Innovation, territorial capital and reginal growth

Research context and objectives Traditionally, economics explains long run growth in terms of innovation rates. When a country commits money and workforce to the innovative process, the north-east shift of the production possibility frontier would also increase growth rates, and in the long run determine a higher steady state level of personal income. This has in various ways been asserted by economists, geographers, regionalists, sociologists. This view takes for granted that innovation and the accumulation of human capital automatically lead to income and productivity growth. This dissertation argues that the previous statement does not fully tell the whole story. Empirical evidence suggests that countries and regions display different economic performances given similar degrees of commitment to innovative activities. In this dissertation I argue that growth differentials may be at least partially due to different endowments of social and relational capital. This dissertation argues that social and relational capital may provide additional insight into the non-material determinants of regional economic growth. Social and relational capital represent insightful ways to conceive of cultural heritage. Societies are made up of individuals, who, when educating children, teach them norms and values that, at a later stage in their lives, will contribute to the formation of their beliefs and priors, and will shape their opinions and actions. How frequently people attend public meetings, how often they visit friends, how strongly they participate to political activities, and above all, how much they trust each other (in a nutshell: how high a society’s social capital), will co-determine the ease with which knowledge is produced, how easily knowledge spreads across people and firms, and how efficiently new knowledge is used in everyday life. Analogously, the higher the intensity and density of personal networks of trustworthy business-oriented relations (in other words, the higher the stock of relational capital), the more efficient the process of knowledge production, diffusion and exploitation will be. The main objective of the dissertation is therefore to critically summarize the existing studies on regional growth, identifying as the crucial gap a relative lack of attention to social and relational capital as possible generators of increasing returns to the accumulation of physical factors. Three empirical studies complete the dissertation with a special reference to three main aspects of regional growth: • The impact of social and relational capital on the formation of increasing returns to regional knowledge; • The role of social and relational capital in the formation of regional knowledge spillovers; • The impact of social and relational capital as magnifiers of agglomeration externalities. Methods Each of the three empirical applications is based on the most recent advancements both in terms of economic theory as well as with respect to econometric specification and technique. The dissertation is based on a first crucial step, viz. the collection of a comprehensive dataset on European NUTS2 regions. This is based on several sources, among which: • EUROSTAT New Cronos Database; • Cambridge Econometrics Regional Database; • European Values Study Database; • International Social Survey Programme database; • Geo-referenced data from different sources; • Data on the cumulated number of EU Framework 5 Programme projects. Methods of data collection, analysis, validation and merging are carefully described in the dissertation. The empirical chapter of the dissertation lay down three major econometric models to be tested, respectively with the following techniques: • Spatial autoregressive models, and spatial panel data models (Elhorst 2003, 2009); • Robust pooled least squares for cross-sectoral estimates; • Instrumental variables estimates for the assessment of cause-effect relations. Results Results show in general a good fit of the model predictions. In particular the following results are to be found in the dissertation: • Social and relational capital contribute to the formation of increasing returns to regional knowledge. Not only they represent a relevant production factor; also, the higher their stock, the higher the returns of regional knowledge. This statement is verified both econometrically (through interaction terms) as well as statistically (with interaction graphs yielding significance bands). • Social and relational capital do magnify the effects of spatial externalities in a Glaeser et al. –like model (Glaeser et al. 1992). Specialization (Marshallian) economies have a higher impact on regional labour growth the higher the stock of social and relational capital; sectoral diversity (Jacobian) externalities have a less negative impact on growth; finally, competition (Porter) externalities become slightly significant and positive when social and relational capital stocks are higher. • Social and relational capital in the region generating new knowledge both help retaining internally the positive effects of new knowledge inside the region; the stock of social capital in surrounding regions determines a higher knowledge leakage outside the region; finally, the stock of relational capital in neighbouring regions is associated with a lower knowledge leakage from the region generating new knowledge