Leaving Paris Twice and What the US Withdrawals Changed

On January 20, 2025, a few hours after taking the oath of office for the second time, Donald Trump sat at a desk set up in a downtown Washington arena in front of a cheering crowd and signed a stack of executive orders, handing the pens out to supporters as he went. One of them directed the United States to withdraw from the Paris Agreement on climate change. It was the second time the same president had pulled the country out of the same treaty. The first exit had taken more than three years to become official; this one was set to take effect in one. Climate advocates braced for a repeat of 2017, when the announcement landed like a verdict on the planet. What actually followed is more complicated, and it says a good deal about where power over American emissions really sits.

In short: The United States left the Paris Agreement in 2020, rejoined in 2021 and began leaving again on January 20, 2025. My argument is that each exit hurt, especially on diplomacy and climate finance, but that states, cities, utilities and the economics of clean power have limited how much a presidential walkout can change what actually comes out of American smokestacks and tailpipes.

The first exit, in slow motion

The first withdrawal was announced in the Rose Garden on June 1, 2017. The president said the agreement disadvantaged the United States and that he was elected to represent the citizens of Pittsburgh, not Paris. The line got a quick response from Pittsburgh's mayor, Bill Peduto, who noted that his city had voted heavily for Hillary Clinton and that he intended to follow the Paris goals anyway.

What the Rose Garden announcement did not do was take the country out of the treaty. The agreement's own withdrawal clause, Article 28, says a party cannot give notice of withdrawal until three years after the agreement entered into force for it, and that the withdrawal takes effect one year after notice. Paris had entered into force on November 4, 2016. So the earliest possible notice date was November 4, 2019, and the earliest possible exit was November 4, 2020.

The administration filed on the first day it could. The exit became official on November 4, 2020, the day after a presidential election that, as it turned out, the president had lost. For a little over three months the United States was formally outside the agreement. Then, on his first day in office, January 20, 2021, President Biden signed the instrument to rejoin, and the country was back as a party on February 19, 2021.

So the first walkout looked enormous and lasted, in legal terms, about a hundred days. The three-year waiting period, whatever its drafters intended, turned a dramatic announcement into a long countdown that an election interrupted.

The second exit, and what was different

The second time, there was no waiting period to run out. The United States had rejoined in 2021, well more than three years earlier, so notice could be given at once and the one-year clock started immediately. The withdrawal took effect in early 2026.

The second exit also came alongside much more aggressive domestic action. In 2017 the main federal tools were regulatory rollbacks, which moved slowly through the courts. In 2025 the administration paired the Paris order with an energy emergency declaration, a push to expand oil and gas leasing, efforts to halt offshore wind projects, the cancellation of grants and loans, and moves to unwind the scientific finding that underpins federal greenhouse gas rules. Then Congress passed and the president signed a budget law on July 4, 2025, that cut back many of the clean energy tax credits the 2022 climate law had created.

Here is where I want to be careful, because the honest version of this argument has to start with the damage. The second withdrawal took the world's largest historical emitter, and its second-largest current one, out of the negotiating room. The United States had been a major contributor of international climate finance, and that money has been cut sharply. Diplomats from other countries no longer have an American partner to bargain with or pressure. The country's own pledge, submitted in late 2024, called for cutting net emissions by 61 to 66 percent below 2005 levels by 2035, and the federal government no longer claims to be pursuing it.

None of that is small. But the treaty was never what made American emissions fall. Something else was.

Who actually controls American emissions

Ask what determines how much carbon dioxide the United States puts into the air, and the answer is not mainly a treaty. It is which power plants run, what cars people drive, how buildings are heated and what factories make. Those decisions are spread across fifty state governments, thousands of utilities, regional grid operators, automakers and hundreds of millions of households.

States hold much of that power directly. They regulate the utilities that own most power plants. Many of them set renewable or clean electricity standards that require utilities to buy a rising share of power from clean sources. A group of northeastern states has run a cap-and-trade program for power plants, the Regional Greenhouse Gas Initiative, since 2009. California runs its own economy-wide cap-and-trade system. States write building codes, set vehicle registration fees, fund transit and approve or block new pipelines.

On the same day as the first Rose Garden announcement in 2017, the governors of California, New York and Washington formed the US Climate Alliance, a group of states pledging to meet the Paris goals on their own. It grew to include governors of both parties and outlasted the first withdrawal. In 2025 its members again said they would keep working toward the goals. Cities and businesses made similar declarations.

It is fair to be skeptical of pledges, and I will get to that. But the jurisdictions in that alliance together account for a large share of the US economy and population, and they have laws on the books, not just press releases. When a governor signs a clean electricity standard into state law, a president cannot repeal it with an executive order.

The market part of the argument

The second limit on a walkout is economics, and the honest version of this point has grown stronger since 2017.

Over the last decade the cost of solar panels, wind turbines and batteries has fallen dramatically. In much of the country, new wind and solar power is among the cheapest electricity a utility can build. That is why some of the biggest growth in clean power has come in places with no particular interest in climate politics. Texas, run by Republicans for a generation, generates more wind power than any other state and has added solar and battery storage at a remarkable pace, largely because its grid rewards whoever can produce power cheaply.

