[Journal:] Zeitschrift für ArbeitsmarktForschung – Journal for Labour Market Research [ISSN:] 2510-5027 [Volume:] 38 [Year:] 2005 [Issue:] 4 [Pages:] 557-572
Taking up the discussion surrounding the reliability of the IAB reports on the stock of vacancies in Germany in this journal (ZAF, issue 1/2005), this paper clears up misunderstandings, corrects incorrect statements, and presents further new arguments. The paper is intended to show that it would be possible, without any additional costs, to use the IAB reports on the stock of vacancies in the economy as a whole to obtain additional information of interest for labour market diagnosis if the 'jobs to be filled immediately' were recorded correctly. This discussion resulted, however, in the well-founded suspicion that the extrapolation results of the IAB reports on the stock of vacancies in Germany are incorrect in two ways: firstly as a consequence of a selection bias in the data basis with regard to the firms' reporting behaviour, and secondly due to the adjustment of the extrapolations to incorrect data from the Federal Employment Service about stocks of registered vacancies. This suspicion is explained separately in the appendix.