The paper studies how the 2004 merger policy reform affected the probability of a merger being challenged by the European Commission. I use a probit model to assess how economic factors drive decisions and to isolate differences between policies before and after the reform. The net sample consists of 341 horizontal mergers from 1990 to 2012. Overall, I find robust evidence of policy shifts due to the reform. By some measures, the policy appears to have slightly softened. There is also an indication that mergers reviewed under the unilateral-effects theory are more likely to be challenged than are those reviewed under the coordinated-effects theory. Market shares and entry barriers are found to influence decisions before and after the reform. However, market shares appear to play a somewhat smaller role after the reform.