THE IMPACT OF TRADE LIBERALIZATION ON DEVELOPING ECONOMIES: A COMPARATIVE STUDY Nevzat Evrim Önal, Lerzan Özkale 1. INTRODUCTION The aim of this study is to examine and evaluate the impacts of liberalisation reforms on the economies of the developing countries of Latin America and Africa. For this purpose, a panel data set for 31 countries will be analysed to define the effects of liberalisation on economic growth, external trade balance and income inequality. Liberalisation policies have been introduced by Bretton-Woods agencies as an alternative to the import substitution policies after the oil crisis of the late 1970s and early 1980s caused the debt problems of the third world countries to deteriorate into insolvency. It was then claimed that liberalisation reforms would provide the needed stimulation to solve the general stagnation problem in these economies by attracting foreign direct investment and raising the external trade balance. The chief reason to this has been the extraordinary success of the East Asian economies after trade liberalisation reforms. However, the results were not as optimistic as they were expected to be. While the servicing of the external debt has been steadier after the implementation of these reforms, the wave of liberalisation in the developing world did not cause any serious improvement that would parallel the success of East Asian countries. By the end of the 1990s, most of the East Asian countries have progressed enough to be classified as developed while the reformers of Africa and Latin America still had to cope with slow or stagnating growth, external trade deficits and unfair distribution of income across their populations. The studies done in this field show various results concerning the effect of liberalisation on growth. This variety is mainly because of the difference between the definitions and indicators of liberalisation and country samples used in these studies. This study suggests that using a complete set of reforming countries in a panel context, while being an easier way to achieve a high level of meaning in the econometric models, does not distinguish the regional differences. When the fact of unequal achievement of growth between regional clusters is considered, this approach seems inadequate to define the differences that cause this problem.