Based upon the results of the European Representation Study, this paper asks two further questions: Why are party voters less favourable towards specific EU policies than party elites? Second, how does political representation of EU preferences actually work, is it an elite- or a mass-driven process? The data-sets of the European Election Studies 1979 and 1994 are analysed which involve both an elite and a mass survey component. In contrast to earlier research, political representation of EU preferences appears to work rather well as far as the grand directions of policy making are concerned; and that party elites behave responsively regarding changing EU preferences among their voters.