IDENTIFICATION OF THE SPATIAL EFFECTS OF INDUSTRIAL SUBSIDIES

Several analyses show that regional capital incentives induce additional investment and growth (Schalk and Untiedt, 2000). The impact of capital subsidies on employment is more doubtful, even if many studies found that the substitution effect outweighs the output effect (Gabe and Kraybill, 2002). However, the spatial effects of capital subsidies on local development are generally neglected. In a recent paper (De Castris and Pellegrini, 2005) we show that several industrial policies have a strong spatial dimension. This study analyzes the presence of spill-overs generated by subsidised firms, disentangling the effects due to the economic links across areas. The identification strategy is based on the introduction of spatial externalities in the spatial autoregressive model (Anselin, 2003). The presence of specific spatial effects of subsidies is tested on the reduced form of the model, evaluating the implicit common factor restriction. The empirical analysis considers subsidies allocated by Law 488/1992, the main regional policy in Italy, in the Southern regions of the country in the period 1996-2001. A difference-in-difference estimator across 365 subsidised and not subsidised local labour systems is applied. The results suggest the presence of a modest spatial crowding out, where incentives attract skills and growth from neighbouring areas.