Is There a Deadweight Loss of Politics in Public Service Provision?

Richard Boylan (Rice University)

Abstract: This study examines how changes in political incentives impact state firms' decisions. We measure changes in political incentives using the timing of elections, since prior scholarly work has emphasized the incentive of elected officials to increase the supply of public goods and decrease taxes right before elections. We examine 503 municipal electric utilities in United States, during the years 1990 through 2013, that operate under the authority of a mayor and a council. We find that the price of electricity charged by municipal utilities is 1% higher two years before mayoral elections than in election years. Harberger's formula is used to compute the loss from this cyclicality in prices as 0.0001% of electricity revenues. Thus our measure of the deadweight losses of politics in public service provision are very close to zero, with a confidence interval that rules out effects greater than 0.0002% of electricity revenues. Our results can be explained by the fact that voters hold mayors accountable for electricity prices. We find that a 10% increase in the municipal utility electricity price reduces the number of votes to the incumbent mayor by 1-2% points, when state electricity prices are held constant. Further, voters seem to be able to distinguish competence from luck: a 10% increase in the municipal utility electricity prices concurrent with a 10% increase in state electricity votes has no effect on the number of votes obtained by the incumbent.


Download the paper