Several governments around the world seek to maximize economic growth and social welfare by creating entrepreneurship programs. These policies seek to propagate innovations, reduce unemployment and increase market competitiveness. However, the literature has not yet been able to infer the causal influence of these incentives on the propensity of an individual to create new businesses. In addition, several of these programs may suffer from moral hazards and adverse selection problems, which can lead to worse performance and the destruction of public value. Therefore, this study aims to empirically access the influence of government-led startup programs on the entry of people into entrepreneurship. Furthermore, the use of monitoring schemes is evaluated to see how it can improve policy efficiency. To this end, a natural experiment is performed, based on a regression kink design with probabilistic outcomes. The results demonstrated that government-led programs generally decrease the likelihood of people starting new businesses. Nevertheless, when exploring the heterogeneity of policies, it was noted that programs with higher levels of monitoring increase venture creation, especially opportunity-driven entrepreneurship. These results are consistent with behavioral economics literature, which states that formal control systems influence the psychological motivation of individuals, reducing problems related to moral hazards and adverse selection.