We performed a number of robustness checks to verify the stability of our results. First of all we estimated pooled OLS regressions (which have the strongest assumptions with regard to heteroskedasticy and auto-correlation as well as distribution of errors) and calculated the variance inflation factors (VIF). The analysis revealed that only in the model for the solar sector, multi-collinearity might be an issue (mean of VIF 9.17). However, this high value stems from the VIF of our control variables and therefore has no influence on the coefficients of our independent variables of interest (i.e. the policy measures). Second, we estimated random effects estimators (REE) for our models (see Table 20 in the appendix). All our models (PCSE, OLS and REE) display consistent results. When using the OLS model, grants and subsidies as well as codes and standards become insignificant. Third, we ran our models, including only the significant variables from our previous analysis. These analysis displayed consistent results throughout all models. Finally, to account for the excessive number of zeros in our sample, we crosschecked the disaggregated results (i.e. individual policy instruments) with aggregated results (categories of policy instruments) for the entire analysis. The models displayed consistent results.