Chapter 30: Environmental and Energy Policy

How a country argues with its own atmosphere

The atmosphere does not vote. It does not file briefs in federal court. It does not call its senator. It does not show up at school-board meetings, attend EPA notice-and-comment hearings, or contribute to political action committees. It does not care, in any sense the word usually means, what Americans believe about it.

It does, however, respond to what Americans do.

When you burn a gallon of gasoline in a Toyota Camry in suburban Atlanta, roughly nineteen pounds of carbon dioxide enter the atmosphere and disperse globally. They mix with the carbon dioxide produced by a coal plant in West Virginia, a cement kiln in China, a household stove in Lagos, and a forest fire in Canada. The atmosphere does not distinguish among them. The molecules from the Camry add to a stock that has grown from roughly 280 parts per million at the start of the Industrial Revolution to roughly 425 parts per million in 2024 — a one-and-a-half times concentration in human history's blink. The molecules' radiative properties — the way they absorb infrared radiation and re-emit some of it back toward Earth — are settled physics, understood since the 1850s.

What humans do politically about that physics is not settled. It is one of the most contested domains in American politics, and the contest is real. There is a science part — what the atmosphere is doing, why, and what is likely to happen as a result — and on the science part, the consensus among working climate scientists is overwhelming. There is a policy part — what to do about it — and on the policy part, reasonable people who accept the science disagree intensely, because the questions involve values, distributional effects, technology bets, federalism, and judgments about state capacity that the science alone does not resolve.

This chapter takes both parts seriously. It states the science. It steel-mans the policy disagreements. It walks through the institutional architecture that translates the disagreement into rules, dollars, and infrastructure — or that fails to translate them, when the politics deadlocks.

The institutions involved are sprawling. The Environmental Protection Agency. The Department of Energy. The Department of the Interior. The Department of Commerce's National Oceanic and Atmospheric Administration. The Council on Environmental Quality. Fifty state environmental agencies. Federal courts of appeals that have become the day-to-day arbiters of environmental rules. Two political parties whose positions on energy and climate have polarized further over the past twenty years, even as the underlying technology and economics have shifted dramatically. A statutory infrastructure built mostly between 1963 and 1990 that the country has not significantly updated since. A body of judicial doctrine — the major-questions doctrine after West Virginia v. EPA (2022), the loss of Chevron deference after Loper Bright (2024) — that has constrained executive-branch climate action.

And underneath all of it: an energy system that is changing for reasons that have little to do with environmental policy. Fracking lowered the price of natural gas and displaced coal. Solar-cell costs fell by roughly 90% in a decade. Battery costs fell similarly. China built more renewable capacity than anyone projected, and also more coal capacity. The American electrical grid in 2024 looks nothing like the American electrical grid in 2000, and it will look nothing like it in 2040, regardless of what any administration does in the meantime. Policy can speed the transition, slow the transition, redistribute who pays for it, and shape who profits — but it cannot stop the transition, because the transition is being driven by costs and physics that no administration controls.

This chapter is about how the United States manages that combination of certainty (the science) and contestation (the policy) within its institutional structure. It begins with the science, briefly. It then maps the institutional architecture, the major statutory tools, and the live policy fights of the 2020s.

The science consensus

Climate science is among the most studied empirical questions in modern science, and the consensus is robust. Greenhouse-gas concentrations in the atmosphere have risen substantially since industrialization. Global surface temperatures have warmed roughly 1.2°C since the late 19th century. The warming pattern — more at the poles, more at night, in the troposphere but cooling in the stratosphere — is the pattern predicted by greenhouse forcing, not the pattern predicted by solar variability or ocean cycles. Sea level has risen, ice mass has declined, ocean heat content has increased, and the timing of seasons in temperate latitudes has shifted. The probability of certain extreme weather events — heat waves above all, but also some heavy precipitation events — has shifted in the direction of more frequent and more intense.

The Intergovernmental Panel on Climate Change (IPCC), which synthesizes the peer-reviewed literature roughly every six to eight years, summarizes the consensus in its Sixth Assessment Report (2021–2023): warming is "unequivocal," human influence is the dominant cause, and continued emissions will produce continued warming. NASA, NOAA, the National Academies of Sciences, the American Meteorological Society, the American Geophysical Union, the American Physical Society, and every major scientific society in the United States have issued consistent statements.

Some readers will encounter this consensus statement and find it congenial. Others will encounter it and feel skeptical, perhaps because their political affiliation correlates with skepticism on this question, perhaps because they have read specific critiques of climate models or temperature records. This textbook does not adjudicate that skepticism. It states the scientific consensus clearly and notes, without condescension, that the political-affiliation correlation is real and well-documented in survey data (Pew, ANES, Gallup time series). What follows assumes the consensus position because the empirical disagreement is not where the political contest lies. The political contest is over what to do.

The institutional structure

The Environmental Protection Agency (EPA)

The EPA was created in 1970 by executive order of President Richard Nixon, consolidating environmental functions that had been scattered across Interior, Agriculture, HEW, and other agencies. Nixon's reasoning was partly substantive — air and water quality had become national political issues after the 1969 Cuyahoga River fire and visible smog crises in major cities — and partly tactical: a unified agency was easier to control politically than scattered pieces. The structural choice mattered. EPA is an independent agency in the sense that it is not housed in a Cabinet department, but its administrator serves at the pleasure of the president and reports to the president directly, which means it is fully under presidential control and changes character substantially with each administration.

EPA's statutory authority comes from a series of laws Congress passed mostly between 1963 and 1990:

• Clean Air Act (1963, with major 1970, 1977, and 1990 amendments). Authorizes EPA to set National Ambient Air Quality Standards (NAAQS) for criteria pollutants and to regulate hazardous air pollutants. The 1990 amendments added the acid-rain cap-and-trade program for sulfur dioxide and the Title V operating-permits program. Section 111 authorizes performance standards for new and existing stationary sources — the legal hook for power-plant rules that became central in Massachusetts v. EPA (2007), which held that greenhouse gases are "air pollutants" within the Act's meaning, and West Virginia v. EPA (2022), which limited how broadly EPA can read 111(d).
• Clean Water Act (1972). Regulates point-source water pollution through the National Pollutant Discharge Elimination System (NPDES). The "Waters of the United States" definition has been litigated repeatedly, most recently in Sackett v. EPA (2023), which narrowed the Act's reach over wetlands.
• Safe Drinking Water Act (1974). Sets standards for public water systems.
• Resource Conservation and Recovery Act (RCRA) (1976). Regulates hazardous-waste handling, treatment, storage, and disposal.
• Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA / Superfund) (1980). Liability and cleanup framework for contaminated sites.
• Toxic Substances Control Act (TSCA) (1976, substantially amended 2016). Regulation of industrial chemicals.

