The impact of German job protection, legislation on job, creation in small establishments: An application of the regression discontinuity design

One proposal frequently raised to increase flexibility of the German labour market is the liberalization of the job protection law. It applies to those establishments with more than a cut-off number of employees. The argument examined in this paper is that this step in legal regulation hinders small enterprises from job creation. Changes in the cut-off number in the late 1990's provide the basis for estimating this effect. The evaluation approach is a Regression Discontinuity Design using these changes as natural experiments. Local treatment effects can be estimated non-parametrically by local linear regression. The data base used is the IAB establishment panel. The paper is the first one to exploit the policy changes named above and controlling for self-selection into the treatment job protection using minimal assumptions concerning model specification. The results are in line with earlier studies finding no evidence for hindering effects on job growth in small establishments.