Abstract

Multi-level governance is one of the most outstanding issues in political science. It is usually depicted as an effective way of designing policies, because of its capacity to incorporate heterogeneous information in policy making. However, the literature has also warned against the coordination problems that can emerge because of the difficulty in satisfying divergent interests of the involved actors. This article aims at contributing to this debate by explaining two specific dynamics: the distributive slippage in vertical governance and the type III error in horizontal governance. In order to address these issues,.process tracing has been applied to analyze two design processes. The application of causal inference tests to the collected evidence has allowed to significantly increase confidence in the two hypothesised inherent dynamics.