At COP31 in Antalya, Turkey, carbon markets are being presented as one way to increase climate finance for developing countries. Turkish Environment Minister Murat Kurum, who is leading the summit, says they could raise more than US$50 billion a year in extra finance by 2030.

The United Nations now expects global warming to exceed the 1.5 C target in the 2015 Paris Agreement, increasing future risks from drought, flooding, rising seas and wildfires. Developing countries need about US$305 billion a year to adapt, roughly 10 times the US$28 billion in international support they received in 2024. Last year, countries called for efforts to at least triple adaptation finance by 2035.

Carbon markets let countries or companies that emit greenhouse gases buy credits to fund projects that reduce emissions elsewhere. Projects often involve planting or protecting trees or changing farming practices. But the markets have grown slowly: the voluntary market was worth about US$90 million in 2006 and US$535 million in 2024.

In 2024, buyers used about 70 million credits from farm and forest projects, each representing one ton of carbon dioxide cut or stored. That is less than one per cent of the eight to 14 billion tons that farms and forests could deliver each year.

Rohit Jindal, who has studied carbon markets on farms and in forests for two decades, says longstanding concerns remain about whether small farmers can benefit. In a 2006 paper, he and his colleagues warned that projects could exclude poor farmers and people without formal land titles. Low carbon prices and the costs of enrolling many small farms can also limit benefits for smallholders.

Jindal’s recently published paper, along with research under peer review, identifies three recurring barriers to expanding carbon markets. First, it can be difficult to establish whether a project caused real emissions cuts. A large study found that fewer than one in six carbon credits reflected genuine emission cuts.

Second, measuring carbon in soils and trees is costly and can reduce farmers’ returns. In Mozambique, Jindal and colleagues found that a community carbon project raised household incomes only slightly, despite generous contract terms.

Third, projects tend to be established where they are easiest to run. China, India and Brazil hosted more than three-quarters of all projects under the UN’s original carbon credit scheme. In Africa, Jindal and colleagues found as early as 2008 that insecure land rights discouraged new investment.

Climate change can also affect projects. In Tanzania’s Uluguru Mountains, Jindal and colleagues offered farmers contracts to plant and protect trees in 2009. Twenty-one months later, 63 per cent of the trees had survived, with some losses linked to the failure of the short rains.

Adaptation projects face additional measurement challenges. Carbon projects track success in tons of carbon dioxide, which can be converted into credits and sold. Adaptation has no equivalent common measure: outcomes might include less damage from floods or harvests that survive drought, benefits that can take years to appear and are difficult to measure.

Jindal sets out three lessons from carbon markets: keep measurement simple and local; reduce implementation costs and ensure funds reach farmers; and direct support to places most at risk, rather than only to locations where projects are easier to establish. Field experiments in Vietnam found that villagers rarely cheated when monitoring their own work, suggesting communities can help verify results.

Carbon projects face upfront expenses to find buyers, negotiate contracts, obtain government approval and hire independent auditors. Governments and international organizations can help lower these costs by simplifying rules and making it easier to work with community groups that can enroll many farmers at once. Small pilot projects can also help countries learn what works.

Jindal says more adaptation finance is badly needed, but without credible projects on the ground, that money will struggle to make a difference. He has previously received funding from the World Agroforestry Centre, the International Maize and Wheat Improvement Center, and the Social Sciences and Humanities Research Council of Canada to carry out the research cited in the article. He is Department Chair & Associate Professor, Department of Decision Sciences, MacEwan University.