Social Capital, Institutions and Growth: Further Lessons from the Italian Regional Divide

Since Putnam's work on social capital, the Italian regional case has been a very rich source of both data and theories about the origins of large and persistent differences in local stocks of social capital, and about the impact of such differences on economic performances. The Italian case is widely interpreted as supporting the idea that persistent regional divides are largely explained by local differences in social capital. In this paper we maintain that this interpretation fails to recognize that the current large regional gap in Italy is significantly linked to two policy decisions taken by the central State at the beginning of the 1970s. In particular, we focus on the possibility that social capital became a binding constraint for the growth of southern Italy's mainly as a consequence of the deep process of governmental decentralization that began in the1970s. We formalize this hypothesis by using an endogenous growth model with public capital. In this model, the accumulation of public capital is characterized by the presence of iceberg costs that depend on social capital. Decentralization affects these costs because the impact of the local stocks of social capital on public investment increases when the latter is managed locally. To assess the role of decentralization as a trigger of the influence of local social capital on growth, we control for the impact of labor market reforms, a second and almost simultaneous institutional shock that took place in Italy and that made regional labor markets far more rigid than in the previous decades. In the second part of our paper, we use the large empirical literature on the Italian regions to restrict the values of the parameters of our model in order to perform a simple simulation exercise. In this exercise, the model turns out to be able to account for the major swings in the convergence of southern regions towards the center-northern regions since 1861. The general lessons we can draw from this further analysis of the Italian regional case are as follows. First, we show that the strength of social capital as a determinant of long-run growth may depend on some well-defined characteristic of the institutional context. Second, our model suggests that the economic success of decentralization policies -- even when the budget constraint is not 'soft' -- depends on the local endowment of social capital.