“Many of the most popular offset project types feature intractable quality problems”
The “most comprehensive review of evidence” finds that carbon offsets just don't work.
A new paper published in the Annual Review of Environment and Resources states that, “A growing number of studies have found that the most widely used offset programs continue to greatly overestimate their probable climate impact often by a factor of five to ten or more.”
The paper provides a “systemic review of the literature on carbon offsets”. It’s the most comprehensive review of evidence on the effectiveness of carbon offsetting so far. The paper is titled, “Are Carbon Offsets Fixable?” and concludes that “many of the most popular offset project types feature intractable quality problems”.
The paper is written by Joe Romm of the University of Pennsylvania, Stephan Lezak of the University of Oxford, and Amna Alshamsi of the University of Sussex.
In a statement, Stephan Lezak says, “We must stop expecting carbon offsetting to work at scale. We have assessed 25 years of evidence and almost everything up until this point has failed.”
The authors note that the decision at COP29 in Baku to adopt standards for carbon markets under Article 6 of the Paris Agreement “contrasts starkly with the scientific consensus”.
The paper summarises why carbon offsets “so rarely work as intended” and looks into whether the problems with carbon offsets are fixable.
The authors look at why carbon credit quality has been a problem for more than two decades. They focus on overcrediting, additionality, leakage, and permanence. A carbon offset may be fine on three of these areas, but if it fails on the fourth, “it may represent a far lower emission reduction than was credited — or no real reduction at all”.
This post also focuses on these four aspects of carbon credits. Romm, Lezak, and Alshamsi’s paper raises a series of other issues, relating to double counting, environmental justice, cobenefits, safeguards, land rights, and the contributions approach.
In a statement, Alshamsi comments that,
Despite efforts to implement safeguards, carbon offset projects continue to face documented cases of weak accountability, risking the perpetuation of neocolonial patterns of appropriation. While nature-based projects can deliver local benefits, these should be financed through mechanisms other than carbon credits, such as contribution claims where projects are financed while still ensuring that purchasing entities are responsible for reducing their own emissions.
Cheap, junk credits
In their paper, Romm, Lezak, and Alshamsi write that,
For more than two decades, peer-reviewed studies have found offset quality to be poor across most of the largest (by volume) carbon offset programs. That literature has increased in recent years even as it has analyzed larger numbers of credits, suggesting that poor quality is widespread.
In theory all carbon offsets represent one ton of CO₂e. This creates an incentive for companies to buy cheaper offsets. And cheaper offsets tend to be of lower quality. A 2024 paper in Nature Communications found that “individual companies are a major cause of persisting quality issues due to their demand for problematic and cheap offset types known to overstate emission reductions”.
The paper notes that the various conflicts of interest that run through carbon markets exacerbate the race to the bottom “where project developers compete on price rather than quality”.
Overcrediting
The authors write that,
Warnings that carbon offset projects are prone to overcrediting date back over two decades. A major analysis of CDM [clean development mechanism] projects found that 85% had “a low likelihood of ensuring environmental integrity (i.e., ensuring that emission reductions are additional and not overestimated)” whereas only 2% had a high likelihood.
The voluntary carbon market has a similar structure and similar problems to the CDM. And Article 6 has an governance structure that is “nearly identical” to that of the CDM, as Danny Cullenward at the University of Pennsylvania points out.
A 2023 review of REDD projects found that for every real carbon credit, 12 fakes ones were issued. A 2024 study found that cookstove projects overcredited by nine times. REDD and cookstove projects make up just under two-thirds of the credits issued in the voluntary carbon market so far.
Additionality
The authors write that,
Imperfect information is a problem inherent to the offset market because when offsets are based on counterfactuals, their additionality is inherently uncertain. As a 2007 Nature Climate Change article put it: “No test for additionality can provide certainty about what would have happened otherwise.”
CDM has had more oversight than any other offset programme, the authors note. But it “has been consistently criticized for awarding a great many offsets to clean energy projects throughout its history that were very unlikely to be additional” Romm, Lezak, and Alshamsi write.
“When people write offset rules, they always ignore the fact that there are 1,000 smart people next door that will try to game them,” Mark Trexler, a former offset project developer, told journalists Lisa Song and James Temple in 2021.
In 2019, the two largest offset certifiers, Verra and Gold Standard, stopped issuing offsets from grid-connected renewable energy projects in richer countries. But a 2024 Carbon Market Watch report found that the use of renewable energy-based offsets had increased significantly since 2019 as Verra and Gold Standard continued to issue offsets from existing projects.
Meanwhile organisations like the Doha-based Global Carbon Council has stepped forward to issue offsets from renewable projects that would almost certainly have gone ahead without carbon funding.
Leakage
“From a practical perspective, the leakage resulting from REDD+ projects that avoid industrial logging likely cannot be known with any great certainty,” the authors write. “There are too many factors involved to observe the regional or even global market impact of marginally constraining supply in a globalized industry.”
Peer-reviewed literature suggests that leakage could be “extremely high”. A 2023 paper found that market leakage estimates “are typically above 70% . . . and can reach >100%”. That paper notes that “the use of arbitrarily low estimates of possible leakage appears widespread”.
A 2022 study of 86 REDD projects in Colombia, Indonesia, and Peru found that 29 claimed no leakage, 28 subtracted leakage at a median rate of 6%, and the rest provided no data.
Concerns about leakage date back to the 1990s. But a 2023 paper concludes that “there is little evidence that current practices to address leakage actually work”.
Permanence
The authors write that,
The issue of permanence (or durability) has long been a salient concern for offsetting CO₂ emissions because when humans put CO₂ into the air, a large fraction stays in the air for centuries.
There is no consensus on what is an acceptable level of permanence. 100 years is often put forward as long enough. But “nature-based solutions project rarely promise such durability”, the authors write.
Issues threatening offset permanence include limited land tenure, ecological disturbance, human disturbance, and mismanagement. Forestry durability concerns have only risen as climate change has worsened, as a number of impacts exacerbated by climate change — particularly pests, drought, and wildfire — directly threaten trees.
A 2024 paper found that “a CO₂ storage period of less than 1000 years is insufficient for neutralizing remaining fossil CO₂ emissions”.
The reality is that carbon offsets do not reduce greenhouse gas emissions. Even if a carbon offset succeeds in reducing greenhouse gas emissions by one ton (which they very, very often do not), the buyer of the carbon offset will use it to legitimise continued emissions.