A composed error model decomposition and spatial analysis of local unemployment

The differences in the regional unemployment rates, as well as their formation mechanism and persistence, have given rise to a great number of papers in the last decades. This work contributes to that strand of literature from two different perspectives. In the first part of our work, we follow the methodological proposal established by Hofler and Murphy (1989) and Aysun et al. (2014). We make use of an estimation of a stochastic cost frontier to breakdown the Spanish provincial effective unemployment (NUTS-3) in two different components: first one associated with aggregate supply side factors, and the other one more related to the aggregate demand side factors. The second part of our research analyzes the existence of spatial dependence patterns among the Spanish provinces in the effective unemployment and in both above mentioned components. The decomposition performed in the first part of our research will let us know the margin that the policymakers have when they deal with unemployment reductions by means of aggregate supply and aggregate demand policies. Finally, the spatial analysis of the unemployment rates amongst the Spanish provinces can potentially have also significant implications from an economic policy viewpoint since we find that there are common formation patterns or clusters of unemployment