This is a 1970s–early-1990s statutory framework. Congress has not passed a major new environmental statute since the 1990 Clean Air Act amendments — a 35-year gap that explains much of what the chapter discusses below. Climate-change regulation is happening under statutes written before "climate change" was a phrase a member of Congress used.

The Department of Energy (DOE)

DOE was created in 1977 by President Carter, consolidating the Atomic Energy Commission's successor agencies and various energy offices into a Cabinet department. Its original mission was substantially nuclear-weapons-related — DOE still runs the National Nuclear Security Administration, the weapons labs (Los Alamos, Lawrence Livermore, Sandia), and the nuclear-stockpile stewardship program. Roughly half of DOE's budget historically goes to weapons-related work.

The other half — the part that matters for environmental and energy policy — funds:

• Energy R&D. Fossil-fuel research (Office of Fossil Energy and Carbon Management), nuclear R&D (Office of Nuclear Energy), renewable energy and efficiency (EERE), basic science (Office of Science, including the national labs).
• Strategic Petroleum Reserve. The 700-million-barrel emergency oil stockpile, drawn down for the first time at scale during the 2022 oil-price spike.
• Loan programs. The Loan Programs Office, which has financed early-stage energy projects (the Tesla loan during the 2009 stimulus is the famous success; Solyndra is the famous failure).
• Energy Information Administration (EIA). Statistical agency that publishes the data this chapter cites.

DOE's policy reach is mostly about money — grants, loans, R&D — not regulation. The regulatory authority lives at EPA, Interior, and the Federal Energy Regulatory Commission (FERC, which is technically inside DOE but functions as an independent regulatory commission for interstate electric and gas markets).

The Department of the Interior

Interior manages roughly 500 million acres of federal land — about 28% of the U.S. landmass, concentrated heavily in the West and Alaska. The Bureau of Land Management (BLM) manages most of it, on a "multiple-use" mandate that includes grazing, timber, mineral extraction, recreation, and conservation. The National Park Service manages roughly 85 million acres of national parks, monuments, and historic sites. The U.S. Geological Survey provides scientific data. The Bureau of Ocean Energy Management (BOEM) and the Bureau of Safety and Environmental Enforcement (BSEE) manage offshore oil and gas leasing and safety. The Fish and Wildlife Service implements the Endangered Species Act.

Interior is the single most consequential federal agency for resource extraction, because so much of the nation's oil, gas, coal, timber, and minerals sits on federal land. Decisions about leasing, permitting, and royalties on those lands are made by Interior under statutes (the Mineral Leasing Act of 1920, the Federal Land Policy and Management Act of 1976, the Outer Continental Shelf Lands Act) that give the Secretary substantial discretion. That discretion is exercised differently across administrations, which produces the policy pendulum described later in this chapter.

NOAA, CEQ, and the rest

The National Oceanic and Atmospheric Administration (NOAA), housed in the Department of Commerce, runs the National Weather Service, fisheries management, ocean and coastal programs, and most federal climate-data production. The Council on Environmental Quality (CEQ), inside the Executive Office of the President, coordinates environmental policy across agencies and writes the regulations implementing the National Environmental Policy Act. The Department of Agriculture, through the Forest Service and the Natural Resources Conservation Service, manages additional federal lands and shapes agricultural environmental practice. The Department of Transportation issues fuel-economy standards (CAFE) jointly with EPA. FERC regulates interstate electric and gas markets and sites natural-gas pipelines.

State environmental agencies

This chapter has a national focus, but most environmental enforcement happens at the state level. The Clean Air Act, the Clean Water Act, RCRA, and the Safe Drinking Water Act are all built on a "cooperative federalism" model: federal statutes set floors, but states implement and enforce. State agencies issue most permits, conduct most inspections, and bring most enforcement actions. The Texas Commission on Environmental Quality, the California Air Resources Board (CARB), New York's Department of Environmental Conservation, and Florida's Department of Environmental Protection are major regulatory bodies in their own right, with budgets and enforcement reach that exceed many federal regulatory agencies.

The state-federal split matters politically. California's CARB has historically had a special waiver under the Clean Air Act allowing it to set vehicle emissions standards stricter than federal standards, a waiver other states can opt into — which has produced something close to a two-track national vehicle-emissions regime, with the size of California's market giving it leverage that no other state has individually.

Policy responses to climate

When climate change went from a scientific finding to a policy issue in the 1980s and 1990s, three broad approaches emerged for how governments could reduce greenhouse-gas emissions: pricing, regulation, and subsidy. Each has a coherent intellectual case. Each has been tried, somewhere, at some scale. Each has political-economy challenges that have made comprehensive U.S. adoption difficult. The contemporary American policy mix is best understood as a partial, halting, sometimes-conflicting deployment of all three.

Carbon pricing

The economic case for carbon pricing is straightforward. Greenhouse gases are an externality — burning fossil fuels imposes costs on people who didn't cause the burning — and the standard Pigovian remedy is to put a price on the externality so that decision-makers internalize it. A carbon tax does this directly. A cap-and-trade system does it indirectly: government caps total emissions, issues or auctions permits, and lets the market price emit through trading.

The conservative case for a carbon tax. Many economists, including a long roster on the political right, have endorsed carbon pricing as the least-bad climate-policy instrument: it is a market mechanism, it produces a predictable price signal that firms can plan around, it reduces the need for prescriptive regulation, and it can be made revenue-neutral by rebating revenue to households or by cutting other taxes. The Climate Leadership Council, founded by James Baker, George Shultz, Henry Paulson, and other Republican former officials, proposed a carbon tax with per-capita dividend in 2017 — a plan endorsed by more than 3,500 economists in a 2019 statement. The intellectual lineage runs through Greg Mankiw, Martin Feldstein, and the broader free-market environmentalist tradition.

The progressive case for cap-and-trade. The cap-and-trade approach has the political-economy advantage of producing a quantitative emissions target rather than a price target, which matters to advocates who want certainty about cumulative emissions. The European Union's Emissions Trading System has operated since 2005 and has produced measurable emissions declines. In the United States, the Regional Greenhouse Gas Initiative (RGGI), a cooperative cap-and-trade program among Northeastern states, has operated since 2009 and is generally judged to have reduced power-sector emissions cost-effectively. California's economy-wide cap-and-trade program, launched in 2013, is the largest sub-national carbon market.

