One policy, different effects: Estimating the region‐specific impacts of EU cohesion policy

Many academic papers have looked at the economic effects of the EU cohesion policy, which still remain an open empirical issue. The focus of the most recent literature has been on the heterogeneous effects of the policy and the identification of regional conditioning factors. However, most of the existing studies generally assume slope homogeneity for different cross‐sectional units (i.e., regions) and they estimate the average effects of the policy for all the European regions and/or selected groups of regions. Past works also employ data covering few programming periods. This paper has two main goals. First, we study the heterogeneous consequences of EU cohesion policy on regional economic growth in Europe over the past three decades, by applying a heterogeneous coefficient approach to new panel‐time series data. We calculate the region‐specific effects of the policy in terms of long‐run gross domestic product growth. Second, we study regional differences in terms of policy effects depending on the level of assistance received by the regions. We make a distinction among cases of effective, ineffective, trigger and marginal policy. We also document that the effectiveness of EU cohesion policy in the long run can be explained by some of the key factors used in the literature. Finally, we discuss the need for ineffective cases to learn from effective and trigger ones.