Decomposing productivity patterns in a conditional convergence framework

In this study we examine regional data on per worker GDP, disaggregated at sectoral level, by focusing our interest on the role of differences in the sectoral composition of activities, and in productivity gaps that are uniform across sectors, in explaining the catching-up process, which is realized through physical and human capital as well as technological knowledge accumulation. Our objective is to investigate how much of the interregional inequality in aggregate productivity per worker is imputable to each component. A methodology for identifying and analyzing sources of inequality from a decomposed perspective is developed in the growth framework by combining a shift-share based technique and a SUR model specification for the conditional-convergence analysis. The proposed approach is employed to analyze aggregate interregional inequality of per worker productivity levels in Italy over the period 1970–2000. With respect to the existing empirical results, our approach provides a more comprehensive and detailed examination of the contribution of each identified component in explaining the regional productivity gaps in Italy. It is argued that region-specific productivity differentials, uniform across sectors, explain a quite large share of differences in productivity per worker. However, sectoral composition plays a non negligible role, although decreasing since the end of 1980s, and very different productivity patterns emerge within geographical areas.