Does Local Autonomy Increase Local Income? Evidence from Italy

Can fiscal autonomy affect per-capita income levels? The existing literature shows mixed results on the impact of fiscal autonomy on GDP growth, it often uses cross-country datasets comparing nations with different socio-economic contexts. Even when it digs into the subnational entities of a nation either financial indexes or institutional dummies are used as proxies for fiscal autonomy: both can imply endogeneity due either to measurement errors or reverse causality. We empirically investigate the impact of fiscal autonomy on per-capita income stimulated by the proper use of local financial resources. We do this by exploiting an Italian natural experiment comparing the impact on per-capita income of the use of own resources in municipalities belonging to the autonomous provinces of Trento and Bolzano, which manage almost all their taxes autonomously, to those belonging to the neighbouring regions of Veneto and Lombardy, which manage only a small fraction of taxes paid by their citizens. We use a spatial fuzzy regression discontinuity design to compare similar municipalities on the border between the provinces of Trento and Bolzano and Lombardy and Veneto. We find that the higher the level of local financial fiscal autonomy, proxied by the ratio of own tax revenue to total revenue, the higher the level of per-capita income. The proxy is instrumented with a dummy indicating municipalities with a real institutional fiscal autonomy : those belonging to the provinces of Trento and Bolzano. This allows us to interpret the proxy as an exogenous variation indicating institutional fiscal autonomy. We find that a 10 percentage points increase in financial fiscal autonomy increases per-capita income by 3%. Hence, our results suggest that local governments that are more accountable and closer to citizens, manage their revenues in a more efficient way than in the case when they receive transfers from the centre.