Some remarks on a social network approach to identifying key sectors

Ostensibly, certain adaptations of social network theory extend and improve the traditional key-sector approaches. Our analysis of the underlying algebraic properties shows that a social-network-based approach proposed by Garcia Muniz et al. [(2008) Key Sectors: A New Proposal from Network Theory. Regional Studies, 42, 1013-1030] does not relate final demand and output in ways comparable to key-sector measures that are based on the static Leontief input-output model. Using the most recent IO table for Poland we show that the modified approach can lead to spurious empirical results and, as a consequence, to false policy implications.