able to participate fully in the EU Single Market.

In addition to trade liberalization with the West the CEE countries liberalized trade among themselves creating a matrix of bilateral and sub regional free trade agreements. The most important of these was the Central European Free Trade Area established by former Czechoslovakia, Hungary and Poland. The CEFTA agreement was signed on December 21, 1992 and entered into force on March 1, 1993. The initial CEFTA agreements eliminated tariffs on approximately 40 percent of industrial goods. Trade in industrial goods and some agricultural products was further liberalized through a series of additional protocols, mostly signed in 1994 and 1995.

By 1996 almost 80 percent of CEFTA trade in industrial products were free of tariffs. By 1999 tariffs were abolished on almost all industrial products except a minor list of “sensitive” products. The CEFTA membership gradually expanded overtime to include Slovenia (1996), Romania (1997), Bulgaria (1991) and Croatia (2003). The CEFTA agreement was supposed to include also three newly independent Baltic States: Estonia, Latvia and Lithuania that emerged from the former Soviet Union after its collapse in 1991. However, these three countries – in about the same time when CEFTA was built – created their own Baltic Free Trade Area (BAFTA).

In contrast to CEFTA, BAFTA did not enlarge its membership but the coverage of the agreement was increased over time at a faster pace than in the CEFTA member states. In particular, by January 1, 1997 BAFTA included not only industrial but also agricultural and fish products. In this way BAFTA became the first free trade area in the region that provided for completely liberalized trade in these politically sensitive areas.

Consequently, the significant differences in the pace and the coverage of trade liberalization between the BAFTA and the CEFTA member states did not allow creating a single free trade area that would embrace all the CEE countries before their accession to the EU. Instead, a number of bilateral trade agreements between the BAFTA and the CEFTA countries was signed that complemented sub-regional trade liberalization in Central and Eastern Europe.

We study here the trade effects of the Europe Agreements for bilateral imports and exports of the CEE countries using the generalized gravity equation that can be derived from neoclassical and new trade theory models that assume incomplete specialization in production. In our study we control for the effects of other both plurilateral and bilateral free trade agreements concluded by the CEE countries among themselves as well as with countries located outside the region. We analyze the impact of the association agreements for both exports and imports separately.

The generalized estimating equation that encompasses particular estimating equations is derived from various theoretical models. Apart of standard variables, related to distance, GDP, capital/labour ratio or regional trade agreements we also used dummies indicating whether countries share a common border, a common language and/or a same colonizer.

We find that the Europe Agreements as well as BAFTA and CEFTA significantly contributed to increase in both bilateral exports and imports of the CEEC countries. Moreover, the estimates obtained for BAFTA were of higher magnitude than those obtained for CEFTA. This is in line with our initial expectations given the fact that the BAFTA agreement included not only industrial but also agricultural and marine products.

1.2. EUROMED, bilateral and Agadir Agreements

Unlike trade liberalization with the Central and East European countries trade liberalization with the MPCs was a gradual process that lasted over three decades and is in fact not completed at the time of writing. Moreover, unlike the CEE countries that radically liberalized their trade with the EU in the 1990s, exports from the MPCs have enjoyed preferential treatment by the EU for many years. The first generation Euro-Mediterranean Association Agreements with the selected MPCs that provided free access for their manufactures exports to the EU countries were concluded already at the end of 1960s and early 1970’s.

In 1972 the European Commission launched the Global Mediterranean Policy (GMP) that was aimed at providing trade concessions to most MPCs. This policy resulted in a series of so-called Cooperation Agreements that were concluded between 1973 and 1980 with most MPCs. In particular, these agreements extended earlier preferences for MPCs exports of agricultural products both in terms of coverage and the margins of preference. However, at the same time these exports were subject to protectionist measures imposed by the European Common Agricultural Policy (CAP).

Following the southern enlargement the EU concluded a new series of agreements with the MPCs called Adaptation Agreements. As a result of this renovated policy by the end of 1993 all tariffs on exports from the MPCs were eliminated, however, non-tariff barriers (NTBs) to trade related to the CAP were still in force.

The final round of trade liberalization between the EU and the MENA countries was initiated in 1995 at the Barcelona conference that set the ambitious goal of creating the Euro-Mediterranean Free Trade Area (EUROMED) by the target date of 2010. The creation of the EUROMED is to be achieved by means of the new generation of the Euro-Mediterranean Association Agreements between the EU and the MPCs accompanied by free trade agreements between the MPCs themselves. In contrast to earlier, mostly unilateral, trade liberalization (with exception of Cyprus, Turkey, Malta and Israel), the new Association Agreements provide for the implementation of bilateral free trade between the EU and the MPCs. The EUROMED foresees free trade in manufactured goods and progressive liberalization of trade in agricultural products.

