Matching and Labour Market Efficiency across Space and through EU accession: Evidence from Latvia, Estonia and Slovenia

During the transition to market economy and the accession to the EU Central and Eastern Eu- ropean countries have witnessed remarkable changes in the structure and functioning of national economies. The aim of this paper is to analyze the dynamics of aggregate and regional labour markets through the last decade in several new EU member states (Latvia, Estonia and Slovenia). The estimation of aggregate matching functions on monthly panel (1999-2006) data allows per- forming the diagnostics of labour market efficiency in terms of worker-firm matching. We exploit regional and country differences, the dynamics and changes over time (we compare pre to post EU enlargement periods) and measure the importance of spatial spill over effects in matching. The potential misspecification of the matching function is addressed by allowing for stock-flow specification and for spatial interactions between regions in terms of worker and job flows. The results reveal that in transition - EU accession context the hiring process is labour demand driven and displays the existence of stock-flow patterns and spatial spillovers. In Latvia due to job shortage and limited labour demand, hires mainly occur between the stock of unemployed and the inflow of new vacancies, while in Slovenia the inflow of new unemployed also plays an important role in match creation. The aggregate efficiency of the labour market in terms of worker-firm matching increases over time in Latvia and seems to decrease in Estonia and Slovenia. The role of labour demand in creating new hires stands crucial in three countries, but the results also feature the development of a new trend: after the accession to the EU the role of labour demand in the matching process becomes weaker, but the role of labour supply is increasing. The efficiency of matching varies across districts and regions and can partially be explained by the population density in the area or by its geographical location (its proximity to the national borders). Spatial spill over effects in matching are confirmed to be statistically significant: unemployed do not limit their search to the region of residence and search in neighboring areas. The asymmetry of spill over effects is weak in Latvia, while in Slovenia the magnitude of the effects depends on economic context in neighboring regions or also on local population density.