Is there a rural-urban divide? Location and productivity of UK manufacturing

Abstract

The focus of the paper is on evaluating the productivity gap between rural and urban locations in the UK using micro data. We build a structural model of the unobservable productivity emphasising the link between productivity and spatial density of economic activity and adapt the semi-parametric estimation approach proposed in Olley and Pakes (1996) to estimate the parameters of production functions at firm level, within 4-digit UK manufacturing industries, for the period 1997 - 2001. We allow market structure to differ by endogenous export status and location choices and model productivity as a second-order Markov process which greatly enhances our ability to obtain unbiased and consistent estimates of TFP measures at firm level. We aggregate the firm TFPs by location category following the 2004 DEFRA definition of rural and find that aggregate productivity systematically differs across urban, rural less sparse and rural sparse locations as the magnitudes of the differentials are 13.2 percent and 18.0 percent, respectively. Our results are in line with several recent studies. Next, we decompose aggregate productivity into productivity index and industry composition index. The productivity index is the highest in urban locations suggesting that productivity is strongly influenced by density of economic activity and proximity to economic mass. Because industry composition index is positively correlated with productivity index it is evident that locations with high productivity are also characterised by industrial structure enhancing productivity. Further, analysing changes in the decomposition indexes over two periods, before and after implementation of the Euro by the UK main trading partners, reveals substantial heterogeneity in responses across location categories under increased competitive pressure. The main finding is that there is a tendency of rural sparse locations catching up with the urban and rural less sparse location categories in terms of aggregate productivity over the period of analysis. We also find evidence that increased competitive pressure as a result of changes in trade conditions after implementation of the Euro by the UK's main trading partners has acted as a substitute for the role of density of economic activity in enhancing industry composition, especially in rural sparse locations.