You Drowned My Country. Now Send a Cheque.

It takes a particular kind of institutional courage to announce, with great fanfare, that you will compensate people for harm you caused — and then spend the next four years perfecting the art of not actually writing the cheque. At COP27 in Sharm el-Sheikh in November 2022, the world's wealthiest nations agreed to establish a Loss and Damage fund to pay climate reparations to the countries least responsible for warming but most vulnerable to its effects. The agreement was hailed as historic. Activists wept. Diplomats embraced. The fund was real, they said. Justice was coming.

Four years later, in April 2026, the Loss and Damage fund has disbursed precisely zero dollars to any country. Not to Tuvalu, which will disappear beneath the Pacific by mid-century. Not to Bangladesh, where 30 million people have been displaced by flooding since 2020. Not to the Maldives, whose capital Malé is protected by a seawall that cost more than the nation's annual GDP. The fund exists. It has a governing board, a secretariat, and a headquarters at the World Bank. What it does not have is money going out the door.

This is not, of course, without precedent. History offers a useful analogue: German reparations after the First World War. The Treaty of Versailles in 1919 established the principle that the defeated power would pay for the damage it caused. The Reparations Commission set the bill at 132 billion gold marks — roughly $442 billion in today's money. Germany paid approximately 20 billion marks before the entire system collapsed in the hyperinflation of 1923. The Dawes Plan restructured the debt. The Young Plan restructured it again. By 1932, reparations were suspended indefinitely. Germany made its final payment in 2010, ninety-one years after the armistice.

One is tempted to observe that climate reparations are following a similar trajectory, except in this case the perpetrators are the victors, the victims have no army, and there is no treaty to enforce. The analogy is imperfect but instructive. Reparations work when the creditor can compel payment. They fail when payment is voluntary and the debtor controls the timetable.

The Argument They Haven't Made

The case for climate reparations rests on three claims, each of which is factually unassailable. First: the nations that industrialised earliest are responsible for the majority of cumulative carbon emissions. The United States and the European Union together account for 47 per cent of all CO₂ emitted since 1850, according to Carbon Brief's 2023 analysis. China, despite its recent surge, accounts for 14 per cent. Sub-Saharan Africa accounts for less than 1 per cent.

Second: the countries least responsible for warming face the greatest harm. A 2021 study in Nature Climate Change found that low-income countries will bear 75 per cent of the economic costs of climate impacts despite producing only 10 per cent of emissions. Pakistan's 2022 floods — which displaced 33 million people and caused $30 billion in damage — were made 50 per cent more likely by human-caused warming, according to World Weather Attribution. Pakistan's per capita emissions are one-tenth those of the United States.

Third: the damage is not theoretical or distant. It is measurable, present, and accelerating. The Global Centre for Climate Mobility estimates that 21.5 million people are displaced annually by climate-related disasters. The World Bank projects that internal climate migration could reach 216 million by 2050. Small island developing states face not economic loss but territorial extinction. The Intergovernmental Panel on Climate Change's 2023 Sixth Assessment Report states with high confidence that some low-lying islands will become uninhabitable before 2100 even if warming is limited to 1.5°C.

The logic is straightforward: if you break it, you buy it. If your wealth was built on a carbon-intensive industrial revolution that is now rendering entire nations uninhabitable, you owe compensation. This is not radical. It is the principle underlying tort law, environmental liability, and the polluter-pays doctrine enshrined in the 1992 Rio Declaration. The novelty is not the argument but the scale.

The Counterargument, Steelmanned

Wealthy nations have not openly refused to pay reparations. That would be politically untenable. Instead, they have deployed a series of procedural and conceptual objections that sound reasonable until examined closely. The strongest version of their case runs as follows:

First, historical emissions occurred before the climate science was settled. It is unjust to assign liability for actions taken in ignorance. Svante Arrhenius predicted greenhouse warming in 1896, but the scientific consensus did not solidify until the 1980s. Should Britain pay reparations for coal burned in 1880?

Second, the largest emitter today is China, which is classified as a developing country and therefore eligible to receive climate finance rather than provide it. China's annual emissions surpassed those of the United States in 2006. By 2025, China had emitted more CO₂ cumulatively than the EU. If reparations are about present harm, why is the world's second-largest economy exempt?

Third, adaptation finance and development aid already flow from rich to poor countries. The Green Climate Fund, established in 2010, has mobilised $13.5 billion. Bilateral adaptation funding reached $30 billion in 2023. Is this not compensation by another name?

Finally, there is the matter of attribution and scale. How do you calculate loss and damage for a hurricane intensified by warming, a drought exacerbated by shifting rainfall patterns, or a heatwave made more frequent by 1.2°C of warming? And even if you could, the bill would be unpayable. One 2022 estimate by the Loss and Damage Collaboration put annual climate damages in developing countries at $400 billion by 2030. Total development aid from OECD countries in 2023 was $223 billion. The money does not exist.

