In this study, we propose an integrated econometric framework incorporating the difference-in-differences model, the propensity-score-matching difference-in-differences model, and the spatial difference-in-differences model to explore the effect of the Air Pollution Prevention and Control Action Plan on per capita carbon emission in China at the national, regional, and administrative levels. Contradictory results are supported under different econometric models, which highlight the importance and necessity of comprehensive analysis. Taking 285 prefecture-level and above cities as an example, the empirical results show that APPCAP has effectively reduced per capita carbon emission in China at the national level without the consideration of the spatial spillover effect. However, with the consideration of the spatial spillover effect, APPCAP has effectively and directly increased per capita carbon emission in local pilot cities at the national level, and reduced it among pilot cities via the spatial spillover effect, but the effects have become invalid in the non-pilot cities neighboring the pilot cities. Furthermore, the spatial heterogeneity of the effects of APPCAP on per capita carbon emission are supported at the regional and administrative levels. Finally, some specific policy implications are provided for achieving the “win-win” situation of energy saving, emission reduction, and economic development.