In addition to “vertical” bilateral trade liberalization with the EU the MPCs are committed to implement “horizontal” trade liberalization among themselves. However, compared to the Euro-Mediterranean Association Agreements, trade liberalization between particular MPCs is still far less advanced. Given the lack of significant progress in liberalizing trade at the regional and sub-regional levels in the MPCs region some countries located both in the Middle East and North Africa decided to liberalize their trade on bilateral basis. This led to the establishment of the grid of bilateral trade agreements by the end of the 1990s that prepared the ground for the Agadir Declaration signed in 2001 by the representatives of four MPCs: Egypt, Jordan, Morocco and Tunisia. The Agadir Declaration included a commitment to the establishment of a free trade area (FTA). The Agadir Agreement was signed in 2004 and was initially foreseen to take effect in 2005. However, it encountered delays in entering into force and eventually came into force only in 2007. At this stage it is too early to evaluate its effectiveness. Nevertheless, it was possible to evaluate empirically at least the effectiveness of those bilateral trade agreements concluded among the MPCs in the late 1990s that later laid the foundations for the Agadir Agreement.

Similar to the case of the NMS we study here the trade effects of the new EU Association Agreements for bilateral imports and exports of the MPCs using the generalized gravity equation that can be derived from neoclassical and new trade theory models that assume incomplete specialization in production. In our study we control for the effects of other both plurilateral and bilateral free trade agreements concluded by the MPCs among themselves as well as with countries located outside the region. We analyze the impact of the association agreements for both exports and imports separately.

As in the case with CEE countries, we derive the generalized estimating equation based on various theoretical models. In addition to the standard variables, related to distance, GDP, capital/labour ratio or regional trade agreements we also used dummies indicating whether countries share a common border, a common language (Arabic or Turkish) and/or a same colonizer.

In all the cases our estimations demonstrate that the EU-Association Agreements significantly contributed to the increase in bilateral imports of the MPCs from EU members. The evidence obtained for bilateral trade agreements concluded in the late 1990s between the members of the Agadir group is, however, mixed. While the simple OLS estimates suggest that all bilateral agreements concluded between the Agadir group member states significantly increased their bilateral imports, this evidence is not robust when panel data estimation techniques are employed. In contrast to bilateral imports the new generation EU Association Agreements do not seem to contribute positively to the expansion of exports from the MPCs to the EU. This means that the EU countries will be the main beneficiaries of these agreements, at least in the short run, as the EU markets for industrial products have been open to the MPCs since the 1960s and the 1970s, while the markets for agricultural products still remain relatively closed.

We also find that not all the MPCs were able to benefit equally from the EU Association Agreements. In the case of bilateral imports we obtain positive and statistically significant coefficients for the Agreements only for Israel, Jordan, Morocco and Tunisia, and in the case of bilateral exports only for Tunisia. The results obtained for bilateral trade agreements concluded between the future members of the Agadir group were not significant.

2. Why trade liberalization is difficult and frequently delayed: testing Grossman-Helpman model

2.1. Testing Grossman-Helpman model for Poland

The degree of trade openness, given the widespread pressures of lobbies in a majority of countries, depends mainly on the governments’ commitment to act in the interest of all consumers and on the scope of external pressure for liberalization. All NMS, by concluding Europe Agreements, and Central Free Trade Area (CEFTA), were committed to remove tariffs and non-tariff barriers in their trade with the EU.

Polish trade policy in 1990’s, during the economic and political transition, was quite similar to that of other Central and East European (CEE) countries. Poland, at the beginning of 1990’s had some degree of freedom in its tariff policy towards non-European countries. At that time Poland’s tariffs were not subject to the discipline of the GATT and the government had freedom to manoeuvre in shaping its tariff structure. Poland had no legal constraint in the form of a “bound” tariff schedule, although it was a GATT member since 1967.

The preferential tariff liberalization started in 1991. In that year, Poland signed the Europe Agreement (EA) with the European Communities (EC). The commercial part of the EA came into force by 1992. The EC and Poland started FTA for non-agricultural products since March 1994 over a maximum period of ten years. The FTA was not applied to agricultural products. In the case of European Union’s (EU) imports this liberalization has taken five years and was completed by the end of 1997. The timetable of tariff liberalization of Polish imports was extended in time.

Poland signed a similar free trade agreement with EFTA countries. The agreement covered mainly trade in non-agricultural products. EFTA members eliminated most import duties in 1993. Poland gradually liberalized its tariffs and quantitative restrictions on EFTA imports by 1999 (except for steel, petroleum products, and automobiles).

