Effects of implausible power plant lifetime assumptions on US federal energy system projected costs, greenhouse gas emissions, air pollution, and water use

Abstract

Abstract Climate change motivates rapid decarbonization of the energy sector, with a particular focus on decarbonizing and expanding the electricity system. Given multiple plausible pathways toward decarbonization, scenario-based analysis of potential energy futures is a common analytical approach in the field of macro-energy systems. Scenario analysis is particularly valuable for assessing multicriteria impacts and identifying potential unintended consequences, especially when policies focus on a specific indicator, like greenhouse gas emissions. In the United States (US), the federal government produces projections for the future US energy system using the National Energy Modeling System, published in an Annual Energy Outlook (AEO) that includes multiple economic and policy scenarios to inform decision making. Although AEO scenarios are not intended as predictions, they are widely used as contextual baselines for evaluating cost, pace, and environmental impacts of the energy transition. As the official energy system scenarios of the US, a wealthy, high-emissions country, AEO scenarios serve an anchoring function that influences perceptions of the relative implications of alternative energy transition strategies. This research shows that projected cost and environmental intensity of the US electricity system are surprisingly stable across AEO scenarios. Why? We show that by not incorporating business-as-usual license expiration and generator lifespans, AEO effectively assumes that power plants can run indefinitely, which overlooks major asset replacement costs and fleet turnover opportunities that would likely accelerate decarbonization. Assuming typical lifespans and licensed operating periods, roughly 700 gigawatts (GW) of fossil and nuclear electricity generation capacity beyond the 170 GW of retirements modeled in AEO2020’s reference case would be expected to close, with an estimated unmodeled replacement cost of 2019$1.3 trillion. This cost is similar to the projected cost impact of a $35/tonne carbon dioxide price. Macro-energy systems analysis can be improved with more attention to physical and infrastructural constraints in models.