Agglomeration, spatio-economic clustering and competitive performance: Methodological approaches and empirical evidence from the U.S.

The objective of this dissertation is two fold. Firstly, the dissertation tests how well econometric and spatial analysis can be integrated for conducting industrial cluster analysis. Based on input-output and spatial autocorrelation based multivariate techniques, the dissertation develops a methodological approach for detecting spatially and economically linked industries. The dissertation then operationalizes this methodololigcal framework and applies tit to identify clusters in the U.S. at different geographical scales. The dissertation shows that the application of the methodological frameworks provides an objective mechanism to identify economically, and spatially, and spatio-economically related industries. Secondly, the dissertations empirically examines the extent to which economic and spatio-economic clustering have an impact on industrial productivity and the impacts of regional economic structure or agglomeration externalities on cluster growth. The dissertation finds a positive and statistically significant relationship between membership both in an economic and spatio-economic cluster and industrial productivity. The empirical finds also suggest that diversity externalities have significant impacts on cluster value added growth. The implications of these findings can assists regional economic planners and decision-makers in prescribing cluster based policies for regional economic development