The political-economy challenge. Despite the intellectual case, carbon pricing has not been adopted at the U.S. federal level, and several state-level efforts have failed. The Waxman-Markey cap-and-trade bill passed the House in 2009 but never got a Senate vote. Washington State voters rejected carbon-tax ballot initiatives in 2016 and 2018. Australia adopted a carbon tax in 2012 and repealed it in 2014 after a price spike on electricity bills became politically toxic. France's gilets jaunes protests in 2018–2019 were sparked partly by fuel-tax increases. The pattern is consistent: carbon pricing imposes visible costs on consumers (higher gas prices, higher electricity bills) before delivering benefits, and the visibility makes the policy politically fragile. Defenders argue that revenue rebates can solve the problem; critics observe that voters don't seem to credit the rebate the way they notice the tax.

Regulation

The alternative to pricing is direct regulation: tell sources what to emit, by sector, technology, or facility. EPA has substantial regulatory authority under the Clean Air Act, and most U.S. climate policy in the 2010s and 2020s has run through that authority rather than through pricing.

Vehicle emissions. The Obama administration in 2012 issued joint EPA/DOT fuel-economy standards (the CAFE standards, paired with greenhouse-gas tailpipe standards) that targeted a fleet-wide average of roughly 54.5 mpg by 2025. The Trump administration in 2020 weakened the standards. The Biden administration in 2024 issued new standards targeting roughly 50 mpg by 2031 and effectively requiring substantial electric-vehicle market share. The Trump-2 administration in 2025 began rolling back the Biden standards. Each rule has been litigated.

Power plants. EPA's Clean Power Plan (2015) under President Obama would have set state-level emissions targets for existing power plants, allowing states to comply through generation shifting (away from coal toward gas and renewables). The Supreme Court stayed the rule in 2016 and ultimately struck it down in West Virginia v. EPA (2022), holding that EPA's authority under Section 111(d) does not extend to "generation shifting" because such a major economic and political question requires clear congressional authorization. This was the case that established the major questions doctrine as a binding constraint on agency action: when an agency claims authority over an issue of "vast economic and political significance," it must point to clear statutory text, not generic provisions.

The Biden administration in 2024 issued power-plant rules drafted to fit within the post-West Virginia doctrinal frame, focusing on within-source compliance options (e.g., carbon capture, hydrogen co-firing) rather than generation shifting. These rules face their own litigation challenges and are being substantially rolled back by the Trump-2 EPA.

Methane. EPA finalized methane rules for oil and gas operations in 2024 under the Biden administration; the Trump-2 administration is rolling them back.

The pattern is clear: the regulatory approach is highly sensitive to administration changes, the major-questions doctrine has narrowed the available toolkit, and durable climate regulation under the existing Clean Air Act is difficult absent congressional updating of the statute.

Subsidies

The third approach is subsidy: pay for the technologies you want to deploy. The U.S. has used this approach since the 2009 Recovery Act, which included roughly $90 billion in clean-energy investments. The approach went much larger with the **Inflation Reduction Act (IRA) of 2022**, which authorized roughly $370 billion over ten years (with subsequent estimates ranging higher as uncapped tax credits scaled up) for clean-energy and climate spending.

The IRA's structure is mostly tax credits: the Investment Tax Credit (ITC) for solar, the Production Tax Credit (PTC) for wind and other generation, an EV credit for consumers and a manufacturer's credit for EV assembly, hydrogen production credits, advanced manufacturing credits, and credits for carbon capture, nuclear production, and clean fuels. The political design of the IRA was deliberate: spread clean-energy manufacturing investment geographically across red and blue districts so that repeal would impose economic costs on Republican constituencies, raising the political cost of repeal.

The bet has been partially successful and partially not. As of 2025, the largest clean-energy manufacturing investments have been concentrated in Republican-held districts in Georgia, Texas, North Carolina, South Carolina, Ohio, and Michigan. Some Republican members of Congress whose districts received major investments publicly opposed full IRA repeal. The Trump-2 administration's 2025 budget bill (the One Big Beautiful Bill Act) repealed or rolled back substantial portions of the IRA — particularly the EV consumer credit and certain manufacturing credits — but left other provisions partially intact. The case study at the end of this chapter explores the IRA's design and partial unwinding in detail.

Public investment

Beyond pricing, regulation, and subsidies, governments invest directly in research and infrastructure. DOE's R&D budget — through the national labs, ARPA-E, and the Office of Science — funds early-stage technology that the private sector underinvests in. The Bipartisan Infrastructure Law (2021) authorized roughly $65 billion for grid and transmission upgrades, $7.5 billion for EV charging infrastructure, and substantial sums for nuclear, hydrogen, and carbon-capture demonstration projects. Federal procurement — particularly the Department of Defense, the largest single energy consumer in the federal government — shapes early markets for clean technologies.

The energy mix

Policy debates about climate happen against the backdrop of an energy system that is changing for reasons of cost, technology, and physics that policy can shape but not control. The current state, in 2024 EIA data:

U.S. electricity generation, 2024 (approximate): - Natural gas: ~43% - Nuclear: ~19% - Coal: ~16% - Wind: ~10% - Hydroelectric: ~6% - Solar: ~4% - Biomass, geothermal, petroleum, other: ~2%

These numbers shift year to year. The trend over the past two decades is unmistakable: coal has fallen from roughly 50% of U.S. electricity in 2000 to roughly 16% in 2024, while natural gas has roughly doubled and renewables (mostly wind and solar) have grown from near-zero to about 14% combined. Nuclear has been roughly flat. Total U.S. electricity demand has been roughly flat for two decades but is now beginning to grow again, driven by data centers (especially AI training and inference), electrification of vehicles and heating, and reshoring of energy-intensive manufacturing.

Coal decline. The dominant cause of coal's decline is not environmental regulation. It is natural gas. Hydraulic fracturing (fracking) for natural gas, which became commercially viable around 2008, dropped U.S. natural-gas prices substantially and made gas-fired power cheaper to operate than coal-fired power for most of the past fifteen years. Environmental regulations — particularly EPA's Mercury and Air Toxics Standards (MATS) and the Cross-State Air Pollution Rule — accelerated coal-plant retirements at the margin, but the economic dynamic was driven by cheap gas. Coal communities in Appalachia, Wyoming, and the Mountain West have experienced substantial economic dislocation as a result, with policy responses (the Partnership for Opportunity and Workforce and Economic Revitalization, or POWER Initiative, under Obama; various Trump and Biden-era transition programs) generally judged inadequate to the scale of the loss.

