The purpose of this paper is three-fold. First, we analyse and compare the welfare effects of two CAP reform scenarios in the grain sector, a fully decoupled one in which compensatory payments for grains are perceived as production neutral and a fully coupled one where compensatory payments are considered as price complements and taken into account by farmers in their production decisions. We show that the decoupled assumption will lead to a larger surplus for EC grain producers than the copuled scenario. Second, we discuss the implications of breaking down grains into three distinct commodities, i.e., wheat, maize and coarse grains, for the analysis of the effect of the CAP reform in the EC grain sector. We istimate policy price transmission equations for these three grains and integrate this additional "domestic market" dimension in the analysis. The estimation results show that policy price transmission, i.e., the transmission of institutional price changes to market prices, is imperfect and that own-price elasticities, on both the supply side and the demand side, differ largely. As a result, the response of each cereal to institutional price cuts is likely to be different. Third, the implications of assuming coupled or decoupled compensatory payments for grains and of considering market price responses to policy price changes are assessed by conducting specific simulation exercises. These simulations are based on a revised and extended version of the model MISS (Modele international simplifie de simulation), which considers explicitly the three cereals and incorporates policy price transmission equations for these three commodities. The disaggregation of grains is based on a weak separability assumption, on both the supply side and the demand side, between wheat, maize and other coarse grains on one hand, and other netputs taken into account in the model on the other hand. The empirical illustration shows clearly the relevance of the theoretical analysis and highlights the sensitivity of grain balances in the Community to both the coupled versus decoupled assumption and the perfect versus imperfect policy price transmission hypothesis. Finally, this analysis raises the question of the compatibility between the CAP reform and a GATT agreement defined on the basis of the Washington compromise.