Abstract

Due to searching processes there is a time lag between the advertisement and the filling of a vacancy. Therefore, the number of job openings is widely considered as indicator for the development of unemployment. However, this purely quantitative view on supply and demand does not account for the qualitative fit between unemployed and vacancies. Relevant aspects are for instance regional, occupational or qualification-related matching. The lower this matching the higher is the risk that the vacancies cannot be filled within the usual period of time. As a consequence, this could affect the value of the number of vacancies for forecasting unemployment. This project investigates whether and to what extent forecasts of German unemployment can be improved by considering not only the number of vacancies, but also the qualitative fit between vacancies and unemployed. For this purpose, we construct so-called matching-indices that reflect regional, occupational or qualification-related matching. In appropriate out-of-sample tests, we then evaluate the contribution of these indices to improving accuracy of unemployment forecasts.