We estimate the impact of the Clean Power Plan (CPP) on SO2 and NOx emissions. We focus on these co-pollutants because while the amount of CO2 emitted from electricity generation is well established, the same is not true for local pollutants, SO2 and NOx. Unlike previous papers, and unlike the EPA’s own approach, we do not rely on engineering-based linear programming models to assess the effects of the policy. Instead, we use historical data from US power plants to empirically estimate the effects of compliance with the CPP on emissions of SO2 and NOx. In doing so, we provide an alternative approach to assessing the effects of the CPP. Our results suggest that the EPA may be substantially underestimating the reductions in SO2 emissions, while likely overestimating the reductions in NOx emissions. These differences are important from a public health perspective, because of the health effects of these co-pollutants. We provide estimates of those health effects using an integrated assessment model. Overall, our results suggest that there is value in using an empirical approach to project the environmental impacts of EPA interventions, rather than relying exclusively on engineering-based analyses.