Marco Caliendo, Jens Hogenacker, Steffen Kuenn, Frank Wiessner

• Offering unemployed individuals a subsidy to become self-employed is a widespread active labor market policy strategy. Previous studies have illustrated its high effectiveness to help participants escaping unemployment and improving their labor market prospects compared to other unemployed individuals. However, the examination of start-up subsidies from a business perspective has only received little attention to date. Using a new dataset based on a survey allows us to compare subsidized start-ups out of unemployment with regular business founders, with respect to not only personal characteristics but also business outcomes. The results indicate that previously unemployed entrepreneurs face disadvantages in variables correlated with entrepreneurial ability and access to capital. Nineteen months after start-up, the subsidized businesses experience higher survival, but lag behind regular business founders in terms of income, business growth and innovation. Moreover, we show that expected deadweight effects related to start-up subsidiesOffering unemployed individuals a subsidy to become self-employed is a widespread active labor market policy strategy. Previous studies have illustrated its high effectiveness to help participants escaping unemployment and improving their labor market prospects compared to other unemployed individuals. However, the examination of start-up subsidies from a business perspective has only received little attention to date. Using a new dataset based on a survey allows us to compare subsidized start-ups out of unemployment with regular business founders, with respect to not only personal characteristics but also business outcomes. The results indicate that previously unemployed entrepreneurs face disadvantages in variables correlated with entrepreneurial ability and access to capital. Nineteen months after start-up, the subsidized businesses experience higher survival, but lag behind regular business founders in terms of income, business growth and innovation. Moreover, we show that expected deadweight effects related to start-up subsidies occur on a (much) lower scale than usually assumed.…