Nuclear. The U.S. operates roughly 94 commercial reactors at about 54 sites, providing roughly 19% of electricity and roughly half of zero-carbon electricity. The fleet is aging — most reactors were built in the 1970s and 1980s — and only a handful of new units have come online in recent decades. Vogtle Units 3 and 4 in Georgia, completed in 2023 and 2024 respectively, were the first new U.S. reactor builds in roughly thirty years. They came in years late and tens of billions of dollars over budget, illustrating the cost-and-schedule problem that has bedeviled large nuclear in the U.S. The current frontier is small modular reactors (SMRs) — designs in the 50–300 MW range that proponents argue can be factory-built and deployed faster — though as of 2025 no commercial SMRs are operating in the U.S. Several reactors that were scheduled for retirement (Three Mile Island Unit 1 / Crane, Palisades) are being restarted under new economic circumstances, partly driven by data-center demand for firm zero-carbon power.

Wind and solar. Wind and solar costs have declined dramatically — solar by roughly 90% in a decade, onshore wind by roughly 70% — making them the cheapest sources of new electricity in many U.S. markets. Their growth is rapid, but from a small base, and their intermittency is the binding constraint. A solar plant produces nothing at night and less in winter. A wind plant produces nothing on still days. As renewable shares grow above roughly 30–40% of generation in a given grid region, the cost of integrating them rises non-linearly, because the value of additional renewable generation falls (because it competes with other renewables at the same hours) while the cost of firming up the grid against low-renewable hours rises.

Storage. The current binding constraint on higher renewable shares is storage. Lithium-ion battery costs have fallen substantially — roughly 90% in a decade — and grid-scale battery storage is now economically deployed at multi-hour durations in California, Texas, and elsewhere. But four-hour batteries, the current standard, do not solve the multi-day low-renewable problem (a winter week with low wind and short days). Long-duration storage (multi-day to seasonal) — pumped hydro, compressed air, hydrogen, novel chemistries — is the next technology frontier, and its economics remain unsettled.

Fracking transformation. The shale-gas and tight-oil revolution beginning around 2008 reshaped American energy. The U.S. went from a substantial net importer of oil and gas to the world's largest producer of both, and a net exporter of natural gas. The macroeconomic effects were large — lower energy prices, manufacturing reshoring in petrochemicals — and the geographic effects concentrated in specific regions (the Permian Basin in West Texas and southeast New Mexico, the Marcellus and Utica Shales in Appalachia, the Bakken in North Dakota). Environmentally, fracking cut U.S. emissions by displacing coal with lower-carbon gas, while also raising local concerns about water contamination, induced seismicity, methane leakage, and air quality. The political economy of fracking has reshaped state politics in Pennsylvania, Ohio, Texas, Oklahoma, and North Dakota. This chapter treats fracking as an institutional fact, not as a contested policy choice — though specific regulations on fracking (methane rules, water-disposal rules, leasing rules) are contested.

Federal lands and resource extraction

About 28% of U.S. land is federally owned, concentrated in the West. Nevada is roughly 80% federal. Utah is roughly 65%. Idaho, Oregon, Wyoming, Alaska, California, and Arizona all have substantial federal-land shares. East of the Mississippi, federal land is much smaller and concentrated mostly in national parks, national forests, and military installations.

Federal land matters for environmental and energy policy because so much oil, gas, coal, timber, and mineral extraction happens on it. Roughly 25% of U.S. oil and gas production — and a majority of U.S. coal production — comes from federal lands or federal offshore waters. Decisions about leasing those lands are made by Interior under the Mineral Leasing Act, the Outer Continental Shelf Lands Act, and other statutes that give the Secretary substantial discretion. Different administrations exercise that discretion very differently:

• Coal leasing. The Obama administration in 2016 imposed a moratorium on new federal coal leases pending a programmatic review. The Trump administration ended the moratorium in 2017. The Biden administration in 2021 ordered a review and partially restored leasing constraints. The Trump-2 administration is expanding coal leasing again.
• Oil and gas leasing. Trump-1 expanded onshore and offshore lease sales. Biden in 2021 paused new federal oil and gas leasing pending a programmatic review; courts ordered him to resume sales, and the IRA tied future renewable-energy leasing to continued oil-and-gas leasing. Trump-2 expanded leasing further, including in areas of the Arctic National Wildlife Refuge (ANWR) that had been closed under Biden.
• National monuments. Presidents can designate national monuments under the Antiquities Act of 1906. Obama designated several large monuments. Trump-1 reduced Bears Ears and Grand Staircase-Escalante National Monuments substantially. Biden restored them. Trump-2 is reviewing them again.

The pendulum is real. It generates regulatory uncertainty for resource industries, lawsuits at every turn, and a sense in many Western states that their economies are subject to political swings far from their voters. The "Sagebrush Rebellion" tradition, dating to the late 1970s, holds that Western states should have more authority over the federal lands within their borders. Some of its arguments are constitutional (the Equal Footing Doctrine), some economic (state-level decisions reflect state-level economic interests), some political (Western voters are underrepresented in federal land decisions). The counter-position holds that federal lands are the property of all Americans, that conservation values transcend state borders, and that history has shown state management often favors short-term extraction over long-term stewardship. Each side has serious adherents, and the policy fight has been continuous for fifty years.

NEPA and environmental review

The National Environmental Policy Act of 1970 is one of the most consequential federal statutes that almost no one outside of policy circles has heard of. NEPA requires federal agencies, before taking any "major federal action significantly affecting the quality of the human environment," to prepare an Environmental Impact Statement (EIS) analyzing the action's effects and considering alternatives. The statute itself is short and procedural — it does not require agencies to choose the most environmentally protective option; it requires them to consider environmental effects and disclose them.

In practice, NEPA review has become extensive. EISs routinely run thousands of pages. Median EIS preparation time has grown from roughly two years in the 1970s to roughly 4–5 years in recent decades (CEQ data). Federal agencies often prepare Environmental Assessments (EAs) and Categorical Exclusions (CEs) for less significant actions to avoid full EIS preparation, but each of those determinations can be litigated.

The contemporary debate over NEPA features unusual cross-spectrum agreement. Two camps argue, for different reasons, that NEPA review has accumulated procedural barriers that block important projects:

• The progressive "abundance" critique. A growing strand of progressive thought, associated with writers including Ezra Klein, Derek Thompson, Jerusalem Demsas, and Yoni Appelbaum, argues that procedural barriers — NEPA, zoning, occupational licensing — block exactly the kinds of investments progressives want: clean-energy infrastructure, electricity transmission, public transit, affordable housing. The 2023 book Abundance (Klein and Thompson) and the broader "supply-side progressivism" movement argue that liberalism's procedural commitments have come into tension with its substantive goals, and that streamlining is necessary to deliver progressive policy outcomes (a fast clean-energy transition, in particular).
• The conservative state-capacity critique. A parallel critique on the right, associated with writers including Tyler Cowen and Marc Andreessen, argues that the U.S. has lost the ability to build large physical infrastructure — at any speed, in any sector — because of accumulated procedural overhead, of which NEPA is a major component. The argument is less environmental than it is about state capacity and economic dynamism.

