The rational-deductive method of decision making, from Plato to Bentham and much of welfare economics, is based on the ideal that ultimate principles can provide policy administrators with operational principles that can guide decision-making (Arrow 1951). 'Disjoint-incrementalism' (cf. Braybrooke and Lindblom) provides an alternative method to conceptualise policy-making where the status-quo is explicitly taken as a point of departure in decision-taking; small or incremental moves on particular problems rather than comprehensive reform programmes are favoured. The paper enquires into the extent to which these modes of decision-making are prevalent today. It will do so by means of an analysis of current policy-documents from an international organisation: the European Commission. The paper will focus on ex-ante evaluations from the Directorate General of Education and Culture, which have been made available to the public during the last five years covering a wide range of its programmes. Ex-ante evaluations are meant to provide evidence on the need for European policies and justify their specific design. I will review in particular their sections on definition of policy objectives and policy alternatives. Implications will be explored in relation to what degree of change can be expected in European policy in the future. Differences as well as inconsistencies between policy areas will be highlighted.

Content analysis is used to analyse the policy documents using the work of Holsti and Krippendorff.

First, there are significant differences and inconsistencies between the policy objectives and alternatives of EU education policies. This raises, from a procedural point of view, fundamental questions about policy making in the Directorate of Education. Second, from an analytical point of view and linked to issues regarding the use of evidence in policy design, 'disjoint incrementalism' would suggest that careful consideration should be given to the different starting points of member states in the design of EU education policies. The empirical evidence provided shows that EU formulation of alternatives tend to use the EU as a unit of analysis, under the risk of falling into well-known problems of 'ecological fallacy'. Third, looking at the political consequences of the procedural and analytical aspects highlighted above, I show how the policies formulated by the EU can be expected to be better suited for and more beneficial to some countries than others.