Blog
How the Paris Agreement Actually Works, From the Gavel at Le Bourget
Why the Paris climate deal runs on national pledges, a five-year ratchet and public scrutiny instead of penalties, and what that design can and cannot do.
On the evening of December 12, 2015, in a convention hall at Le Bourget, the old airfield on the northern edge of Paris where Charles Lindbergh landed in 1927, France's foreign minister Laurent Fabius looked out at delegates from nearly every country on Earth, asked whether there were objections, heard none, and brought down his gavel. The hall erupted. People who had spent two decades in windowless negotiating rooms hugged strangers. A few cried. What they were cheering was a document that commits no country to any particular cut in its emissions and contains no fines, sanctions or trade penalties for falling short. That sounds like a failure dressed up as a party. It was a deliberate design, and to understand climate politics in 2026 you have to understand why the people in that hall thought a treaty without teeth was the only kind that could work.
In short: The Paris Agreement asks every country to set its own climate target, report honestly on progress and raise its ambition every five years. It has no penalties because the earlier, binding approach collapsed, and because the US Senate would never ratify a treaty with them. Its power comes from transparency, peer pressure and domestic politics, which means the real enforcement happens at home.
The treaty that came before, and why it broke
To see why Paris looks the way it does, start with what it replaced. The Kyoto Protocol, adopted in 1997, took the opposite approach. It set legally binding emissions targets for wealthy, industrialized countries and set none for developing ones, including fast-growing economies such as China and India. The logic was historical fairness: the countries that had filled the atmosphere with carbon for two centuries should move first.
The politics did not survive contact with Washington. Months before Kyoto was finished, the US Senate passed the Byrd-Hagel resolution by 95 to 0, warning that it would not accept any treaty that bound the United States while exempting major developing economies, or that would seriously harm the American economy. The Clinton administration signed Kyoto anyway but never sent it to the Senate for a vote it was certain to lose. The George W. Bush administration walked away entirely in 2001.
Kyoto limped on, but the arithmetic was brutal. As China's emissions overtook those of the United States in the mid-2000s, a treaty that covered only rich countries covered a shrinking share of the problem. The attempt to write a successor at Copenhagen in 2009 ended in a chaotic, last-minute political statement that the conference only "took note" of rather than adopting. Many negotiators came home from Copenhagen convinced that the top-down model, in which diplomats agree on a global number and divide it up among countries, was dead.
Paris was built on its grave. Instead of handing countries targets, it asked each one to write its own.
Pledges written at home
The heart of the agreement is a clumsy phrase: the nationally determined contribution, universally shortened to NDC. Under Article 4, every party prepares, communicates and maintains a plan describing what it will do to cut emissions. The country decides the target, the base year, the sectors covered and the policies it will use. A small island state and a superpower file the same kind of document.
What the treaty binds countries to is process, not outcome. A country must submit an NDC. It must pursue domestic measures aimed at achieving it. It must report. It must come back with a new one on schedule. What it does not legally have to do is hit the number it wrote down.
That distinction was so carefully guarded that one word nearly sank the whole deal on the final day. A late draft said developed countries "shall" undertake economy-wide emission reduction targets. In treaty language, "shall" creates an obligation. American negotiators warned that the word could turn the agreement into one requiring Senate consent, and the text was changed to "should" in what was presented as a technical correction just before the gavel came down. One verb, four letters, and a global deal hung on it.
There is a reason the US cared so much. The Obama administration intended to join Paris as an executive agreement, relying on existing presidential authority and existing law, without a two-thirds Senate vote that was never coming. That choice made American participation possible in 2016. It also made American withdrawal easy later, a trade-off that would matter twice.
The overall goal, in Article 2, is to hold the increase in global average temperature to "well below" 2 degrees Celsius above pre-industrial levels and to pursue efforts to limit it to 1.5 degrees. The 1.5 figure was there because small island and low-lying nations, for whom 2 degrees could mean the loss of their territory, fought for it. It was the most ambitious thing in the text, and it belonged to the countries with the least power at the table.
The ratchet: why five years matters
If countries set their own targets, what stops them from setting easy ones? The first answer is that the first round of pledges was, in fact, too weak. Everybody knew this in 2015. Analyses at the time found that even if every country met its first pledge, the world would still be on track for warming well above 2 degrees. Paris did not hide that gap. It built a machine to close it.
That machine is usually called the ratchet. It has three parts.
First, countries must submit a new NDC every five years, and each successive one is expected to represent a progression beyond the last, reflecting the country's "highest possible ambition." A ratchet turns one way. Backsliding is not formally banned, but it is meant to be politically embarrassing.
Second, there is the global stocktake, set out in Article 14. Every five years the parties assess collective progress toward the long-term goals, and the result is meant to inform the next round of pledges. The first stocktake concluded at the climate conference in Dubai in 2023, and its outcome called on countries to transition away from fossil fuels in energy systems, the first time a decision under the UN climate process had named fossil fuels in that way. That language was a compromise, and many countries wanted something stronger, but it was a line that had not existed before.
Third, there is the enhanced transparency framework in Article 13. Countries report their emissions and their progress under common rules, and those reports go through technical expert review and a public, multilateral exchange where other governments can ask questions. The first biennial transparency reports under the framework were due at the end of 2024.
Put those together and you get the theory of change: commitments that start low but must rise on a fixed calendar, measured with common rules, assessed together in public. The pressure is meant to come from comparison, from investors and voters reading the reports, and from the simple discomfort of standing up every five years and explaining yourself.
