Social and economic policies are likely to affect individuals not only by the change in financial incentives, but also by a change in the behavior of their social or work environment. It is our goal to determine social interaction effects in the context of employment behavior in Germany. In particular, we analyze whether the labor supply decisions of mothers with young children depend on the decisions made by their coworkers. The identification of social interaction effects bears various challenges that need to be overcome by the design of the empirical strategy. To this end we follow a reduced-form approach exploiting quasi-random variation in the costs of parental leave induced by the introduction of the new Elterngeld in 2007. Administrative linked employer-employee panel data enables us to assign a peer group to each individual and provides us with individual employment histories of all individuals in these social groups. This data set, combined with microeconometric and quasi-experimental methods, allows us to overcome the challenges associated with the identification and determine existence and magnitude of social interaction effects in employment decisions.