Are they worth it?: A counterfactual impact evaluation of a decade of investment subsidies to Italian firms

In this paper we investigate the impact on employment, payroll costs and labour productivity of investment subsidies awarded to industrial firms in Italy during 2000-2009, focusing on estimating heterogeneous impacts by economic values of the subsidy and by relevant firm characteristics. The analysis exploits desirable natural experiment conditions in terms of exogenous treatment exclusions and an unique availability of administrative microdata that cover the universe of Italian firms. Our impact estimates are obtained by means of two different matching estimators and a discontinuity approach, combined with a difference-in-difference scheme, and dynamically implemented in terms of separately considering each consecutive cohort of treated firms. The results of our analysis show that large non-repayable subsidies, particularly when given to large firms and in underdeveloped regions, are an ineffective way to stimulate additional private investments leading to employment growth and/or improvements in labour productivity. Small subsidies given to small firms, not in the context of severely distressed areas, are instead the programme interventions with the best cost-effectiveness.