Both critiques agree that NEPA, as currently implemented, blocks projects on both political wishlists. They disagree about which projects matter most. They generally agree that the underlying statute could be amended to maintain environmental disclosure and analysis while reducing the procedural barriers — particularly through page limits, time limits, and tighter standing rules.

The defenders of robust NEPA review argue that the statute's procedural requirements have prevented genuinely harmful projects, especially in low-income and minority communities; that the median NEPA timeline overstates the typical case (most federal actions don't require an EIS at all); and that the real cause of permitting delay is often agency under-staffing rather than statutory burden. Streamlining, they argue, will mostly benefit fossil-fuel infrastructure that progressives oppose.

The 2023 Fiscal Responsibility Act — the debt-ceiling deal that emerged from the Biden-McCarthy negotiations — included the first significant NEPA amendments since 1982. It imposed a 150-page limit on standard EISs (300 pages for "extraordinarily complex" projects), a two-year deadline for EIS completion, and tighter scope and standing rules. Senator Joe Manchin's "permitting reform" effort, which dominated his final two years in the Senate, sought further amendments — including specific authorization for the Mountain Valley Pipeline — and produced the "Energy Permitting Reform Act of 2024," which passed the Senate Energy and Natural Resources Committee but not the Senate floor. The case study on permitting reform explores this in depth.

Environmental justice

The empirical pattern is consistent across many studies: pollution exposure in the United States correlates with race and income. Communities of color and low-income communities are more likely to live near hazardous-waste sites, large industrial facilities, ports, freeways, and airports. Their air, water, and soil exposures are systematically higher. The patterns persist after controlling for income, suggesting race-specific factors in addition to income-specific ones. The literature dates to Robert Bullard's 1990 Dumping in Dixie and the GAO's 1983 study on hazardous-waste siting, and has been replicated extensively.

The institutional response has evolved over time:

• EPA Office of Environmental Justice (OEJ), created in 1992 under President George H.W. Bush, charged with environmental-justice considerations in agency decision-making.
• Executive Order 12898 (Clinton, 1994), directing federal agencies to identify and address disproportionate environmental and health impacts on minority and low-income populations.
• Justice40 Initiative (Biden, 2021), committing that 40% of the benefits of certain federal climate, clean-energy, and environmental investments would flow to "disadvantaged communities" defined by a federal screening tool (the Climate and Economic Justice Screening Tool, or CEJST).
• State-level environmental-justice statutes, including New Jersey's 2020 EJ law (the strongest in the country, requiring permit denial when cumulative impacts on overburdened communities exceed thresholds), New York's Climate Leadership and Community Protection Act (2019), and Washington's HEAL Act.

The conservative critique of environmental-justice programs runs along several lines. The first is constitutional: race-conscious program design raises 14th Amendment equal-protection concerns, particularly after Students for Fair Admissions v. Harvard (2023). Defenders respond that Justice40 is technically race-neutral — it uses geographic and socioeconomic variables, not race directly — though critics observe that the variables are race-correlated by design. The second is effectiveness: critics argue that "disadvantaged community" definitions have been gamed (some programs have classified large fractions of states as disadvantaged), reducing the redistributive precision the program was supposed to provide. The third is opportunity-cost: dollars spent on environmental-justice add-ons to projects can slow or block the projects themselves, including clean-energy projects whose benefits accrue heavily to the same communities.

The defenders of environmental-justice programming argue that the empirical exposure pattern is robust, that institutional remedies are warranted, that race-correlated geographic variables are the constitutionally permissible way to address race-correlated harms, and that procedural delay concerns are typically overstated relative to actual program implementation. Both positions have serious adherents and are present in current policy fights.

Climate adaptation

Adaptation — adjusting to climate impacts that are already locked in — is becoming a larger part of U.S. environmental policy as physical impacts accumulate.

Insurance markets. Several major insurers have stopped writing new homeowners' policies in California (citing wildfire exposure) and Florida (citing hurricane exposure). State-backed insurers (the California FAIR Plan, Citizens Property Insurance in Florida) have grown substantially. Reinsurance costs have risen. The actuarial mismatch between current premium levels and current risk levels is a slow-motion fiscal crisis in several states.

Federal flood insurance. The National Flood Insurance Program (NFIP), administered by FEMA, has been technically insolvent for years, with the Treasury covering shortfalls after major hurricane events. Reform proposals — moving toward risk-based pricing — have been politically blocked by representatives of coastal districts whose constituents' premiums would rise.

Migration patterns. Survey and demographic data suggest some climate-driven migration within the U.S. is occurring, though the magnitude is contested. Wildfire and hurricane exposure correlate with out-migration in some studies; heat exposure may also correlate, though the pattern is harder to disentangle from other migration drivers.

Federal adaptation funding. The IRA and Bipartisan Infrastructure Law provided substantial adaptation-related funding (resilience, coastal protection, fire mitigation). The Trump-2 administration is rolling back some adaptation programs while continuing others.

International climate

Climate is a global commons problem, and U.S. policy operates within an international context.

The Paris Agreement (2015) is the operative international framework. Each country submits "Nationally Determined Contributions" (NDCs) describing its emissions plans, with reporting and review mechanisms but no enforcement teeth. The U.S. joined under Obama, withdrew under Trump-1 (2017), rejoined under Biden (2021), and withdrew again under Trump-2 (2025). The pendulum has produced a structural problem for U.S. climate diplomacy: other countries cannot rely on continuity in U.S. commitments, which weakens the U.S.'s ability to extract reciprocal commitments.

The deeper issue is that U.S. emissions are no longer the dominant global emissions story. U.S. emissions peaked around 2007 and have been declining slowly since, driven by the coal-to-gas shift, efficiency gains, and renewable growth. The U.S. is now responsible for roughly 11% of global emissions; China is responsible for roughly 30%, and India is rising rapidly. Global emissions continue to rise, primarily because Chinese, Indian, and developing-country emissions growth exceeds developed-country declines. Any serious assessment of global climate trajectories has to engage with what China and India do, not just what the U.S. does. American policy can model and enable the transition; it cannot, by itself, change the global trajectory.

Reform proposals

Several reform proposals organize the contemporary policy debate. Each has serious adherents; this textbook does not endorse any of them.

