Eu Preferences Revisites

1. INTRODUCTION
The European Union (EU) has a long history of non-reciprocal preferential trade agreements with developing countries. 1 For almost 40 years, since the signature of

Yaoundé I in 1963, the EU has constantly enlarged the scope of these agreements and today over 140 countries benefit from some form of preferential access to the EU market. The main schemes through which the EU hasput in place this policy have been the Lomé Convention and the Generalized System of Preferences (GSP). The former grants preferential access to European markets to a group of 78 African, Caribbean and Pacific states (ACP) and the latter to over 140 low and medium income countries spread all over the world. 2 However, in the past ten years, the EU has begun to consider a change in its tradestrategy with ACP countries. Among the reasons for this change is the failure to integrate ACP countries in the world economy. The extension of this argument is that the Lomé Convention has failed in its purpose of boosting ACP economies through trade. Consequently, it would be reasonable to conclude that non-reciprocal preferential access has been ineffective in promoting ACP exports. Given that theLomé Convention is the most comprehensive preferential trade agreement granted by the EU, other systems like the GSP should work no better. This, in turn raises the question of whether preferential trade agreements have been (and continue to be) beneficial for developing countries or not. In spite of this change in policy with ACP countries, the EU has continued to show its belief in preferentialtreatment and in 2000 launched the Everything But Arms (EBA) Initiative, which grants quota and duty free access to the European market to all Least Developed Countries (LDCs). In this paper we intend to analyze whether trade preferences have actually been beneficial to developing countries (and are a good policy option) by studying the last 30 years of exports to the EU from a large sample oftrading partners, including both developing and developed economies. Our analysis builds on the work of Nilsson (2002)
In strict legal terms, the body which has the legal capacity to sign agreements is the European Communities. However, since the European Union remains the more widely known term we use the latter throughout the article. 2 This includes most ACP countries which in practice do not usethis scheme because it is less comprehensive than Lomé
1

1

who, using a gravity model, found that both the Lomé Convention and the GSP had a positive impact on the beneficiary countries. Interestingly enough, there are very few studies evaluating the exporting capacity of ACP, GSP and MED countries using a gravity setting. In the last decades the gravity model has become one of the mostpopular devices to analyze trade flows. 3 In its basic form, the gravity equation predicts trade flows as a function of the size of the trade partners and the distance between them. Within this framework, practically any factor can be controlled for to examine its impact on trade flows. The original gravity equation, proposed by Tinbergen (1962), did not have a strong theoretical foundation. Inaddition, most extensions to the model have used ad hoc arguments to explain its validity, which makes every result potentially disputable on economic grounds. However many authors have since tried to compensate for this shortcoming by providing a theoretical foundation to the model. The best known works are those of Anderson (1979), Bergstrand (1985, 1989) and Deardorff (1995, 1998). Our intention isto apply the gravity framework to the analysis of these flows. Consequently, we conduct a number of gravity equations incorporating different degrees of complexity. Starting from a simple cross-section model replicating Nilsson’s we then construct a number of panel data gravity models which take into account the new developments coming both from economic theory and from econometrics. Namely, by...