European Union and ACP-Countries : are preferential trade agreements to be taken for granted ? Exports of developing countries may be differently stimulated. First, devel-oping countries' exports may benefit from gradual cuts in custom-duties negociated under the auspices of the GATT. Second, developing countries' exports may be favoured by preferential trade agreements like those granted by the European Union towards the ACPs. The theoretical issues of the effects of preferential trade agreements rely on the Viner's analysis of trade flows creation and diversion. The Viner's analysis is however limited to the welfare costs and benefits of the donor. Moreover, those kinds of cost-benefit analyses are by definition static. In this article, we propose a theoretical and econometrical analysis of the Yaounde and Lomé conventions effects on the ACPs' manufactured exports. From a theoretical point of view, the exports of tnanufactured goods depend on the income of import buyers, on sectorial capacity and competitiveness indicators. The two latter variables are either unobservable or unavailable on a panel of developing countries. We substitute the sectorial capacity indicator by a domestic income indicator, making further additional econometric assumptions that allow us to estimate a log linear function of the export rates of ACP's. We secondly substitute the sectorial by a global competitiveness indicator : the Real Effective Exchange Rate (REER). The use of the latter is only allowed if we control for labor and human capital endowments. The finally estimated export rate estimated function depends – positively on the income of OECD countries, human capital endowments and – negatively on the REER and the labor force. Preferential trade agreements are taken into account by binary dummies denoting ACP membership. We check separately the effects of the Conventions on the manufactured goods exports of ACPs to the European countries and to the world. Explanatory variables have the predicted effect. ACP-countries do perform better than other developing countries when they export to European countries. But the latter positive effect is more prevalent under the early conventions. Moreover, ACP-countries do not perform significantly better than other developing countries when they export towards European and non-European countries. We notice however that the Yaoundé convention granting reciprocal trade advantages has solely favoured the ACPs' exports. Some pieces of explanation are proposed. Trade agreements may have postponed economic reforms in ACP-countries, that appeared to be necessary in the face of emerging exporting countries in the South-Asia and South America. Trade agreements appear to have allowed for the durability of distorsions. In last ive diagnose more for a trade flows diversion instead of a stimulation that may question the relevance of trade agreements.