skip to content

Cambridge Centre for Carbon Credits (4C)

A closeup image of an insect sitting on a leaf in a tropical forest.

A letter to Science, co-authored by 17 conservation and climate researchers, argues that carbon credits can be a valuable tool for climate change mitigation and forest conservation, but their success depends on improving their credibility.

The co-authors are from the Universities of Cambridge, Oxford, Exeter, Bangor, São Paulo, North Carolina, VU Amsterdam, the Pontifical Catholic University in Rio de Janeiro and the London School of Economics. They include the lead researchers from the papers which formed the foundation of The Guardian’s investigation into carbon credits earlier this year (West et al, Guizar-Coutiño et al).

The short letter proposes some solutions, for example, drawing on work from economics and public health to better understand the ‘counterfactual’, or what would have happened to a forest without the credit-funded intervention.

Tom Swinfield, lead scientist at the Cambridge Centre for Carbon Credits and co-author of the letter, says that scrutiny of REDD+ forest preservation projects is welcome, and gives the opportunity to demonstrate the robust analysis which supports them.

“REDD+ projects are some of the few types of credits in the voluntary carbon market that can be evaluated using only remote sensing and econometrics. The lack of these assessments for other types of credits does not mean they are more robust: they lack the same level of scrutiny,” says Tom.

You can read the letter in Science here: DOI 10.1126/science.adh3426

It was published in the Friday 5th May 2023 edition of Science, and is posted in full below.

Credit credibility threatens forests

Addressing global warming requires increased investment in conserving and restoring carbon-dense natural habitats. Some companies that emit carbon have turned to certified carbon credits to offset their environmental impact. However, the effectiveness of carbon credits depends on the methods used to quantify them. If carbon credits do not accurately represent their environmental benefits, relying on them could exacerbate climate change (1). To ensure that carbon credits are robust, the methods used to calculate them must be improved.

So far, credits from tropical forest conservation have been generated by estimating project effects through comparisons with historical trends in reference areas identified by project proponents (2). However, there is considerable evidence that these methods substantially overestimate the degree to which projects are changing outcomes (i.e., their “additionality”) (3–5).

Carbon crediting procedures also need to get better at accounting for the risk of increased emissions elsewhere. If food or timber production is prevented by a carbon offset project, that production and its carbon impacts may simply move to a different location (6, 7). Most assessments of this carbon “leakage” focus on rough estimates of small-scale leakage and ignore or greatly underestimate longer-range displacement of production (8).

Finally, the carbon credit system needs more robust ways of accounting for the impermanence of carbon held in vegetation and soils. Current certification methods try to underwrite credit permanence claims by maintaining a shared pool of nontradable credits which can be drawn from in the event of reversals (9). However, this system provides no incentive for future stakeholders to prevent carbon releases after credits have been issued (10).

Better methods are available. Additionality can now be estimated more reliably using statistical techniques developed over many decades in economics and public health to eradicate bias in estimating the counterfactual—i.e., what would have happened without the intervention (11, 12). Likewise, there are new methods to better adjust for leakage effects (8), and properly value impermanent storage (10). All of these improvements should be routinely deployed in credit quantification. Carbon credits can be a valuable tool for climate change mitigation and forest conservation, but their success depends on improving their credibility. 


Andrew Balmford1*, Pedro H. S. Brancalion2, David Coomes3, Ben Filewod4, Ben Groom5, Alejandro Guizar-Coutiño3, Julia P. G. Jones6, Srinivasan Keshav7, Andreas Kontoleon8, Anil Madhavapeddy7, Yadvinder Malhi9, Erin O. Sills10, Bernardo B. N. Strassburg11, Frank Venmans4, Thales A. P. West12, Charlotte Wheeler3, Tom Swinfield1

  1. Department of Zoology and Conservation Research Institute, University of Cambridge, Cambridge CB2 3EJ, UK.
  2. Department of Forest Sciences, University of São Paulo, 13.418-900, Piracicaba-SP, Brazil.
  3. Department of Plant Sciences and Conservation Research Institute, University of Cambridge, Cambridge CB2 3EA, UK.
  4. Grantham Research Institute on Climate Change and the Environment, London School of Economics and Political Science, London, WC2A 2AE, UK.
  5. Land, Environment, Economics, and Policy Institute, Department of Economics, University of Exeter Business School, Exeter EX4 4PU, UK.
  6. School of Natural Sciences, Bangor University, Bangor, UK. 7Department of Computer Science and Technology, University of Cambridge, Cambridge CB3 0FD, UK.
  7. Department of Land Economy, University of Cambridge, Cambridge CB3 9EP, UK.
  8. Environmental Change Institute, School of Geography and the Environment, University of Oxford, Oxford OX1 3QY, UK.
  9. Department of Forestry and Environmental Resources, North Carolina State University, Raleigh, NC 27695–8008, USA.
  10. Rio Conservation and Sustainability Science Centre, Department of Geography and the Environment, Pontifical Catholic University, Rio de Janeiro, Brazil.
  11. Institute for Environmental Studies (IVM), VU University Amsterdam, Amsterdam, Netherlands.

* Corresponding author.


  1. P. Greenfield, “Revealed: More than 90% of rainforest carbon offsets by biggest provider are worthless, analysis shows,” The Guardian (2023).
  2. Verra, “VCS Standard 4.4” (2022).
  3. T. A. P. West et al., Proc. Natl. Acad. Sci. U.S.A. 117, 24188 (2020).
  4. A. Guizar-Coutiño et al., Conserv. Biol. 36, e13970 (2022).
  5. P. Delacotte, G. le Velly, G. Simonet, Res. Energ. Econ. 67, 101277 (2022).
  6. C. Streck, Clim. Pol. 21, 843 (2021).
  7. N. Villoria et al., Nat. Comm. 13, 5476 (2022).
  8. B. Filewod, G. McCarney, “Avoiding leakage from nature-based offsets by design” (London School of Economics and Political Science, 2023);
  9. Verra, “Proposal to create a long-term reversal monitoring system” (2021).
  10. A. Balmford et al., “Realising the social value of impermanent carbon credits,” Cambridge Open Engage 10.33774/coe-2023-5v93l-v3 (2023).
  11. G. W. Imbens, J. D. Angrist, Econometrica 62, 467 (1994).
  12. P. J. Ferraro, M. M. Hanauer, Ann. Rev. Env. Res. 39, 495 (2014).


A.B. is a director of the Cambridge Centre for Carbon Credits, a research partnership aimed at improving the robustness of estimates of the performance of carbon credit–generating projects. P.H.S.B. is a founding partner of and scientist at, a company dedicated to restoring native ecosystems. B.B.N.S. has equity stock in and is a board member of T.S. is funded by the Tezos Foundation gift to the Cambridge Centre for Carbon Credits (grant code NRAG/719) and has an unpaid role as a member of the Cambridge Offsetting Working Group advising the University of Cambridge on their carbon offsetting decisions.