Poland, along with the Czech Republic, Hungary and the Slovak Republic, established also CEFTA (Central European Free Trade Area) in 1992. Afterwards Slovenia 1996, Romania and Bulgaria joined the CEFTA. The CEFTA Agreement established a free­trade area by 2001. CEFTA covered all goods, except for a few agricultural products. Thus, by the end of 1990’s almost all Polish duties on non-agricultural imports from European countries have been eliminated. The share of all these European countries exceeded 65% of Poland’s total imports.

The non-preferential (MFN, conventional) liberalization of Poland’s trade policy towards non-European countries started in 1995. The country took part in the GATT Uruguay Round as the only state having the formal status of a developed country without any “bound” customs duties. After submitting its initial offer on tariff concessions Poland had bilateral negotiations with several countries. Poland’s main commitments on trade in goods in the Uruguay Round were to bind 94% of its duties and to reduce tariffs by 38% on industrial products and by 36% on agricultural goods over six years. The simple average bound MFN Polish tariff rate for non-agricultural products was reduced from 16.73 to 9.89 per cent.

The tariff structure was determined, almost from scratch, in the early 1990s by governmental decision. It seems that the interest groups probably did not have very strong influence on the process. The organizations of producers (chambers) were just being established. However, the trade unions were quite powerful. It is possible to show some anecdotal evidence that the tariff changes were influenced by lobbies’ pressure. Afterwards the level of import duties was gradually reduced over the next years. The scope of reductions was quite impressive in the case of preferential duties.

In order to verify the possible impact of different lobbies we applied the Grossman-Helpman (G-H) “Protection for sale” model to analyze tariff policy. The governmental policy –in this model -is determined by elected politicians. They simultaneously consider the consumer welfare of the electorate and contributions of lobbies, representing various sectors of the economy. Thus, in the model, various lobbies in organized industries provide contributions to the government in return for influencing the tariff schedules.

Our empirical implementation (similar to that of Maggi and Goldberg for United States) is dealing with Polish trade policy in the late 1990s. We have used the instrumental variable approach to estimate the model, taking into account possible endogeneity of the regressors (import penetration). In the absence of direct measures of industry contributions, we have used similar variables to those used in literature, as a proxy for industry organization. The Herfindahl concentration index seems to be the best proxy for the industry organization. It means that more concentrated industries with fewer companies have better chances to coordinate their actions and are more effective in lobbying. The model was estimated for years 1996-1999.

Our results are broadly in line with the predictions of the Grossman-Helpman (G-H) model. Most of our regressions support the theory and we find support for the (minor) significance of lobbies in the formation of trade policy in Poland. The G-H model seems to work well in the case of MFN tariffs. The importance of the lobbies is, however, significantly lower than in the case of the United States. According to the model the Polish government attached very high weight to social welfare, paying almost no attention to pressures from lobbies. The model finds only very weak support using preferential data. This result seems plausible since in late 1990’s Poland had already no freedom in its preferential policy, due to tariff reductions, in line with provisions of the Europe Agreement.

Thus, we can conclude that the Polish government paid almost no attention to organized protectionist lobbies, and due to external pressure from the EU, EFTA and WTO liberalized its tariff policy very rapidly in the second half on 1990’s. It seems that a similar situation prevailed in other Central and East European, future NMS. All of them signed Europe Agreements, EFTA and CEFTA agreements and reduced their conventional tariffs in line with Uruguay Round commitments.

2.2. Testing Grossman-Helpman model for Israel

Since 1991 Israel has moved slowly but steadily towards a very open trade regime. The average applied MFN tariff was only 8.9% in 2005. Some 46.4% of the tariff lines carry duties between zero (excluded) and 15% (included). MFN tariffs on agricultural products remain high, with an average tariff of 32.9%, and rates varying considerably among product groups. Israel has bound rates on just half of its tariff lines. The bound rates are often above the applied MFN rates, giving Israel the possibility to unilaterally raise its applied tariffs. However the significance of MFN tariff schedules is limited. The bulk of Israel’s imports is conducted within the framework of free-trade agreements.

The oldest NTB in Israel is the kosher certificate requirement. Because of the power of the Rabbinate, any overseas exporter who aims to reach maximal market access and share must comply with kosher requirements. Another idiosyncratic NTB originates from its centralistic structure regarding imports. In many instances, only one firm is allowed to import a certain label.

Protectionism via the use of government procurement has plagued the Israeli economy since Independence in 1948, as public expenditure is unusually high for an OECD-type economy, with the Army and the Ministry of Defence playing a central role. While Israel has been for years party to the plurilateral Agreement on Government Procurement (GPA), it has always invoked developing-country status, allowing it to implement offset arrangements. Foreign governments, particularly the one of the US, have been complaining for years, about the lack of transparency in Israeli practices.