• Carbon tax with dividend. The Climate Leadership Council's plan and the Citizens' Climate Lobby's Energy Innovation and Carbon Dividend Act both propose a steadily rising carbon tax with revenue rebated per capita to households. Has bipartisan policy support — including some Republican former officials and some progressive economists — but limited current legislative traction.
• Green New Deal. A progressive comprehensive industrial-policy approach, emphasizing public investment, jobs guarantees, and equity. Influences the IRA's design but never passed in the original form.
• Permitting reform. A bipartisan coalition (Manchin, Schatz, Capito, Whitehouse, others) supports streamlining NEPA and federal-energy siting to accelerate both clean-energy and traditional-energy infrastructure. Steady progress but no comprehensive reform.
• Nuclear renaissance. A growing bipartisan coalition supports federal action to enable new nuclear builds — including SMRs, advanced reactors, and existing-fleet preservation. Recent legislation (the ADVANCE Act of 2024) modernizes Nuclear Regulatory Commission processes.
• Geoengineering research. Solar radiation management research (Harvard's SCoPEx, before its 2024 cancellation; ongoing research at NOAA) has serious adherents and serious critics across the spectrum.
• Adaptation focus. Some conservative analysts (Bjorn Lomborg, parts of AEI's environmental work) argue that adaptation deserves greater priority relative to mitigation, given the difficulty of cutting emissions enough to prevent significant climate impacts. Critics respond that mitigation and adaptation are complements, not substitutes, and that adaptation alone is insufficient at higher warming levels.

How environmental policy travels through the institutions

The policy approaches surveyed above — pricing, regulation, subsidy, public investment — do not act on the world directly. They are translated into outcomes through specific institutional channels, and the channel matters as much as the policy. A reader who wants to understand why a particular climate policy is or is not working has to track the policy through the institutions that implement it.

Statutory authority and its limits. The starting point for any federal environmental rule is statutory authority — a specific provision of an act of Congress. EPA cannot simply decide to regulate carbon; it has to point to language in the Clean Air Act that authorizes carbon regulation. Massachusetts v. EPA (2007) was a watershed because it held that greenhouse gases fit the Clean Air Act's definition of "air pollutant," which means EPA must regulate them once it makes an "endangerment finding" that they threaten public health and welfare. EPA made that finding in 2009. Every subsequent climate regulation under the Clean Air Act traces back to the Massachusetts holding and the endangerment finding. Trump administrations have considered, but not actually attempted, to revisit the endangerment finding — partly because the scientific record makes the finding hard to reverse on the merits, and partly because revisiting it would invite years of litigation regardless of outcome.

Notice-and-comment rulemaking. Once EPA has statutory authority, it issues a proposed rule, accepts public comments (usually 60–90 days), and issues a final rule. Major climate rules typically generate hundreds of thousands of comments — the Clean Power Plan generated more than four million — which the agency must summarize and respond to. The response-to-comments document for a major rule can run thousands of pages. The Administrative Procedure Act requires that the rule reflect "reasoned decisionmaking" and not be "arbitrary and capricious"; courts review final rules under that standard. Every major climate rule in the past two decades has been litigated, often through to the Supreme Court.

The Office of Information and Regulatory Affairs (OIRA). Inside the White House, OIRA reviews all economically significant regulations before they are published. OIRA review typically takes 90 days but can take longer for major rules. OIRA enforces the Office of Management and Budget's circulars on cost-benefit analysis (Circular A-4) and conducts inter-agency review. Climate rules generate particularly intense OIRA scrutiny because the cost-benefit numbers depend on the social cost of carbon — a parameter that ranges from roughly $7 per ton of CO2 (Trump-1's central estimate) to roughly $190 per ton (Biden-era updated estimates). The choice of social cost of carbon dominates the cost-benefit math of climate rules, and that choice has shifted substantially across administrations.

Congressional Review Act. Congress can disapprove a final rule within 60 legislative days of its publication, by simple-majority vote in both chambers, immune to the filibuster. The Trump-1 administration used the CRA to repeal more than a dozen Obama-era environmental rules in early 2017. The Biden administration used it to repeal a few Trump-1 rules in 2021. The Trump-2 administration used it to repeal Biden-era rules in 2025. The CRA is a powerful tool for the party that controls both chambers and the presidency at the start of an administration; outside that window, it has limited reach.

Judicial review. Final rules are reviewed by the federal courts of appeals, with most environmental challenges going to the D.C. Circuit. The Supreme Court grants certiorari in a small number of major cases. The doctrinal trajectory of the past two decades has constrained agency authority: West Virginia v. EPA (2022) established the major-questions doctrine; Loper Bright Enterprises v. Raimondo (2024) overruled Chevron deference, requiring courts to determine the "best" reading of statutes themselves rather than deferring to reasonable agency interpretations; Sackett v. EPA (2023) narrowed the Clean Water Act's reach. The combined effect is to make ambitious agency interpretations of older statutes substantially harder to sustain in court. Defenders of these doctrinal moves argue that they restore the constitutional separation of powers and require Congress to legislate, rather than allowing agencies to pretend old statutes authorize new programs. Critics argue that they freeze regulatory authority in a 1970s-era framework that Congress has not updated, leaving the country institutionally incapable of responding to a problem (climate change) that the original statutes did not anticipate. Both arguments have serious adherents.

State implementation and litigation. Most environmental enforcement happens at the state level under cooperative federalism, but states are also major litigants in federal environmental cases. Republican-led state attorneys general (Texas, Louisiana, West Virginia, others) routinely sue Democratic administrations' EPA rules. Democratic-led state attorneys general (California, New York, Massachusetts, others) routinely sue Republican administrations' EPA rules. The state-AG litigation track has become a major channel of policy contestation, and the choice of forum (D.C. Circuit, Fifth Circuit, Tenth Circuit) substantially shapes outcomes.

The political economy of the transition

Energy policy is unusual among policy domains in the directness with which costs and benefits land on identifiable people. A coal-plant closure puts identifiable workers out of identifiable jobs in identifiable counties. A wind farm puts identifiable wind turbines on identifiable land owned by identifiable landowners. A gas-pipeline expansion lowers identifiable home-heating bills for identifiable customers. The political economy of energy is not abstract; it is concrete, geographic, and visible.

This visibility shapes the politics in several ways.

Concentrated losses, diffuse gains. Climate policy's benefits are mostly diffuse and long-term — slightly less warming, slightly less sea-level rise, slightly less extreme weather, distributed across the global population over decades. Its costs are mostly concentrated and immediate — coal plants closed, gas prices raised, jobs displaced. Public-choice theory predicts that concentrated losses generate more political mobilization than diffuse gains, which is part of why climate policy has been politically difficult in democracies. The IRA's geographic-distribution strategy — putting clean-energy manufacturing in red districts — was specifically designed to address this asymmetry by concentrating gains in places that would otherwise see only the costs.

