Spatial externalities between Brazilian municipios and their neighbours

Economic growth certainly is not a uniform process over space, especially in a country as vast and diverse as Brazil. Among the determinants of local growth, the role of externalities has been much discussed in the recent literature (Glaeser et al., 1992). These externalities not only matter for growth within a given city or region but also for growth in neighbouring localities (Lopez-Baso et al., 2004). Our paper aims at providing a causal decomposition of spatial externalities occurring in the growth process of Brazilian municipios. Local growth may impact neighbours through a variety of externalities, and understanding how these externalities operate is crucial for the elaboration of public policies. Previous works on related topics include a recent paper by Lall and Shalizi (2003). Focusing on the Brazilian Northeast, the authors find that growth in municipios is negatively influenced by growth in their neighbourhood. Clearly, it seems worthwhile to extend this kind of analysis to the whole country to try and find if this phenomenon is a Northeastern specificity or a nationally valid result. Moreover, providing a more detailed account of the role of various sources of spatial externalities, and of their potential heterogeneities between regions, would be valuable. Spatial externalities Why may growth at a location affect growth at a neighbouring location? Several causes can be invoked. First, through technological externalities, a locality may benefit from improved economic conditions in another. For instance, if some firms in a locality have developed innovative processes, knowledge spillovers may favour the diffusion of new technologies to firms at neighbouring locations. Linkages between input suppliers and final producers may also be critical: if a final consumption good produced at a particular location benefits from a booming demand, upstream firms in the same region will thrive. Finally, proximity of an important economic centre may improve matching on the labour market, thus reducing costs and increasing labour productivity. Pecuniary externalities may also matter in spatial growth differentials: growth at a location may attract new firms and workers, thus increasing land rents. Transmission of this land market tension to nearby localities can reduce incentives for firms to locate there, and therefore attenuate growth prospects. Finally, local economic growth may foster immigration from less dynamic places. The impact of this migration on both the departure and arrival locations depends on various factors, notably the differences in education levels between the two localities, the substitutability between skilled and unskilled workers in production and the state of local labour markets. Empirical strategy: In order to disentangle which channels matter the most among the various kind of externalities, and to evaluate their spatial scope, we focus on the most rapidly growing Brazilian municipios. From a qualitative point of view, selecting a sample of fast growing localities permits to have a better understanding of the local growth process, since part of these localities are “polar cases” owing their rapid growth to a restricted combination of factors. Moreover, the strength of spatial externalities is likely to be greater in these locations.We evaluate the effects of these externalities on the performance of neighbouring municipios using spatial econometrics methods (Anselin, 2003), and controlling for various local characteristics of the neighbourhood (economic specializations, education, density, public infrastructures, etc.). Using different neighbours’ sets permits to measure the geographical scope of these externalities: some types of externalities only operate at short distance, while others may impact more distant locations. (Neighbours’ sets can be geographically defined, but also sets designed following other similarity or complementarity criterions.) Policy implications: poverty traps and land market issues Understanding how local growth may spread to neighbours or may hinder their economic performance is critical for policy design. Many Brazilian regions are characterized by important spatial inequalities between municipios, which seem to be very persistent over time. These poverty traps result from disparities in growth among neighbours and reducing them requires a better understanding of their formation. Moreover, local policies aiming at fostering growth may have adverse effects on nearby localities, certainly not a desirable outcome. Knowing which are the “bad” channels may help designing more efficient policies. Land and transportation policies are also a closely related issue: some spatial externalities are driven by the functioning of the land market. When rising rents in a growing locality are transmitted to adjacent locations, for instance, public policies may be needed to reduce market tensions through the development of new land plots or the improvement of transportation networks. In this case again, evaluating the strength and spatial scope of pecuniary externalities can help improving these policies.