Addressing spatial dependence in technical efficiency estimation: A Spatial DEA frontier approach

This paper introduces a new specification for the nonparametric production-frontier based on Data Envelopment Analysis (DEA) when dealing with decision-making units whose economic performances are correlated with those of the neighbors (spatial dependence). To illustrate the bias reduction that the SpDEA provides with respect to standard DEA methods, an analysis of the regional production frontiers for the NUTS-2 European regions during the period 2000-2014 was carried out. The estimated SpDEA scores show a bimodal distribution do not detected by the standard DEA estimates. The results confirm the crucial role of space, offering important new insights on both the causes of regional disparities in labour productivity and the observed polarization of the European distribution of per capita income.