These objections are not without merit. But they collapse under scrutiny. Yes, early industrial emissions occurred before the science was settled — but emissions since 1990, when the first IPCC report was published, account for more than half of all cumulative CO₂. Ignorance is not a defence for the past three decades. Yes, China is now the largest annual emitter — but its per capita emissions remain far below those of the United States, and the cumulative stock of CO₂ in the atmosphere, which drives warming, is still dominated by Western emissions. Yes, adaptation finance exists — but it is loans, not grants, and it funds future resilience, not compensation for past harm. And yes, the bill is large — but so was the Marshall Plan, which transferred 5 per cent of U.S. GDP to rebuild Europe. The United States is far wealthier now than it was in 1948.

What the Fund Has Actually Done

Since its establishment in November 2022, the Loss and Damage fund has achieved the following: a transitional committee met five times in 2023 to draft governance structures. At COP28 in Dubai in December 2023, the fund was formally operationalised. Pledges totalling $792 million were announced by the EU, United States, UAE, and several smaller donors. The World Bank was designated as interim host for four years. A 26-member governing board was constituted, with representation from developed and developing countries. The board held its first meeting in March 2024. By April 2026, the fund has approved project guidelines, established fiduciary standards, and issued a call for proposals.

What it has not done: distributed a single dollar. The $792 million pledged has not been fully deposited. The United States pledged $17.5 million but appropriated only $10 million in the 2025 budget; the remainder is pending Congressional approval. Several European pledges are conditional on governance reforms that have not been implemented. The fund's secretariat has no staff beyond seconded personnel. The project pipeline is empty. Even under optimistic assumptions, the first disbursements will not occur before 2027 — five years after the agreement.

This is where the wit column traditionally delivers the punchline. But the punchline here is not funny. It is this: the fund is designed to fail slowly enough that no one can be accused of killing it, but comprehensively enough that it will never fulfil its mandate. The bureaucratic infrastructure ensures process without outcome. The pledges ensure headlines without accountability. The timeline ensures that by the time the fund is operational, the damage will have compounded beyond what any realistic level of funding could address.

The Precedent We've Forgotten

There is, in fact, a model for large-scale international transfers based on historical harm. It is not the Treaty of Versailles. It is the agreement that ended it. After the Second World War, the Allies did not repeat the mistake of punitive reparations. Instead, they designed the Marshall Plan: $13 billion in grants and loans to rebuild Europe. Adjusted for inflation and GDP, that would be equivalent to $1.4 trillion today. The programme ran for four years, from 1948 to 1952. It worked because it was adequately funded, quickly disbursed, and clearly in the strategic interest of the United States, which wanted stable democracies and open markets in Europe.

Climate reparations lack that strategic clarity. The harm is diffuse, the victims are distant, and the political constituency in donor countries is weak. But the analogy holds in one respect: the cost of inaction exceeds the cost of action. Europe in ruins would have been a breeding ground for extremism and instability. A world in which 200 million people are displaced by climate impacts by mid-century will produce the same. The question is not whether we can afford reparations. It is whether we can afford the alternative.

What Should Happen

The first step is to stop pretending that $792 million is a serious response to $400 billion in annual damages. The Loss and Damage fund needs to be capitalised at a scale commensurate with the harm: at minimum, $100 billion annually by 2030, rising to $300 billion by 2040. This is not a fantastical number. It is less than 0.1 per cent of combined OECD GDP. It is less than global military spending increases in 2024 alone.

The second step is to fund it through mechanisms that do not rely on annual appropriations. A carbon tax on international aviation and shipping — sectors exempt from the Paris Agreement — could raise $50 billion annually. A financial transactions tax on fossil fuel trading could raise another $30 billion. Redirecting just 10 per cent of fossil fuel subsidies would raise $700 billion. The money exists. What is missing is political will.

The third step is to recognise that this is not charity. It is not even justice in the abstract. It is strategic self-interest. Climate displacement drives migration, which destabilises regions, which produces conflict, which costs far more than prevention. The European Union spent €21 billion on migration and border control in 2023. The United States spent $18 billion. That money bought walls, detention centres, and coast guard patrols. It did not address the root cause. Climate reparations would.

The Closing Argument

It is worth noting, in closing, what the world looks like when reparations are not paid. Germany's failure to meet its Versailles obligations contributed to economic collapse, political radicalisation, and the conditions that produced the Second World War. The Allies learned from that mistake. They chose reconstruction over retribution, investment over punishment, and pragmatism over moral satisfaction. It worked.

Climate reparations offer a similar choice. Pay now, through a managed, multilateral process that addresses harm and builds resilience. Or pay later, through border enforcement, disaster response, and the political and humanitarian costs of mass displacement. The bill is coming either way. The only question is whether we write the cheque ourselves or wait for history to cash it.

Four years after the Loss and Damage fund was agreed, the people of Tuvalu, Bangladesh, and the Maldives are still waiting. Their patience, one suspects, is not infinite. Ours should not be either.