This thesis sets out to analyse the effects of diverse forms of geographical location and network structures on industrial growth and business cycle co-movement. The initial focus is on the analysis of the effects of clusters on industrial growth to determine whether clusters can cope with industrial shocks. Next, in order to fully understand the complex relationships that characterise modern industries, the concept of industrial network is used to analyse the input-output
global trade relationships to determine if the network characteristics of an industry can affect growth and co-movement. In that sense, this thesis offers a comprehensive review of the determinants of growth at industry level by analysing industrial growth in a broad range of settings; from a geographically-tied clusters definition to a broader network approach using novel techniques and datasets.
On one hand, the results suggest that clusters are relatively neutral; they do not promote higher growth in the presence of a positive national shock and neither do they generate a lower probability of being hit adversely by an economic downturn; location alone may not capture the full scale of relationships that constitute a cluster. On the other hand, the empirical study of industrial networks suggests that some network characteristics affect industrial growth. The most
important finding is that industries that have greater co-movement are also those that have higher rates of growth during periods of global economic expansion as well as exhibiting more rapid declines in the face of global economic contraction.
If the centrality of an industry is revealed to be an important transmitter of both global economic growth and downturns, as this thesis suggests, then investigating these phenomena more deeply should be high on the global research agenda.