Local labor market effects of public employment

Abstract

This paper quantifies the impact of public employment on local labor markets in the long-run. We adopt two quantitative approaches and apply them to the case of Spanish cities. In the first, we develop a 3-sector (public, tradable and non-tradable) search and matching model embedded within a spatial equilibrium model. We characterize the steady state of the model, which we calibrate to match the labor market characteristics of the average Spanish city. The model is then used to simulate the local labor market effects of expanding public sector employment. In the second empirical approach, we use regression analysis to estimate the effects of public sector job expansions on decadal changes (1980-1990 and 1990-2001) in the employment and population of Spanish cities. This analysis exploits the dramatic expansion of public employment that followed the advent of democracy in the period 1980 to 2001. The instrumental variables' approach thatwe adopt uses the capital status of cities to instrument for changes in public sector employment. The two empirical approaches yield qualitatively similar results and, thus, cross-check each other. One additional public sector job creates about 1.3 jobs in the private sector. However, these new jobs do not translate into a substantial reduction in the local unemployment rate as better labor market conditions attract new workers to the city. Increasing public employment by 50% only reduces unemployment from 0.156 to 0.150.