This paper investigates the relationship between economic growth performances of countries and their structural input-output network characteristics. We employ a new centrality measure developed by Blöchl et al. (2011) for directed networks with self-loops to determine sectoral heterogeneities in IO tables of 33 OECD countries over the period 1995-2011. Relating the gini indices of these centrality measures to output growth reveals that countries with less heterogenous IO networks tend to grow faster. Such finding implies a key role for the inter-linkages across sectors in economic growth, and underlines the importance of designation of sectoral policy measures to counteract heterogeneity of IO networks.