This paper demonstrates that kinks or discontinuities in incentive schemes (e.g., taxes, subsidies, or prices) simultaneously identify agents' intensive and participation margin responses. The proposed semi-nonparametric estimator enables the evaluation of such schemes when existing kink and discontinuity methods are inapplicable due to the presence of both margins. The paper applies the estimator to evaluate the German subsidy for rooftop solar panels, a cornerstone in the global efforts to transit towards a carbon-free economy. Compared to a linear scheme, the government's nonlinear subsidy reduces costs by 0.14 per cent; an optimal nonlinear scheme would more than triple this gain. Ignoring the participation margin when optimising the subsidy would increase costs substantially. The results highlight the importance of estimating both margins for optimal policy design.