Forests store enormous amounts of carbon, and tropical deforestation releases it at a scale few other activities match. REDD+ was designed to make standing forests worth more than cleared ones. It is also the part of the carbon market that has faced the sharpest criticism, and the part that has changed most in the past two years.
REDD+ earns carbon credits for deforestation and forest degradation that a project prevents, measured against a baseline of the forest loss that would otherwise have happened. That baseline is where both the value and the controversy sit. This article explains how REDD+ works, what the evidence on over-crediting shows, how new methodologies and jurisdictional REDD+ change the picture, and what a buyer should check today.
How REDD+ works
REDD+ began as a UN framework for paying tropical countries to reduce deforestation. In the voluntary market it mostly takes the form of projects:
A forest under threat. The project area faces deforestation, planned (for example a concession for plantations) or unplanned (gradual clearing by agriculture, logging or settlement).
A baseline. The project estimates how much forest would be lost without it, usually from historical deforestation in a comparable reference area.
Protection activities. Patrols, land tenure support, alternative livelihoods, fire prevention, community agreements.
Credits. The difference between the baseline loss and the actual, monitored loss, converted into tonnes of CO2, after deductions for leakage and a contribution to the buffer pool, becomes credits.
REDD+ credits are avoidance credits: they keep carbon in the forest rather than removing it from the atmosphere (avoidance vs removal).
Why REDD+ is criticised
Measuring a standing forest is straightforward with satellites. Measuring the forest that would have disappeared without the project is not. If the baseline assumes too much deforestation, the project earns credits for forest that was never going to be cut.
Independent research has tested this. A synthesis published in April 2026 combined six independent evaluations of 44 REDD+ projects. It found that about four in five projects did reduce deforestation, but that together they issued almost eleven times more credits than the avoided deforestation justified, mainly because reference areas and baseline models overstated the threat (Swinfield et al., Nature Communications, 2026).
Three other issues come up often:
Leakage: protecting one forest can push clearing to the next one.
Permanence: protected forest can still burn or be cleared later (permanence and buffer pools).
Rights and benefit sharing: many REDD+ areas are home to indigenous peoples and local communities, whose consent and share of revenue must be secured.
The criticism targets how credits were calculated; the same research found most projects did protect forest. The challenge is paying for that protection without overstating the tonnes.
What changed: new methodologies and the CCP label
Verra’s VM0048 replaces Verra’s older REDD+ methodologies with one approach in which baselines are not chosen by each project. Verra sets deforestation risk and activity data per jurisdiction, and projects receive their share. Legacy methodologies such as VM0007 were excluded from Verra’s submission to the ICVCM, and existing projects have to move to VM0048 at their next baseline reassessment or crediting period renewal. The rollout depends on Verra’s jurisdictional risk maps; by April 2026 the first credits under the new framework were close to issuance.
Jurisdictional REDD+ takes the approach one level higher: a whole country or state reduces deforestation and is credited for the result, under standards such as ART TREES or Verra’s Jurisdictional and Nested REDD+ (JNR) framework. Individual projects can be “nested” inside a jurisdictional programme, so that project and government accounting add up.
In November 2024 the ICVCM approved three REDD+ methodologies for the CCP label: ART TREES v2.0, Verra VM0048 and Verra JNR v4.1 (ICVCM). Credits under the legacy methodologies cannot carry the label (Core Carbon Principles explained).
What this means for credits on the market today
Most REDD+ credits available today still come from projects under legacy methodologies, because the new framework is only starting to issue. That creates a practical split:
Legacy-methodology REDD+ credits | VM0048 or jurisdictional credits | |
|---|---|---|
Baseline | set by the project | set at jurisdiction level |
CCP label | not possible | possible |
Availability today | large | small, growing |
What to check | the project’s own baseline, independent ratings, leakage, community agreements | the jurisdiction’s data and nesting rules |
A legacy credit is not automatically weak, and a new one is not automatically strong. But a legacy credit needs more project-level due diligence, because the methodology itself is no longer the quality floor.
Two projects in the Regreener portfolio show the range: Katingan Mentaya in Indonesia (Verra VCS 1477), which protects about 150,000 hectares of peatland forest originally slated for plantations, and Matavén in Colombia (Verra VCS 1566), led by indigenous communities across more than a million hectares. Both use a legacy methodology, both have faced public questions, and both went through our full screening before we offered them.
Tip from Bernard de Wit, Founder of Regreener: “For any REDD+ project, ask for three things: the deforestation rate assumed in the baseline, the actual rate in the reference area since the project started, and the latest independent rating. If the assumed rate is far above what actually happened next door, you have found the over-crediting risk before you pay for it.”
How to buy REDD+ credits responsibly
Check the baseline: assumed deforestation rate vs observed rates in comparable areas.
Prefer recent methodologies where available (VM0048, JNR, ART TREES) and look for the CCP label.
Use independent ratings from BeZero, Sylvera or Calyx; REDD+ is where ratings differ most (rating agencies compared).
Check community agreements: consent processes, land tenure and how revenue is shared.
Claim it correctly: REDD+ is avoidance. It supports a contribution claim; it does not count as a removal for SBTi’s removal requirement (SBTi carbon removal requirements).
For the projects we currently rate highest, see the best REDD+ carbon credits. Our free credit quality checker scores a specific project.
“Forest protection is some of the most important climate action there is, and some of the hardest to measure. We still buy REDD+, selectively, because the alternative is letting those forests go. What changed is that we now check every REDD+ baseline ourselves before we offer a project.”
Bernard de Wit, Founder of Regreener
Check a REDD+ project before you buy
Our free credit quality checker scores a project on baseline, additionality and permanence, with the questions to ask your supplier.