Invoking the status of developing country for applying Uruguay Round resolutions has not been limited to the domain of government procurement, but to other highly relevant issues in the case of Israel, such as TRIPS, where Israel was given until 2000 to change its law on intellectual property.

Since July 2003, Israel has lifted its general prohibition on imports from WTO Members that had no diplomatic relations with it or prohibited imports from Israel. However, a licensing requirement remains in place for some other countries that prohibit imports from Israel.

The agricultural sector is still highly protected, especially in comparison to industrial goods. The only exception to agricultural self-sufficiency is feed grains such as wheat and soybeans, where Israel is a net importer. MFN tariffs on agricultural products did actually rise more than 2.5 times from 1993 to 1999 with the adoption by the GATT’s principle of tariffication. The huge tariffs protect all local producers, including those kibbutzim and Arab meat producers which are allowed to supply non-kosher meat. Because there are still tariff quotas on some product groups, there is some scope for lobbying the Ministry of Agriculture and Rural Development as well as the one of Industry, Trade, and Labour, which are the ones in charge of the administration of those quotas.

Regarding textiles and wearing apparel, Israel has accepted since at least a decade that the survival of its firms under Israeli ownership and management implies that they have to be “de-localized” to neighbouring countries such as Jordan and Egypt. The process has taken place in an orderly way and is now almost completed.

In our project we also tried to estimate the “Protection for sale” model for Israel. But, in this case we faced some serious problems resulting from limited availability of statistical data. The core estimation equation of the Grossman-Helpman model involves the inverse import penetration ratio as an explanatory variable that is clearly endogenous. The original modeling framework uses instrumental variables to overcome that problem. However, it was impossible to get a robust estimate of predicted import penetration ratios for Israel. The problems stemmed mainly from the lack of sectoral data on capital stock required to predict import penetration in a Heckscher-Ohlin, factor-intensity fashion. Thus, it was impossible to estimate the G-H model correctly using the original framework. Nevertheless, an attempt has been made to estimate the model neglecting the endogeneity problem. Estimates are in line with the theory only for the case of the export intensity ratio (i.e. share of exports in domestic production) used as industry organization variable for both applied MFN tariffs and weighted average of applied duties (for preferential and non preferential imports).

Poor performance of the G-H model may result from the fact that the Israeli government paid relatively little attention to organized lobbies or that they were poorly organized. Our results suggest that only the export oriented industries in Israel, managed – to a very limited extent – to exert some influence on the government and receive some tariff protection in the framework of preferential agreements. This statement has to be treated with extreme caution, since these estimates are not robust and are subject to statistical problems. Moreover, this policy was probably not a real obstacle in Israeli trade liberalization, since export oriented industries are, almost by definition, not seriously threatened by import competition.

Perhaps, these results explain partially why Israeli governments managed to follow fairly liberal trade policies. Unfortunately, sectoral data, necessary to estimate G-H model, appeared to be unavailable for other more protectionist Mediterranean Partner countries.

3. Business environment: the causes and role of corruption in MPCs and NMS: ways for improvement

The shape of tariff policy can be influenced by lobbies and vested interests groups, as shown in the previous research. But the real access to a given domestic market by foreign and domestic suppliers depends on business environment as well. Bureaucratic delays, costs, and the number of procedures to carry out business activities can severely limit the access to a given market. Thus, the unfavourable business environment can acutely restrict real liberalization of the economy and lead to distortions in production and consumption.

The proliferation of unnecessary regulation can benefit narrow, but powerful groups inside the state administration with privileged access to legislature. With such actions, top bureaucratic officials may organize the business environment as to enrich themselves. In such cases, even though the government aims to realize public wants, regulatory policies can be implemented in wasteful and incompetent ways, which results in the over-regulation of the businesses. Seeing this, lower-level officials may take advantage of this situation and may seek to delay bureaucratic decisions in order to extract bribes. Thus, they can use their public power to extract bribes from those who need the authorizations or permits. As a result, an excessive number of regulations may result in widespread corruption.