Energy prices as political flashpoint. Gasoline prices, electricity bills, and home-heating costs are among the most politically salient prices in the U.S. economy. Voters notice them; voters punish the incumbent party when they spike. The 2022 gas-price spike, driven primarily by Russian invasion of Ukraine and OPEC+ supply decisions, generated substantial political damage to the Biden administration despite being almost entirely outside its control. The lesson — that voters punish incumbents for energy-price spikes — has shaped what subsequent administrations are willing to attempt. It is a major reason carbon pricing has been politically difficult: any policy that visibly raises gas prices is electorally risky, even if economic theory says the price increase is the point.

Geographic concentration of fossil-fuel employment. Coal mining is concentrated in West Virginia, Wyoming, Kentucky, Pennsylvania, and a handful of other states. Oil-and-gas extraction is concentrated in Texas, North Dakota, Oklahoma, New Mexico, Pennsylvania, and Louisiana. These industries' workers and supporting communities are politically organized, geographically concentrated, and overrepresented in the Senate (the small-state advantage works for fossil-fuel-producing states). The Senate's structure means that a national policy with substantial regional costs in a small number of states faces a structurally elevated obstacle to passage.

Geographic distribution of clean-energy potential. Conversely, the best wind resources are in the Great Plains; the best solar resources are in the Southwest; the largest manufacturing capacity for batteries and EVs has been built in the Southeast. Many of these regions are politically conservative, which has produced the partial bipartisanship that surrounds the IRA — Republican governors and senators from states with substantial clean-energy investments often dissent from full party-line opposition to clean-energy spending.

The transition as labor question. What happens to workers in declining industries? Coal-mining employment in the U.S. has fallen from roughly 178,000 in 1985 to roughly 40,000 today, with most of the decline driven by mechanization and gas displacement rather than environmental policy. The communities that lost those jobs have generally not been replaced by clean-energy jobs in the same locations, partly because the geography is wrong (wind farms are sited where the wind is, not where the coal mines were) and partly because the skills do not directly translate. Federal "just transition" programs have existed since the Obama administration but have generally been underfunded relative to the scale of the dislocation. The combination of economic decline and policy emphasis on the dislocated regions has been a major theme of the politics of the past two decades.

Federalism in environmental policy

Environmental policy is one of the clearest demonstrations of American federalism in action. The federal floor sits on top of state implementation, which sits on top of local enforcement. Each layer responds to different political pressures, operates under different constraints, and produces different outcomes. Understanding the layered structure is essential to understanding why policy outcomes vary across states with similar federal frameworks.

California's outsized role. Section 209 of the Clean Air Act allows California to set vehicle-emissions standards stricter than federal standards if it obtains a waiver from EPA. This provision dates to a 1967 compromise: California already had air-quality regulations more stringent than what was about to become federal law, and Congress did not want to preempt them. The waiver provision, modest at the time, has become one of the most consequential federal-state arrangements in environmental law, because California's market is large enough that automakers cannot ignore it, and because Section 177 of the Clean Air Act allows other states to opt into California's standards (rather than choosing between California's standards and federal standards). Roughly 17 states, representing about 40% of U.S. vehicle sales, have opted in. The result is something close to a two-tier national vehicle-emissions regime, with California's standards effectively setting the bar for the larger market segment. The Trump-1 administration revoked California's waiver in 2019; the Biden administration restored it in 2022; the Trump-2 administration is revoking it again, with substantial litigation pending.

State carbon pricing. Beyond RGGI and California's economy-wide program, Washington state launched its Climate Commitment Act cap-and-invest program in 2023, after voters rejected a carbon tax in 2016 and 2018. Oregon has a Climate Protection Program (in litigation as of 2025). New York's Climate Leadership and Community Protection Act sets emissions targets but the operational pricing mechanism remains under development. Sub-national carbon pricing has moved further than federal pricing, partly because the political economy is more favorable in single-party-dominant states.

State renewable portfolio standards. Most states have some form of renewable portfolio standard or clean-energy standard, requiring utilities to source a specified fraction of generation from renewable or zero-carbon sources by specified dates. The most ambitious targets are in California (100% zero-carbon retail electricity by 2045), Hawaii (100% renewable by 2045), New York (70% renewable by 2030, 100% zero-carbon by 2040), Massachusetts, Washington, and a handful of others. These state targets, taken together, cover roughly 40% of U.S. electricity demand. They are not federal policy, but they substantially shape the U.S. electricity transition.

State-level pendulums of their own. State environmental policy is not insulated from political swings. Texas under Republican governance has been the largest single source of new wind and solar capacity in the U.S. (driven by abundant resources, cheap land, and a deregulated electricity market) while simultaneously being a major focus of conservative pushback against ESG investing and against specific renewable-energy projects. Florida under Governor DeSantis removed climate references from state law in 2024 while maintaining renewable-portfolio policies. Illinois passed the Climate and Equitable Jobs Act in 2021 with substantial Democratic majorities. State-level swings, combined with federal swings, produce a complex policy landscape in which the same renewable-energy company can face very different regulatory environments in adjoining states.

Tribal sovereignty. Federally recognized tribes have substantial authority over environmental matters within their reservations under several Clean Air Act and Clean Water Act provisions allowing tribes to be treated as states for regulatory purposes (TAS — "treatment as state"). Tribal lands include substantial fossil-fuel, mineral, and renewable-energy resources, and tribal-government decisions on extraction, leasing, and transmission have been increasingly important. The Standing Rock protests against the Dakota Access Pipeline (2016–17) brought tribal-sovereignty environmental issues to national attention; the legal questions remain partially unsettled.

What does not get measured

A recurrent challenge in environmental and energy policy is that what gets measured shapes what gets managed, and substantial environmental impacts fall outside the measurement frameworks of major statutes.

Cumulative impacts. Most environmental statutes regulate individual sources — a specific factory, a specific power plant, a specific facility. They do not naturally aggregate cumulative impacts on a community from multiple sources. New Jersey's 2020 environmental-justice statute is one of the few legal frameworks that requires permit decision-makers to consider cumulative impact. Most other regulatory frameworks consider new sources in isolation, which means a community already overburdened by existing sources can receive permits for additional sources because each new source, considered alone, meets standards.

