The Spatial Dimension of Public Performance Benchmarking: Empirical Evidence from Swedish Municipalities

Benchmarking is used to measure and compare the performance of a given local government against that of its peers. It is commonly held that there are two main benefits of doing so: the first being the prevention of under-performance, and the second being improvement of performance by sharing knowledge about innovative practices. Yet in a world of blame games and electoral competition, the local politicians and bureaucrats also have good reasons to be cautious about presenting and comparing performance information. There has been only limited research on the factors behind engagement in voluntary benchmarking exercises. This paper analyzes why some Swedish municipalities have taken part in a newly established voluntary performance benchmarking project while others have not. The number of participants in this project has skyrocketed from less than 60 (out of 290) in 2006 to 211 in the year 2013. Theory and anecdotal evidence suggest that there are spatial spillover effects, i.e. whether a municipality joins a benchmarking club seems to be affected by the value of the dependent variable Y in the contiguous municipalities. The binary dependent variable in the analysis equals 1 if a municipality participated in the project in a year t, and 0 otherwise; with t = 2007,...,2013. We use linearized GMM logit models. We control for a set of politico-economic variables, including localities' size (in terms of inhabitants), organizational efficiency, fiscal health and fragmentation in the local council.

Surprisingly, we find only limited statistical evidence for spatial spillover of participation in benchmarking projects. Most notably, the probability of participation in year t is determined by participation in the year before (t-1). The benefits of knowing "how do we perform" and mutual exchange among practitioners seem to outweigh the participation fees and opportunity costs. Social group pressure, a pivotal element of the consensual Swedish political system, might add to localities’ reasons to stay in the project.