Labor productivity growth in EU28 : spatial panel analysis

This paper is an attempt to explain variations across EU regions in productivity growth and takes into consideration the important structure of the age-productivity relation of Human Capital. The study is fundamentally based on the theory of Fingleton's model which analyses the spatial process of productivity growth on the on the foundations of the theory of New Economic Geography. The applied specification links manufacturing productivity growth to the growth of manufacturing output by the means of Verdoorn's law. The model incorporates productivity-adjusted human capital understood as Total Human Capital Productivity corrected with age structure with the use of productivity as a function of age. Moreover, a new approach to defining the age-productivity curve has been introduced. Based on the previous studies found in the literature the age-productivity function has been interpolated by the means of Radial Basis Function method with thin-plate spline. The age-productivity function allows to describe how the work performance differs over the life period and thus allows for differences in age structure of employees in regions under research. This study covers 261 NUTS 2 regions of EU excluding some French, Portuguese and Spanish regions due to their isolated position and Croatia because of the lack of comparable data. All data used in the empirical part of this study are published by Eurostat and refer to the years 2000-2013. The regional productivity is explained by the quotient of regional GDP and the number of Economically Active Population. The productivity growth is approximated by the exponential change of regional productivity in these years to regional productivity in the year 2000. The regional GDP is expressed in millions of Euro in constant prices (year 2000), where Economically Active Population is in thousands of people 15 years or over. The Human capital is defined by the Employment in Technology and Knowledge-intensive Sectors as a percentage of Economically Active Population. The model has been tested through implemented methodology, namely a spatial panel model with fixed effects. The model presented provides evidence of the importance of increasing returns to scale for regional economic growth, which lead to divergence effects for EU regions. Similar implications can be observed in the case of regionally differentiated human capital. Furthermore, the country fixed effects turned out to be significant. The findings also suggest that productivity in jobs requiring problem solving and learning skills reaches a plateau for the 35-45 age bracket and has its peak around the age of 40. We suggest that the applied approach constitutes an innovation providing additional information hence a deeper analysis of the investigated problem.