When a utility decides to build solar instead of gas, it is usually responding to fuel costs, financing costs and what its regulators will approve, not to whether the United States is a party to a UN agreement. The clean power by state figures show how differently states have moved, and how little that movement tracks the partisan map.

Now, the strongest counterargument. The market case can be overstated. Clean energy got cheap partly because of public policy: tax credits, state mandates and research money. The 2025 budget law cut a large part of that support. Cheap solar is no help if a project cannot connect to the grid, cannot get a permit, or loses the tax credit that made its financing work. Electricity demand is rising again because of data centers and new factories, and some utilities are keeping coal plants open longer or planning new gas plants. Federal agencies can make life very hard for specific projects, as offshore wind developers learned in 2025 when work was halted on projects that already had their permits.

All of that is true. The argument is not that federal policy does not matter. It is that the federal government's ability to stop the transition is much weaker than its ability to speed it up. Removing subsidies slows things down. It does not make a coal plant cheaper to run than a solar farm in a place where it is not.

Consider what a president can and cannot reach. He can direct federal agencies to stop issuing grants, to slow permits on federal land and waters, to rewrite pollution rules and to lean on the grid operators the federal government regulates. He cannot order a privately owned utility in Iowa to stop signing wind contracts, tell a school district in Colorado to tear out the heat pumps it installed, or cancel the long-term power purchase agreements that technology companies have signed for solar and wind farms. Those contracts run for years and are enforceable in court. Each one is a small piece of the transition that sits outside his pen's reach.

There is also a time dimension. Power plants are built to run for decades, and the companies that build them plan around the next twenty or thirty years, not the next four. A utility executive who has watched federal climate policy flip three times since 2017 has every reason to discount whatever Washington says this year and to bet on whatever looks cheapest over the life of the plant. Political whiplash, oddly, can push long-term planners toward the options that do not depend on Washington at all.

Where blue states have their own problems

If states are the backstop, progressives should be honest about how well the backstop works. It is uneven, and some of the failures are self-inflicted.

States that pass ambitious clean energy targets often make it very hard to build what those targets require. Transmission lines take a decade or more to approve. Local zoning boards block solar farms and wind projects, sometimes with the support of the same residents who vote for climate candidates. Environmental review laws that were written to stop polluters are now used to delay clean energy projects too. California, New York and Massachusetts have all struggled to build fast enough to meet their own goals.

Electricity prices are another vulnerability. Several blue states have some of the highest power bills in the country. Some of that comes from policy choices, some from geography, some from aging grids and the cost of wildfire and storm damage. Whatever the cause, high bills make climate policy easier to attack, and they fall hardest on working families. A climate agenda that cannot keep the lights affordable is one that will be voted out.

The fix is not to abandon the goals. It is to treat building as the goal: faster permits for transmission and clean power, simpler interconnection rules, public financing where private capital is slow. You can follow how states have used their own power against federal rollbacks on the blue states versus Trump page.

What the walkouts taught

There are three lessons I would draw from leaving Paris twice.

The first is about durability. An executive agreement can be undone by the next executive. Climate policy that lives only in presidential orders is fragile; policy written into statute, state law and private contracts is much harder to reverse. The 2022 climate law turned out to be partially reversible too, but the factories it helped finance, many in Republican districts, created a constituency that defended pieces of it.

The second is about who the climate coalition includes. The strongest defense of clean energy in 2025 often came not from environmental groups but from manufacturers, construction unions, farmers with wind leases and Republican members of Congress whose districts had new battery plants. That coalition is broader than the one that cheered in Paris in 2015, and it is worth building on rather than lecturing.

The third is about the world. Other countries did not follow the United States out. The European Union, China, India, Brazil and nearly every other party stayed in. The global clean energy industry, from Chinese solar factories to European wind manufacturers, kept growing. An American withdrawal changes the country's influence over that industry more than it changes the industry. The risk for Americans is less that the transition stops than that it happens with other countries making the products and holding the jobs.

What you can do

The exits moved the fight to statehouses, utility commissions and the next Congress, so that is where your effort goes furthest.

• Find out who decides your state's energy rules. State legislators write clean energy standards and fund building programs. Use the representatives lookup to find your state lawmakers alongside your members of Congress, with their phone numbers and latest votes.
• Show up where power plants get approved. Public utility commissions hold hearings on rate cases and resource plans, and they hear from very few ordinary customers. Local groups often help people prepare testimony; the organizations directory lists environmental and consumer groups in every state.
• Comment on the federal rollbacks. Rules that weaken power plant and vehicle standards must go through a comment process, and the record those comments create matters when the rules are challenged in court. The comment page lists open federal rules, closing soonest first, with how to write one.
• Keep an eye on the courts. Many of the 2025 actions against clean energy projects and climate rules are being litigated. Court Watch follows the Supreme Court's opinions this term, its emergency docket and Senate votes on judges.

A president can take the United States out of a treaty with a pen. He cannot, with the same pen, make the wind stop blowing across the Texas panhandle or repeal a law passed in Sacramento. The question for the next few years is whether the people who run those places keep building.