Spatial Spillovers and Regional Growth: A Cognitive Approach

Abstract

Spatial spillover effects are widely studied in the literature, as an important source of externality and therefore of discrepancy between private and social optimum, creating the emergence for ad-hoc policy interventions. Within the literature, their spatially-bounded nature is highly emphasised, but the mechanisms for their spatial diffusion are generally considered as a black box, with no reference to territorialized channels through which these externalities spread around. The study aims to present a literature review on the spatial spillover concept, emphasising two main stages of reflection, the so-called spatial approach and the functional approach. This work identifies in a third approach, the cognitive approach, a new perspective to the interpretation of spatial spillovers and their conditions, and on this approach it elaborates theoretically and empirically. The main assumption is that the capabilities of economic agents to absorb and utilize spillovers—implicitly assumed by the spatial spillover theory to be invariant with respect to space—in reality highly depend on strongly territorialized and un-imitable assets, such as local trust, sense of belonging and creativity. This assumption is tested empirically on all NUTS 2 of the 27 European member countries, thanks to a regional forecasting growth model developed by the author, called MASST.

Acknowledgements

The author acknowledges two anonymous referees for their fruitful comments. Any mistakes or omissions remain the author's responsibility.

Notes

Although not explicitly mentioned, the concept of spatial spillover was present in many regional growth theories, like the growth pole theory of Perroux (1955), the export-base theory of North (1955), the potential development theory of Isard (1954), and Giersch (1949), just to quote those theories in which the liason with the modern concept of spatial spillover is evident.

See among others, Almeida and Kogut (1999), Maier and Sedlacek (2005), Coe and Helpman (1995), Acs et al. (1994), Anselin et al. (2000), Audretsch and Feldman (1996), Caniëls and Romijn (2005), Feldman and Audretsch (1999), Fischer (2006), Jaffe et al. (1993), Holod and Reed (2004), Jaffe (1989), and Karlsson and Manducchi (2001).

A collective learning process of this kind was first hypothesized by the GREMI group (Camagni, 1991; Perrin, 1995) and subsequently widely adopted as a sound theoretical concept for the interpretation of knowledge-based development and innovation (Keeble & Wilkinson, 1999, 2000; Capello, 1999; Cappellin, 2003).