Public infrastructure, private capital and the performance of manufactures: short and long run effects

Despite the general idea among economists that public capital is a necessary but not sufficient condition for regional economic development, there is not consensus on the quantification of its effect. It is our belief that the variety of results on infrastructure impact in the economic literature is due to the fact that it depends on a series of factors. Among them we can think of a different sectoral effect, the level of economic development in the area, so as the amount of the existing public capital stocks and their interaction with infrastructures in other regions. In this paper we focus on this latter condition, having into account that the evidence on public capital spillovers obtained in other studies is not homogeneous. This way, whereas in Holtz-Eakin and Schwartz (1995) the stock of highways in a state does not suppose benefits in productivity further its borders, Mas et al. (1997) obtain that the infrastructural endowment in the neighbours can be important. Using different functional forms, the first purpose of this paper is to shed light into the existence of public capital spillovers considering not only the physical interaction among regions but also other forms of relationship such as trade linkages. In a second step, we follow a discusion on the existence of other economic externalities that in case of not being considered could be the reason why we obtain biased estimations for regional returns to scale. Both purposes are treated through the use of spatial econometrics techniques in the Spanish regional case. Keywords: Public capital, Infrastructure Spillovers, External Effects, Spatial Econometrics