The elasticity and efficiency of job matching in Dutch regional labour markets

In many matching function studies, the measure of job matches and the stock of job searchers does not correspond. New hires or filled vacancies are related to unemployed job searchers only whereas this should be the total stock of job searchers, as any job searcher, besides unemployed, can be hired or fill a vacancy. Furthermore, in many studies the matching function exhibits constant returns to scale. We show that there is a relation between the choice of variables representing matches and job searchers and the returns to scale of the matching function. If only unemployed are assumed to search for a job, then a matching function of total hires with constant returns to scale yields a matching function of hires from unemployment with decreasing returns to scale. A matching function is estimated, which corroborates this point, using quarterly regional data for The Netherlands. The differences between our model and other regional matching studies can be linked to the role that employed job searchers and job searchers not in the labour force play in the matching process. It implies that if policy measures to improve the functioning of the regional labour market are required, they are best addressed to the northern provinces in The Netherlands.