Competitive network positions in trade and structural economic growth. A geographically weighted regression analysis for European regions

Mark Thissen, Thomas de Graaff, Frank van Oort
(2016) Papers in Regional Science, 95(1), 159-180

DOI Online

In this paper, we introduce network dependence in European regional growth analy- ses in two new ways. First, we use detailed trade-flow data across European regions to decom- pose regional economic growth into two components: demand-led growth due to growing export markets and structural growth due to growing market shares in those export markets. Only structural growth, constituting on average approximately 20 per cent of total growth, is po- tentially affected by regionally varying locational characteristics. The second network novelty we introduce is revealed competition as a measure of regional network membership in the growth analysis. Applying a neo-classical regional growth model using geographically weighted regression, we show that the degree of revealed competition in trade and services between sec- tors moderates regional structural growth. Regions operating on similar international markets share more favourable structural growth prospects, with locational determinants being more – albeit still limited – significant due to similar externalities. With regional characteristics being of limited sectoral and geographical importance for structural growth, our results suggest that the recently advocated place-based development strategies of European regions should be complemented with competitive regional network strategies.