United Nations-backed framework for protecting tropical forests could allow governments to collect income from carbon credits without advancing forest conservation. The weakness lies in how the program calculates baselines, which is the expected rate of deforestation without intervention. There is no evidence that enrolled jurisdictions — countries, states, and provinces — have acted on that opportunity, but the incentive structure favors those who do, Yale researchers found. It also penalizes the jurisdictions that are in most need of intervention.
Emissions from land use changes come mostly from deforestation and account for about 10-12% of total anthropogenic carbon dioxide emissions, according to the Global Carbon Budget . The forest credit program at issue in the study is jurisdictional REDD (JREDD+), a variation of the Reducing Emissions from Deforestation and Forest Degradation (REDD+) program, which faced widespread criticism for generating credits that don't represent real emissions reductions.
REDD+ is a project-based approach to credits, in which landowners enroll to receive payments for reducing deforestation on their plots. It was developed under the United Nations Framework Convention on Climate Change and included in the 2015 Paris Agreement. However, REDD+ projects can include enrolled forests that were never seriously at risk for deforestation. Project-based REDD+ also carries the risk of leakage: while landowners enrolled in the program cut their deforestation rates, others would increase theirs to meet agricultural and market demand.
JREDD+ was developed gradually in response to the problems in the REDD+ program. JREDD+ pays state, provincial, and national governments to reduce deforestation within their borders. Brazil was the first country to create the program in 2008. The voluntary carbon market has increasingly turned to JREDD as a more credible alternative to project-based carbon credits, Yet, the study, published in the Proceedings of the National Academy of Sciences , found that there were also weaknesses in the design of this iteration of the program. These include:
- Jurisdictions that are already decreasing the rate of deforestation can generate credits even if they aren't taking any new action.
- Locations where deforestation is rising — and most in need of funding — can be discouraged from joining the program because they would have to reduce forest loss dramatically before generating their first credit.
- Deforestation spiked temporarily in half the jurisdictions that enrolled in JREDD right before the crediting period began.
"People have talked a lot about what the benefits are of getting carbon credits from jurisdictional REDD. This study points out that there are definitely some things we should worry about, even if we think that those actors haven't taken advantage of them — yet" said study coauthor Luke Sanford, assistant professor of environmental policy and governance at the Yale School of the Environment.
There are now over $3 billion in committed credit purchases under ART TREES, one of the two major JREDD registries. U.S. companies, including Amazon, Walmart, and Salesforce, are some of the largest purchasers of credits from JREDD+ through the LEAF Coalition, a public-private initiative aimed at halting tropical deforestation by 2030.
"There's billions of dollars of commitments, but there hasn't been that much evaluation of what those credits are going to look like, their potential, and the strengths and weaknesses of the program," Sanford said.
The study's most concerning finding, the authors note, is the potential for adverse selection — the possibility that jurisdictions could enroll in JREDD+ precisely because they know the program will credit them for reductions they were already planning to make. The mechanism is baked into the program's design: JREDD protocols calculate baselines using a simple average of deforestation over a previous reference period. That means a jurisdiction already trending toward less deforestation can enroll and collect credits without implementing any new conservation policies.
The researchers also found the reverse is true: jurisdictions with rising deforestation — the places arguably most in need of intervention — are effectively penalized by the baseline structure. This is because they would have to cut their deforestation dramatically just to reach the threshold at which credits kick in, which may not be affordable for them.
Despite the clear structural opportunity for gaming, the study found no evidence that enrolled jurisdictions have exploited it. Governments, unlike private companies, can't easily optimize enrollment timing to maximize profit. They also face political constraints — election cycles, legislative processes, constituent pressures — that make purely strategic behavior harder, Sanford said.
However, the authors identified an additional issue of concern. Deforestation spiked significantly in the years just before the JREDD+ crediting period began and then dropped back down afterward. Pre-enrollment spikes in deforestation can inflate the baseline, making reductions during the crediting period look larger than they actually are. Sandford and coauthor Alberto Garcia, a former postdoctoral researcher at YSE and assistant professor at the University of Utah, identified this as an "anticipatory moral hazard."
They suggested a new approach using dynamic baselines. The baseline would be calculated after the crediting period ends based on deforestation data from other jurisdictions deemed comparable, instead of being set in advance. Thus, jurisdictions wouldn't know their own baselines in advance, reducing the opportunity to game the system. This would, however, also make it harder for governments to anticipate the benefits of the program.
"We took seriously how the different incentives faced by both jurisdictional governments and landowners themselves can shape the integrity of forest carbon credits in jurisdictional programs," Garcia said. "I'm optimistic that better understanding those incentives can help inform the design of more credible carbon markets, especially as jurisdictional REDD+ continues to grow and evolve."
