From sectoral industrial composition to employment and reverse. The Italian Case

Abstract

The diffusion of knowledge among firms and workers, which, in turn, depends on the nature of the knowledge itself and the relatedness or unrelatedness of the industrial composition, is an essential driver for the growth and employment of territories. The proximity between related industries allows for faster diffusion of specific knowledge, as happens in the Marshall-Arrow-Romer (MAR) theory of externalities (agglomeration) and Jacobs’s theory of knowledge transmission. The literature (notably the Evolutionary Economic Geography one, EEG) has empirically investigated the relationship between employment (or GDP) and some measures of industrial composition (e.g., the related and unrelated variety). In this paper, we contribute to the literature in two ways. First, we propose a more coherent use of entropy indexes to measure the relatedness or unrelatedness of industries in an economy. Second, we empirically investigate the nexus between industrial composition and employment level by exploiting these measures. Unlike the existing literature, we argue that such a nexus could be circular. For such a reason, an empirical investigation using a spatial vector autoregressive (SpVAR) model will be performed for the Italian economy at the provincial level. The results show that such a circular nexus exists, shedding new light on the debate.

Keywords

• entropy
• variety
• sectoral composition
• knowledge diffusion
• spatial var