Wasting public money to buy turnout: quasi-experimental evidence from the Polish battle for fire trucks

ABSTRACT

It is well known that incumbents enjoy advantages in elections. Among others, they may use public resources to increase their electoral chances by mobilizing specific social groups to vote. In the 2020 presidential elections, the Polish government organised a competition: small municipalities with the highest turnout could win a fire truck. We exploit the threshold of 20,000 that arbitrarily divided municipalities into competing and non-competing for this public good and apply the quasi-experimental Regression Discontinuity Design. The competition aimed to increase the turnout in districts supporting the incumbent president. However, we find no evidence of any significant impact of the competition on the turnout or support for the incumbent around the threshold. We discuss why similar practices are common despite this result.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 After the elections, the NGO Watchdog (2020) confirmed that the fire trucks were indeed given to the municipalities. Interestingly, in two cases, the truck was not accepted by the awarded municipality because it did not have any fire brigade. One of those municipalities forwarded the truck to a neighbouring municipality whose fire brigade protects it. The other one sold the truck for around 10% of the value to another municipality, although the inhabitants wanted to create a voluntary fire brigade to be able to accept the truck. The latter case, as well as the lack of municipalities that would state no need for a fire truck, shows that the state of local fire brigades most likely did not affect the behaviour of the local community.

2 Also, the rules of the competition did not specify how the decisive number of inhabitants would be defined. Usually, for electoral purposes, the population evidence (ewidencja ludności) is used, in which residents are permanently registered.