Accountability in amalgamated municipalities: an incumbent's nightmare?

Electoral accountability crucially depends on retrospective evaluations of voters. Its basic principle focuses on the idea that citizens reward a good performance with electoral support for the incumbent government and punishes a bad performance by voting for the opposition. The retrospective voting literature includes several studies proving that voters hold local representatives accountable for various outcomes such as unemployment, crime, test scores in schools, or the maintenance of roads. However, the exact ways how accountability works at the local level when considering amalgamation are still being investigated. Therefore, we ask: how does accountability work in amalgamated municipalities? There is a relatively straightforward argument for why amalgamation should be important. The essential precondition for functioning accountability is information. On this point, we can follow the logic of the neighbourhood effect, suggesting that individuals are influenced by the nature of the politically relevant information circulating within their social networks. The mass media do not cover local elections in most municipalities. Although the situations are distinctly different in big cities and smaller municipalities, the amount of information about local elections circulated in the media is dramatically lower than those about parliamentary elections. In this sense, information is most likely spread by direct contact, to which pre-merger borders can be felt as a barrier. Considering how representatives are related to different parts of municipalities, it seems plausible that in amalgamated municipalities, especially in peripheral locations, the rewarding effect should be weaker than in central areas and non-amalgamated municipalities. As a result, we expect that the inhabitants of peripheral communities in amalgamated municipalities should reward incumbents less than the inhabitants of geographically central communities and inhabitants of non-amalgamated municipalities. To statistically test this hypothesis, we use data from the exit poll conducted during the 2022 Czech local elections covering answers from 11 000 respondents in 60 municipalities.