The search for the optimal size of political systems is one of the most enduring in political thought. Given the validity of arguments for and against small units, one might expect variation in rearrangements of unit sizes. However, the reform trend is uniform: units, often at the local level, are amalgamated to harvest scale effects. The purpose of this article is to evaluate the argument on economies of scale in the economic costs of running political systems. Our testing ground is a recent Danish reform. It allows us to avoid endogeneity problems often facing researchers of size reforms. The reform was directed by the central government and constitutes an exogenous shock to 239 municipalities, whereas 32 municipalities were left untouched. We thus have a quasi-experiment with pre- and posttreatment observations for both an experiment and a control group. Our findings show that scale effects, measured as administrative costs per inhabitant, are considerable.
American Journal of Political Science, 2014, Vol 58, Issue 4, p. 790-803