This paper analyses the causal effects of weaker dismissal protection on the incidence of long-term sickness ( six weeks). We exploit a German policy change, which shifted the threshold exempting small establishments from dismissal protection from five to ten workers. Using administrative data,...