Cem Ozguzel (p)

This paper provides the first direct evidence on how labor mobility of immigrants cushions natives during a labor demand shock. Spain was one of the hardest hit economies during the Great Recession. Faced with a drop in the local labor demand, immigrant workers moved to other locations in Spain or left the country. Focusing on this episode, using microdata from municipal registers and longitudinal Spanish administrative data, I study the effects of out-migration of the immigrant population on the wages and employment of the remaining natives. I build a shift-share instrument based on the past settlements of the immigrant population across Spain to instrument outflows and argue for a causal relationship. I find that out-migration of immigrants accelerated employment and wage growth of the natives, especially for those with higher substitutability with the leaving population. Moreover, I find that employment effects are driven by the entry to the employment of individuals who were unemployed or inactive, while wage effects were limited to those who were already employed. In sum, through their mobility, immigrants diffused the incidence of local shocks and cushioned the fall of the natives.

Simone Maria Grabner (p)

This paper examines the role of interregional input-output linkages and relatedness on industrial diversification of U.S. counties. The hypothesis is that flows of intermediate goods and services among regions are a vehicle for the diffusion of external capabilities, upon which regional economies can thrive and diversify. Interregional linkages may also relax the role of relatedness, as external knowledge is likely to be unfamiliar, which provides opportunities for new and unrelated recombinations. To proxy interregional input-output linkages of counties we use a county centrality measure derived from the national input-output network. We find a positive relationship between county centrality and industrial diversification, where the latter is measured by the entry of new industry specializations. The results also show, that the impact of relatedness is weakened, if local industries are strategically interconnected within the whole input-output network. Interregional linkages via local industries that are prominently positioned within the national production system appear to stimulate regional diversification in general and unrelated diversification in particular.

Carolina Guevara Rosero (p), María Belén Romero

There is a large heterogeneity of firms at the spatial and the industrial scale. In those contexts, heterogenous firms interact with each other. However, such interaction can take place within similar firms. Big companies tend to exchange with other big companies. The same occurs across micro firms. While the existent literature has paid attention mainly to the interaction across big companies, micro firms are larger in number, in consequence their relationships might be more numerous, leading to higher benefits of agglomeration economies. In developing countries, micro firms have an important role as the engines of employment and poverty alleviation mechanisms (Gomez, 2008). For this reason, this study seeks to determine the effect of agglomeration economies on the productivity of micro-firms of the manufacturing sector, by differentiating the types of agglomeration by nature, namely, diversity, specialization and competition; and by types of firms, namely, big, medium-sized, small and micro firms. To do so, the database of the Economic Census of 2010 of Ecuador is employed. The productivity equation is estimated using a multilevel model. To deal with the correlation of firms within locations and industries, multilevel models are appropriate. Using OLS models would lead to misleading results because the assumption of the independence of observations does not hold true. In contrast, a multilevel approach allows (i) capturing simultaneously the effects of the predictors of group level and individual level, (ii) considering the dependence of the observations within the groups, and (iii) disentangling and examining the inter-individual and inter-group variation (Deleeuw and Meijer, 2008). Thus, the multilevel method allows measuring the variance percentage explained by industry and regional effects (Burstein et al. 2011). Moreover, this methodology acknowledges the hierarchical structure of the productivity, i.e. firms’ productivity responds to a higher level, industry or region (Fazio and Piacentino, 2010).