Local production and innovation systems in the state of São Paulo, Brazil

This paper applies a specific methodology to geographically locate and delimit local production and innovation systems in the state of Sao Paulo. A first task, and a bulk part of the paper, is the elaboration of a number of quantitative indicators for the geographic concentration of manufacturing industries and for the location of industries at regional and local levels in that state. The elaboration of indices and other indicators for measuring regional location and specialization of economic activities is an old practice and has been an important object of study in regional economics since the seminal contributions by the pioneers of Regional Science. However, the statistical work developed in this paper is based on two most important recent contributions by P. Krugman (in Geography and Trade, 1991) and D. B. Audretsch & M. Feldman (in R&D spillovers and the geography of innovation and production, The American Economic Review, 86 (3), 1996). These authors calculated locational Gini coefficients for branches of U. S. manufacturing industry (Krugman) and for the geographic concentration of innovative activities and the location of U. S. manufacturing industries (Audretsch & Feldman). This paper applies the same methodology to calculate locational Gini coefficients for manufacturing industries in the state of Sao Paulo, the most advanced state in Brazilian industrialization. The Gini coefficients were calculated from data in RAIS - Relacao Anual de Informacoes Sociais, elaborated by the Brazilian Ministry of Labor. This database provides detailed information on employment and number of plants by branches of manufacturing industries at micro-region and municipality levels. On the basis of the locational Gini coefficients the paper assesses which manufacturing industries are mostly spatially concentrated. Once these industries are identified, the paper proceeds by adding other indicators such as locational quotients, share of the local industry in the total manufacturing employment in the state, number of jobs and number of plants in the local industry. Combined, the Gini coefficients and the other indicators allow to identify, locate, and geographically delimit local production and innovation systems. Additionally, they make it possible to assess to what extent the local system is integrated. These findings are essential for guiding field research and, afterwards, for designing policies oriented to local production and innovation systems.