Search

There is ample evidence of regions diversifying in new occupations that are related to pre-existing activities in the region. However, it is still poorly understood through which mechanisms related diversification operates. To unpack relatedness, we distinguish between three mechanisms: complementarity (interdependent tasks), similarity (sharing similar skills) and local synergy (based on pure co-location). We propose a measure for each of these relatedness dimensions and assess their impact on the evolution of the occupational structure of 389 US Metropolitan Statistical Areas (MSA) for the period 2005–2016. Our findings show that new jobs appearing in MSA’s are related to existing ones, while those more likely to disappear are more unrelated to a city’s jobs’ portfolio. We found that all three relatedness dimensions matter, but local synergy shows the largest impact on entry and exit of jobs in US cities, thus being the strongest force of diversification.

The Evolutionary Economic Geography literature has extensively investigated the impact of local capabilities on innovation and regional diversification. However, the role of external linkages has only received limited attention. The general concern of this thesis is how external linkages impact innovation and regional diversification with an evolutionary approach. This is further elaborated in four research questions. First, what is the impact of returnees on firm innovation? Second, what is the impact of MNEs on regional industrial diversification? Third, through what knowledge spillover channels can MNEs impact local export performance? Last, what is the impact of co-inventor networks on regional technological diversification? The findings of this thesis show that external linkages do matter for innovation and regional diversification both in the context of China and Europe. And external linkages could interact with relatedness and complexity in regional diversification, thus triggering different diversification paths. In more detail, the main content of each chapter could be summarized as follows. Chapter 1 presents the motivation, theoretical background, research questions, and outline of the thesis. Chapter 2 investigates the relationship between the global mobility of Chinese returnees and firm innovation for Chinese public listed firms. We construct a new dataset from CNRDS, PatentsView, CSMAR, and LinkedIn data. Our findings show that, in general, both returnee employees and managers have positive and significant impacts on firm innovation. The roles of returnee employees and returnee managers are complementary, that is, returnee employees as the agents of domestic patenting, and returnee managers as the ones for international patenting. We also find that the heterogeneities of returnees matter for firm innovation. Chapter 3 studies the relationship between foreign MNEs and industry entry in Chinese regions by linking the MNE research with regional diversification literature. The association between industry entry and the presence of MNEs in the same industry and related industries is tested for 331 Chinese prefectures between 1998 and 2007. Our findings indicate that the presence of MNEs in the same industry and related industries is positively and significantly associated with a higher probability of industry entry, and this association also holds for the entry of unrelated and complex industries. The effect of MNEs is larger for knowledge-intensive industries. Chapter 4 focuses on the role of relatedness and forward-backward linkages in the process of MNEs knowledge spillovers to domestic firms in the context of industrial export quantity and quality in Chinese regions. The effect of relatedness and forward-backward linkages is tested for 279 Chinese prefectures between 2000 and 2007. The results in Chapter 4 show that relatedness linkages matter in the context of export quantity and new exporting destination entry, whereas forward-backward linkages matter for the complexity of export products. Chapter 5 explores the role of both internal and external co-inventor networks in regional technological diversification, considering the effect of both relatedness and knowledge complexity with the OECD REGPAT dataset. Our results show that both internal co-inventor network connectivity and external co-inventor network linkages could facilitate regional technology entry and inhibit technology exit. However, too inwardly oriented, or outwardly connected co-inventor networks would hamper patent production. We also find suggestive evidence that both internal co-inventor network connectivity and external co-inventor network linkages could induce related and complex regional technological diversification. Chapter 6 concludes and discusses policy implications. It also discusses the limitations of this thesis and proposes directions for future research.

This research stresses the role of geography in the formation of network ties that channel ideas, knowledge and technology. It introduces novel aspects to consider while studying the drivers and outcomes of spatial network formation. The thesis establishes that spinoff background matters for tie formation in cluster knowledge networks. It shows that by searching for the drivers of network tie creation and tie persistence separately, we can better understand the evolution of industry clusters. It also provides evidence that co-location and complexity of technologies determine their relatedness ties. Furthermore, this research establishes a connection between forming structurally important linkages and being successful in urban agglomeration of creative industries. Besides extending the literature on spatial networks with new aspects, each of the chapters reinforce that geography is an important determinant of network tie formation.

This research stresses the role of geography in the formation of network ties that channel ideas, knowledge and technology. It introduces novel aspects to consider while studying the drivers and outcomes of spatial network formation. The thesis establishes that spinoff background matters for tie formation in cluster knowledge networks. It shows that by searching for the drivers of network tie creation and tie persistence separately, we can better understand the evolution of industry clusters. It also provides evidence that co-location and complexity of technologies determine their relatedness ties. Furthermore, this research establishes a connection between forming structurally important linkages and being successful in urban agglomeration of creative industries. Besides extending the literature on spatial networks with new aspects, each of the chapters reinforce that geography is an important determinant of network tie formation.

The question of why some countries or regions experience higher growth rates and levels of development than others has been debated and studied for long time and remains one of the most important issues in political and economic debates. The aim of this thesis is to investigate some of the factors that contribute to the unevenness of economic performance among European regions, focusing in particular regional industrial composition, spatial and network relations and the technological and institutional characteristics of the regions. The role of industrial structure and composition of regions is assessed in relation to their knowledge and technological features (Chapter 1). Our results indicate that related variety exerts positive effects in only regions characterized by higher technological and knowledge capabilities, whereas no significant effect is found for average or less advanced areas. More generally, we conclude that different levels of knowledge endowment may condition the way knowledge-related externalities affect the local economy. Following this insight, we study spatial and network relations as sources for new knowledge (Chapter 2). In line with the previous literature we find that spatial and trade linkages indeed contribute to channeling new ideas and technologies in the regional economy, with a positive effect on local productivity. However, network connections to most advanced regions – which would potentially grant access to highly valuable knowledge - have no beneficial effect unless the “linking-in” region presents high level of absorptive capacity. Recent literature on institutions has shown how quality of government and social capital are important determinant of economic and innovative performance. Following some contributions in Evolutionary Economic Geography, we investigate how institutional factors interplay with industrial compositions. In Chapter 3, we show how industrial relatedness and bridging social capital are key drivers of regional diversification. Unlike, formal institutions do not directly affect regional branching, though some evidence suggests that they contribute to limiting the potentially negative effects of bonding social capital. The role of industrial relatedness in allowing knowledge and ideas to flow is further studied in relation to the presence, entry and exit of multinational enterprises (MNEs) (Chapter 4). We find that the number of MNEs in related sectors has significant effects on industrial employment only in knowledge-intensive industries in advanced regions. Additionally, the positive relation between employment level and proximity to industries attracting foreign companies provides supportive evidence of short-term knowledge spillovers. However, both intra- and inter-industry spillovers of MNEs are heterogeneous across industries and regions. Positive MNE externalities appear to be stronger in advanced sectors and, more surprisingly, in Eastern EU regions. These regions, however, are also those that suffer the most in the case of MNE exits.