Explaining Differences In Efficiency: The Case Of Local Government Literature

One learns two main lessons from the efficiency literature on local governments. The first lesson regards the heterogeneity in the efficiency scores reported in primary papers. The second lesson is that there is no quantitative evidence on the role played by the features of each paper (i.e. estimation method, sample size, dimension, returns to scale) in explaining the differences in results. In order to fill this gap, we review the related empirical literature and perform a Meta Regression Analysis (MRA) by examining 360 efficiency scores retrieved from 54 papers published from 1993 to 2016. The meta-regression is based on a random effect model estimated with the Random Effects Maximum Likelihood (REML) technique, because it controls for within- and between-study heterogeneity. We also run a fixed effect unrestricted Weighted Least Squares (WLS) regression. Due to its main research focus, that is measuring the impact of potential sources of heterogeneity on local government efficiency, the paper contributes to the debate in two ways. One of this concerns the role of methodological choices made by researchers when performing an efficiency study. The second regards the role of deregulation in local government, which is a policy-issue in a number of countries. Results show that efficiency scores are highly heterogeneous. To be precise, significant differences in means are found when grouping efficiency by different criteria. The meta-regression estimates indicate that studies focusing on technical efficiency provide higher efficiency scores than works evaluating cost efficiency. Using panel data in primary studies allows researchers to obtain higher efficiency of local government than papers using cross-section data. Interestingly, FDH studies yield, on average, higher efficiency scores than DEA papers, thereby suggesting that in this literature the convexity hypothesis of the production set is a matter. Furthermore, we find that primary papers evaluating the efficiency of European municipalities provide lower efficiency scores than studies focusing on other countries (USA, Africa, Asia and Latina America). We also provide evidence that the estimated efficiency scores in primary papers focusing on the municipalities of a region are, one average, lower than those retrieved from studies addressing the efficiency of the national system of local government.