Since the mid-1980s, differentiated integration (DI) has become a core feature of European integration. In most cases, DI has been an unintended and unanticipated outcome of intergovernmental negotiations on widening and deepening that governments intended to result in uniform integration initially, but failed to produce unanimous agreement. In other words, DI has been a second-best solution to prevent complete integration failure. By contrast, based on existing theory and research on the causes and patterns of differentiation in the EU, this paper explores ways in which DI could be used as a feature of ex ante integration design rather than an ad hoc integration fix. The paper starts with a discussion of demand and supply factors that contribute to the efficiency and legitimacy of DI: heterogeneity of integration preferences and capacities; a large size of the integrationist group; low externalities between the levels of integration; and the absence of prior uniform integration. The paper goes on to discuss the suitability for DI of various combinations of these factors and to assess DI opportunities in the current reform agenda of the EU.