There is a growing statistical evidence that corruption slows down economic growth and foreign direct investment. Corruption in candidate countries had been one of the EU’s major concerns time after time, when the European Commission begun publishing its annual progress reports on candidates in 1997. The grounds of that anxiety were simple: – the European legal system works under the assumption that law will be implemented, controlled, and enforced by the public administration and judiciary of the member states. Corruption endangers the implementation and execution of rules or makes their adoption merely formal. An implication of the Accession Partnerships is that that the candidate countries must fight against corruption. Moreover, the EU-Commission has adopted a requirement of an anti-corruption framework for all candidates. What is more important, the EU was a major force behind de-regulation in the NMS. Therefore, the EU accession process has had a major impact on corruption in candidate states. This can be confirmed by comparing corruption scores of non-EU post communist countries and New Member States. In our opinion, the recent accession to the EU required from the NMS to put in place measures that enforce transparency and competition.

In MPCs, there has been a tendency for reforms and economic liberalization programs to lag. This underdevelopment of bureaucratic powers and regulation has created opportunities for rent seeking associated with corruption. The MPCs had no such external pressure as the NMS had, applied on them by the European Commission. Hence, the position of interest groups within the MPC still allows rent-seeking behaviour through artificial barriers to entry to internal markets and discourage competition.

The main goal of the project was to analyze the causes of corruption and ways of improvement. Several hypotheses were tested here. Some of them appear quite general,

e.g. that (i) corruption is lower in higher economically developed countries and that (ii) stable democratic institutions and corruption are expected to be positively correlated. Two other hypotheses were more important for practical reasons for MPCs; i.e. that (iii) corruption is associated with the number of bureaucratic procedures and (iv) that it rises with bureaucratic delays.

The testing started with the two exogenous variables explaining corruption: the logarithm of GDP per capita and the level of democracy proxied by the political rights variable. This pair of control variables proved to have correct positive sign and be statistically significant. Hence, the first two hypotheses have been positively verified.

Then, we dealt with licenses variables. This topic tracks the procedures, time, and costs to build a warehouse, including obtaining necessary licenses and permits, completing required notifications and inspections, and obtaining utility connections. As expected, it came out in the investigation that longer bureaucratic delays and higher number of procedures were a significant factor in the prevalence of corruption.

On the other hand, official costs of dealing with licenses came out significant only in some of the specifications, with low levels of confidence. The positive signs in all specifications indicate that official costs of bureaucratic procedures do not play a major role in corruption, since businesses prefer predictable official payments to unpredictable informal ones so they substitute one by the other. This means that official incentive payments may be one way to achieve a less corrupt and speedier bureaucracy.

To accomplish a successful reduction in the level of corruption, liberalization and improvements in working of the bureaucracy will be required in most MPCs. Decreasing the number of procedures, streamlining of the bureaucracy and other ways to limit time spent waiting for its decisions may significantly reduce corruption, even if it occurs at a higher official cost of conducting administrative procedures. Given the long history of corruption in the region, the proposed regulatory measures would probably not be sufficient to eradicate it to the level of new members of the European Union, at least not in the short term. No doubt, it may take years or even decades to bring corruption down.

The possible improvement in the level of corruption in MPCs has been analyzed as well. First, we compared the current level of each of the variables used in the regression for every MPCs country with an average level for the eight NMS from Central Europe.

Then we analyzed the difference between the average level for the eight NMS countries and the level for each MPCs country of each of the variables used in the regression. We have estimated the ‘predicted improvement’ in the level of corruption in MPCs by multiplying the differences in the level of regulation by the coefficients obtained in the regression (only significant variables were used).

The result is that if the MPCs reduced their level of business regulation to that of the NMS, we would observe a predicted fall in corruption. This fall is quite large, especially for Syria and Algeria. The improvement would move these countries almost to the level of corruption observed currently in Turkey and the improvement in the corruption level in Turkey would move it to the level of corruption observed in the Czech Republic.

However, Israel and Tunisia have a lower level of regulation than the 8 New Member States’ average. As a result, a change in the level of regulation to the NMS 8 level would mean an effective increase in regulation. Therefore, this would theoretically increase corruption instead of decreasing it. This hypothetical result seems plausible too, though it is very important to remember, that post-communist countries suffer from legacies of the now obsolete system of central planning. By definition, central planning involved vast and substantial direct governmental intervention in economic decision-making. Therefore, these countries have a very high level of regulation even though liberalization is still undergoing.

This is not in any way connected to the bureaucratic rigidities often ridiculed by opponents of the European Union. In fact, the level of regulation and the time spent dealing with the bureaucracy is much lower in the EU-15 countries than in the 8 NMS by almost an order of significance. Furthermore, as discussed above, the European Commission was a major force behind deregulation in many areas of economic activity in the NMS.

[1] The group MPCs consists of: Algeria, Cyprus, Egypt, Israel, Jordan, Lebanon, Malta, Morocco, the Palestinian Authority, Syria, Tunisia and Turkey

[2] The group of NMS consists of: Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia and Cyprus and Malta. In our project we analyze only first eight countries