- Many countries support business start-ups to spur economic growth and reduce unemployment with different programmes. Evaluation studies of such programmes commonly rely on the conditional independence assumption (CIA), allowing a causal interpretation of the results only if all relevant variables affecting participation and success are accounted for. While the entrepreneurship literature has emphasised the important role of personality traits as predictors for start-up decisions and business success, these variables were neglected in evaluation studies so far due to data limitations. In this paper, we evaluate a new start-up subsidy for unemployed individuals in Germany using propensity score matching under the CIA. Having access to rich administrative-survey data allows us to incorporate usually unobserved personality measures in the evaluation and investigate their impact on the estimated effects. We find strong positive effects on labour market reintegration and earned income for the new programme. Most importantly, resultsMany countries support business start-ups to spur economic growth and reduce unemployment with different programmes. Evaluation studies of such programmes commonly rely on the conditional independence assumption (CIA), allowing a causal interpretation of the results only if all relevant variables affecting participation and success are accounted for. While the entrepreneurship literature has emphasised the important role of personality traits as predictors for start-up decisions and business success, these variables were neglected in evaluation studies so far due to data limitations. In this paper, we evaluate a new start-up subsidy for unemployed individuals in Germany using propensity score matching under the CIA. Having access to rich administrative-survey data allows us to incorporate usually unobserved personality measures in the evaluation and investigate their impact on the estimated effects. We find strong positive effects on labour market reintegration and earned income for the new programme. Most importantly, results including and excluding individuals׳ personalities do not differ significantly, implying that concerns about potential overestimation of programme effects in the absence of personality measures might be less justified if the set of other control variables is rich enough.…