Why there are no penalties
Here is the part that frustrates people. Article 15 creates a committee to facilitate implementation and promote compliance. The text describes it as expert-based and facilitative, operating in a manner that is transparent, non-adversarial and non-punitive. If a country misses its target, there is no fine, no suspension, no trade sanction. The committee can talk to the country about why.
It is worth stating the strongest case against this design, because it is a serious one. Critics argue that a treaty without consequences invites free riding: every country benefits when others cut emissions, so every country is tempted to let others do the cutting. Some economists have proposed "climate clubs," groups of countries that commit to real carbon pricing and impose border charges on goods from countries that do not. The European Union's carbon border adjustment mechanism is a step in that direction. On this view, Paris is a confession that the world could only agree on the things that cost nothing.
Here is the answer, and I think it is the stronger one. Binding targets were tried, and the result was not tougher action but a treaty that the largest historical emitter refused to ratify and the largest current emitter was never asked to join. A penalty that the biggest players will not accept is not a penalty; it is a reason to stay out. Paris traded legal force for nearly universal participation, and participation is what makes the other mechanisms work. You cannot pressure a country through peer review if it is not in the room.
There is also a quieter point. International law has few ways to punish sovereign states for anything. Even trade agreements with formal dispute systems rely heavily on states choosing to comply. Climate action depends on thousands of domestic decisions about power plants, buildings, cars and farms. No international committee can make those decisions. Parliaments, ministries, courts and voters can. Paris was built on the bet that the real enforcement would happen at home, through domestic laws that turn pledges into obligations people can sue over and vote on.
The money and the people at the edges
Paris is not only about emissions. It also deals with adaptation, with finance and with harm already done, and this is where its human-rights side comes into view.
The preamble acknowledges that parties should respect, promote and consider their obligations on human rights, the right to health, the rights of Indigenous peoples, local communities, migrants, children, persons with disabilities and people in vulnerable situations, along with gender equality and intergenerational equity. It is preamble language, not an operative obligation, but it was fought for, and it gives advocates a text to point to.
Article 9 says developed countries shall provide financial resources to help developing countries with both cutting emissions and adapting. Rich countries had promised in 2009 to mobilize $100 billion a year by 2020 for climate action in developing nations, a pledge they were widely reported to have missed on time. At the 2024 conference in Baku, governments agreed on a new goal of at least $300 billion a year by 2035. Many developing countries called that figure far too low, and they said so loudly on the conference floor.
Article 8 recognizes loss and damage, the harm from climate impacts that cannot be adapted to: a coastline that is gone, a harvest that will not come back. For years wealthy countries resisted any mechanism that might sound like liability. At the 2022 conference in Egypt, they agreed to create a fund for loss and damage, and it was made operational the following year. The amounts pledged so far are small next to the need, but the principle that people living with the worst effects deserve help was finally written down.
This is the moral center of the agreement, and it is easy to lose in arguments about targets. The people who did least to cause the problem, farmers in the Sahel, fishing families in Bangladesh, residents of atolls in the Pacific, are the ones living with it first. Paris treats their governments as full parties. Whether the money follows is a political question, and it is answered in the budgets of countries like ours.
What the design means for Americans now
The agreement entered into force on November 4, 2016, after enough countries had joined to cross a double threshold: at least 55 parties accounting for at least 55 percent of global emissions. The United States and China announced their joint ratification together in September 2016, which helped clear that bar fast.
Because the treaty's force runs through domestic politics, its fate in any one country depends on who wins elections there. The United States has now left twice, under the same president, and that story deserves its own telling. But the structure explains why the exits did less damage than many feared. When Washington walked out, the pledges of other countries stayed in place, the reporting cycle continued and the five-year calendar kept turning. COP30 met in Belém, Brazil, at the edge of the Amazon, in November 2025, with the United States federal government sending no high-level delegation.
The structure also explains why progressives cannot treat Paris as a substitute for winning arguments at home. The treaty gives an ambitious American government a framework and a deadline. It cannot force an unwilling one to do anything. The power to turn a pledge into falling emissions lies with Congress, with federal agencies writing rules, with states setting their own standards and with the utilities and companies that respond to all of the above. You can follow the domestic fights over climate policy and see that pattern repeat.
And the design carries a lesson about persuasion. Paris works only to the extent that people inside each country believe the pledges matter and hold their leaders to them. A farmer in Iowa selling wind leases, a union electrician installing heat pumps, a mayor in Phoenix dealing with record heat: those people are the enforcement mechanism. They do not have to call themselves environmentalists. They have to see the stake.
What you can do
The agreement's real teeth are domestic, so the useful moves are close to home.
- Watch the numbers the reports are built on. The climate data page tracks carbon dioxide at Mauna Loa, the global temperature record, the US power mix and clean power by state. When someone says the transition has stalled or is unstoppable, you can check which direction your own state is moving.
- Weigh in on the federal rules that turn pledges into policy. Power plant standards, vehicle rules and methane limits all go through public comment. The open comment periods page lists federal rules accepting comments now, soonest deadline first, grouped by issue, along with guidance on writing a comment that an agency has to consider.
- Ask your members of Congress where they stand on climate finance. International climate money comes from appropriations bills, and those are written by people you can call. You can look up your representative and senators by address, see their recent votes and reach their offices directly.
- Support groups doing adaptation and justice work. Organizations working on environmental justice, disaster recovery and immigrant communities often handle the human side of climate impacts. The organizations directory lists nonprofits by cause in every state.
The gavel at Le Bourget did not save the climate, and nobody in that hall thought it would. It set a clock. Every five years the world has to stand up, show its work and say whether it can do better. Whether the answer is yes depends less on diplomats than on the rest of us.