Repealing power-plant rules deletes carbon capture's federal customer

The EPA has put a number on its deregulation, and the transition finance market should read it in the direction the agency did not intend: by the agency's own estimate, repealing the Biden-era greenhouse gas standards for fossil fuel-fired power plants will save the power sector more than $300 billion in compliance costs through 2047. Read from the other side, $300 billion measures the spending the standards would have compelled, and the slice of that spending with a federal requirement behind it was carbon capture. The requirement is gone.

The package is broader than a single rule and had been signaled since last year, when the EPA first moved to scrap the standards on the argument that power-plant greenhouse gases "do not contribute significantly to dangerous air pollution" within the meaning of the Clean Air Act. Announced by EPA Administrator Lee Zeldin at a conference of G20 energy ministers, the repeal reaches back past the Biden rules to their foundation: the Obama administration's 2015 Clean Power Plan, which set CO2 limits for coal and gas plants to push generation toward natural gas and renewables. The Supreme Court ruled in 2022 that the EPA lacked authority to devise emissions caps aimed at shifting generation from high- to low-emitting sources, and the Biden EPA answered in 2024 with plant-level standards requiring more CO2 controls and carbon capture and storage at modified coal plants and new gas plants. The agency has now repealed most of that 2024 package, including the emissions guidelines for existing fossil fuel-fired plants and the capture requirements.

Spread evenly across the 21 years to 2047, the EPA's compliance-cost figure is about $14 billion a year the power sector will not spend, and the useful question for a lender is who was going to receive it, because what a capture requirement on new gas plants really supplies is a customer. A developer building under the rule had a compliance reason to contract for capture capacity regardless of marginal cost — a rare thing in transition infrastructure: a buyer whose demand does not depend on the commodity price, the corporate sustainability budget, or the depth of a voluntary market. Remove the requirement and the same project has to be financed against whatever stands in its place: offtake contracts, state procurement, an anchor buyer with its own motives, every one of them recourse to a counterparty rather than to a rulebook, and the difference shows up in tenor, in the cost of debt, and in how much equity a sponsor leaves in the deal.

The more consequential item in the announcement is the one that is not final: alongside the repeal, the EPA proposed rescinding the 2015 greenhouse gas findings for fossil fuel-fired power plants, which the agency says would make it substantially harder for a future administration to impose limits on the sector. A rule can be repealed and reinstated as administrations change; a findings rescission is an attempt to raise the price of the reinstatement. For a lender underwriting a twenty-year asset, how political risk changes over time matters more than where it stands today, and the effect here is to shorten the window in which a federal mandate can be assumed to exist at all. That pushes capital toward revenue that survives a change in the rulebook and away from projects that only close with a compliance obligation attached.

Environmental groups criticized the announcement sharply and indicated they will take legal action to preserve the rules, which means the compliance relief the EPA granted is contested rather than banked. A sponsor treating the $300 billion as a permanent improvement in power-sector economics is underwriting a legal outcome nobody can schedule, and the defensible posture is to model the repeal as reversible in part and to price the capture pipeline on both branches — mandate returns, mandate does not. Either way the project needs a buyer who is not the Clean Air Act.

Transition capital has already moved from labels to project-level risk, and Monday's decision is the next repricing round arriving on schedule — it will separate lenders who priced construction from those who priced intent. The decision also cuts against the companion view that the public balance sheet absorbs early transition risk and private capital follows once the state has taken construction and policy risk. The state can withdraw the absorber, and the EPA just did, at least for the US power sector and at least until a court says otherwise, leaving capital that followed a federal requirement into a project carrying policy risk it did not have to price while the rule was in force.

The arithmetic the EPA published in 2022 has not moved: the electric power sector accounts for roughly a quarter of US greenhouse gas emissions. Monday's change is in the financing question beneath that number, now that the federal government has repealed most of the standards that required anyone in the sector to bring it down. For the moment, the answer is voluntary buyers, state programs, and lenders willing to underwrite capture against a contract instead of a rule.