Observable and non-observable social interactions in labor supply

This paper analyzes determinants of individual work force participation in a bivariate probit model using data from the GSOEP. From Monte Carlo simulation, we know that there is low probability to detect social interactions in medium data sets. Nevertheless, we find that apart from influences such as age, number of children and local employment rates, educational and gender differences as well as the work force participation of the eldest brother or sister also have a significant influence on a person´s work force participation. The results show that the second-born takes a similar job participation decision as his/her peer