Indoor air quality. Outdoor air quality is heavily regulated under the Clean Air Act. Indoor air quality, where Americans spend roughly 90% of their time, is largely unregulated at the federal level. Indoor air-quality issues — radon, gas-stove combustion byproducts, mold, formaldehyde from building materials — receive much less institutional attention than outdoor pollution despite arguably comparable health impacts.

Embedded emissions. A substantial fraction of the emissions associated with U.S. consumption occurs outside U.S. borders, in the manufacturing of products imported into the United States. Carbon-border adjustment proposals — fees on imports based on their embedded emissions — would address this gap. The European Union has implemented a Carbon Border Adjustment Mechanism (CBAM) starting in 2026. U.S. proposals (the Foreign Pollution Fee Act introduced by Senator Cassidy and others) have bipartisan but limited traction.

Methane leaks. The conventional accounting for natural gas's emissions advantage over coal assumes methane leak rates of perhaps 1–2% across the gas supply chain. Direct measurement using satellite, aircraft, and on-the-ground methods has consistently found higher leak rates, sometimes substantially higher in specific basins. The actual climate footprint of U.S. natural gas depends sensitively on the leak rate, and the leak rate has been chronically under-measured.

Forest, soil, and ocean carbon. Land-use change, agricultural practices, and ocean carbon dynamics interact with climate in ways that are scientifically substantial but policy-marginal in U.S. environmental law. Forest fires release substantial CO2 (and the U.S. fire frequency has been rising); agricultural practices affect soil carbon; ocean carbon dynamics affect both atmospheric concentrations and ocean acidity. None of these falls cleanly within the major federal environmental statutes.

The gap between what is measured and what matters is part of why the policy debate often feels disconnected from the underlying physical reality. Measurement creates the categories the policy system can act on; un-measured categories tend to fall through the system.

Specific live policy fights, 2024–2025

The chapter has surveyed the institutional architecture and the policy categories. To make the discussion concrete, here are several specific live fights as of 2025.

EV mandates and consumer credits. The Biden EPA's 2024 vehicle rules effectively required substantial EV market share by 2030. The Trump-2 administration is rolling them back. The IRA's $7,500 EV consumer credit was substantially repealed by the One Big Beautiful Bill Act in 2025. Several states (California most prominently, joined by states that opt into California's standards under the Clean Air Act waiver) have their own EV mandates, which the Trump-2 administration is challenging legally. The litigation over California's waiver is one of the central environmental cases of 2025.

Power-plant rules. Biden's 2024 power-plant rule, drafted to fit within the post-West Virginia doctrinal frame, is being rolled back. The Trump-2 EPA has proposed alternative rules that are substantially less restrictive. The litigation track is parallel: industry groups challenging the Biden rules; environmental groups challenging the Trump-2 rules.

Methane rules. Biden's 2024 methane rules for oil and gas operations are being rolled back. The IRA's methane fee — a per-ton charge on methane emissions from large oil-and-gas facilities — is being administratively delayed.

Permitting and transmission. The need for substantially expanded electricity transmission to accommodate higher renewable shares is widely acknowledged across the political spectrum. The Department of Energy's 2024 transmission siting authority — a new federal authority created under the Bipartisan Infrastructure Law — is being implemented. Federal interregional transfer capacity targets are under FERC consideration. State-level opposition to transmission siting (often driven by local environmental, aesthetic, or property-rights concerns rather than partisan politics) remains a binding constraint.

Federal lands. ANWR oil-and-gas lease sales are being expanded under Trump-2. Coal-leasing constraints are being lifted. National monument boundaries are being reviewed for potential reduction.

Critical minerals. Lithium, cobalt, nickel, rare-earth minerals, and other critical minerals are required for batteries, magnets, and other clean-energy technology. The U.S. has very limited domestic supply for most of them. Both Trump and Biden administrations have pursued domestic mining and processing expansion (Inflation Reduction Act tax credits, Defense Production Act invocations, lease approvals on federal land), partly to reduce dependence on Chinese supply chains. Environmental groups have generally opposed specific mining projects (the Resolution Copper project in Arizona, the Pebble Mine in Alaska, lithium projects in Nevada) while supporting the broader clean-energy transition that depends on these minerals — a tension within environmental coalitions that is unusually visible.

Nuclear. Bipartisan support for nuclear has grown substantially. The ADVANCE Act of 2024 streamlined Nuclear Regulatory Commission processes. Three Mile Island Unit 1 (now Crane Clean Energy Center) is being restarted under a Microsoft power-purchase agreement — symbolic of the data-center demand pull on firm zero-carbon power. SMR projects (NuScale's project in Idaho was canceled in 2023; X-energy, TerraEnergy, Kairos, and others are advancing) are at varying stages of development.

Carbon capture. The IRA's expanded 45Q tax credit substantially raised the dollar amount per ton of CO2 captured. Direct air capture demonstration projects (Occidental's Stratos in Texas, Climeworks' projects, others) are being built. Critics argue that DAC is too expensive at scale to be a primary mitigation strategy and that subsidies risk supporting an indefinite continuation of fossil-fuel use. Defenders argue that some hard-to-abate sectors (cement, steel, aviation) likely require carbon capture and that the technology curve should be ridden down.

Conclusion

American environmental and energy policy is a system that handles a settled scientific question (climate change is real and human-caused), a partially settled technological question (the energy transition is happening), and a deeply contested political question (what to do, how fast, who pays, and at what trade-off with other goals). The institutional architecture is large but not well-designed for the climate problem as it has materialized. The major statutes pre-date the climate problem. The pendulum between administrations has produced regulatory whipsaw. The economic transition is moving faster than the political process. The international dimension complicates everything.

The chapter's running theme — the gap between how the system was designed to work and how it works in practice — is on full display. The system was designed to handle local-pollution problems with bipartisan political consensus. It is being asked to handle a global, century-scale, deeply partisan problem. It is doing so imperfectly. Reasonable people disagree about how to make it work better.

The chapter asks readers to take the disagreement seriously, on the strongest version of each side's case, and to bring the empirical evidence to bear without pretending the empirical evidence resolves the values disagreements. It asks them to notice what the institutions can and cannot do — and to ask what would change if Congress passed a major new environmental statute for the first time in 35 years, or if it didn't. It asks them to track the energy transition that is happening for reasons of cost and physics, and to ask which policies speed or slow that transition, which redistribute who profits and who pays, and which simply trade rhetorical positions while the underlying technology and economics roll on. The Camry in Atlanta will keep emitting until the Camry's owner replaces it. The atmosphere does not vote, but the people who decide whether to replace the Camry — and what they replace it with — do. The chapter is about the institutions that shape that decision, and the contested politics that shape those institutions.