We use the 1952 Swedish municipal amalgamation reform to study free-riding and the common pool problem in politics. We expect municipalities that were affected by the reform to increase their debt in anticipation of a merger, and this effect to be larger if they were merged with many other populous municipalities (i.e. facing a large common pool). We use ordinary least squares and matching on the complete cross section of rural municipalities for the period 1947-1951, fixed effects when exploiting the panel features, as well as a geographical instrumental variables strategy. We find an average treatment effect close to the amount that the average merged municipality increased its debt with during this period, which corresponds to 2.8 percent of average income or 63 percent of the average increase in income. However, we do not find larger increases in municipalities that were part of a larger common pool.