CO2 cap-and-trade mechanisms and CO2 emission taxes are becoming increasingly widespread. To assess the impact of a CO2 price, marginal abatement cost curves (MACCs) are a commonly used tool by policy makers, providing a direct graphical link between a CO2 price and the expected abatement. However, such MACCs can suffer from issues related to robustness and granularity. This paper focuses on the relation between a CO2 emission cost and CO2 emission reductions in the power sector. The authors present a new methodology that improves the understanding of the relation between a CO2 cost and CO2 abatement. The methodology is based on the insight that CO2 emissions in the power sector are driven by the composition of the conventional power portfolio, the residual load and the generation costs of the conventional units. The methodology addresses both the robustness issue and the granularity issue related to MACCs. The methodology is based on a bottom-up approach, starting from engineering knowledge of the power sector. It offers policy makers a new tool to assess CO2 abatement options. The methodology is applied to the Central Western European power system and illustrates possible interaction effects between, e.g., fuel switching and renewables deployment.