Do municipal mergers reduce public expenditure? Evidence from the MTE approach

This study examines whether municipal mergers reduce public expenditure by examining the marginal treatment effect (MTE) of municipal mergers. Extant papers in the literature have paid little attention to bias from self-selection among municipal mergers or to heterogeneity in the treatment effects of the mergers. Corresponding to these issues, we use the instrumental variables used in Miyazaki (2018) [Applied Economics, 50(10), pp. 1108-1121] and estimate the MTEs of the mergers. From the estimated MTEs, we construct several estimands, and we show that the municipal mergers resulted in an increase in public expenditure on average. Moreover, we confirm that the local average treatment effect (LATE) was quite large from FY2006 to FY2015, although it decreased suddenly in FY2016, when some incentives that promoted the mergers ended. This implies that the incentives offered by the national government negated the cost reductions resulting from the municipal mergers.