Caveat emptor: Auditors can’t save carbon offsets

The authors of an editorial in Science magazine reply to carbon offsetting industry responses to their critique.

PHOTO Credit: Chris Lang.

In July 2025, Science magazine published an editorial under the headline, “Auditors can’t save carbon offsets.” The editorial was written by by Cynthia Giles, an independent scholar who has worked in senior roles at the US Environmental Protection Agency under the Obama and Biden Administrations, and Cary Coglianese, Professor of Law and Political Science at the University of Pennsylvania.

The editorial was based on a longer report by the same authors that reviewed 95 projects that significantly overclaimed carbon credits. The projects were all registered with Verra.

In the Science editorial, Giles and Coglianese highlight the conflicts of interest in the carbon offsetting industry. “All key market participants benefit from inflated claims about carbon credits,” they write, from project developers, to registries, to auditors. “Although ongoing efforts to modify methodologies can help, they won’t eliminate subjective judgments underlying these credits that introduce the opportunity for bias.”

The editorial was covered by Natasha White for Bloomberg, Rishika Pardikar for Drilled, Carla Ruas for Mongabay, Isa Mulder for Carbon Market Watch, and elsewhere. It was also featured in a podcast by the Kleinman Center for Energy Policy at the University of Pennsylvania and a Reclaim Finance webinar .

And, of course, REDD-Monitor wrote about it:

Verra and the International Association of Validation and Verification Bodies (IAVVB) responded to the Science editorial. This post gives Coglianese and Giles’ responses to Verra and the IAVVB.

Response to Verra

Five days after the editorial in Science, Verra wrote what it calls a “constructive response”. It’s typical of Verra’s responses to any criticism. Verra argues that its system is rigorous, that it is introducing new methodologies, and that the criticism relied on research and studies that “appear to have worked backward from a predetermined conclusion”.

Here is Coglianese and Giles’ response — posted on LinkedIn:

Reply to Verra’s “Constructive Response”

Cary Coglianese and Cynthia Giles

We recently published an editorial in Science — “Auditors Can’t Save Carbon Offsets” — explaining that third-party auditors for carbon offsets cannot provide the guarantee of credit integrity they are claimed to provide. Our Science editorial is supported by a longer report — “Third-Party Auditing Cannot Guarantee Carbon Offset Credibility” — that details our research and provides an extensive review of the auditing bias literature to support our conclusions.

The climate registry Verra, a key entity that registers offset projects and awards carbon credits, has recently posted a response to our research: “Beyond the Headlines: A Constructive Response to Criticism of Validation and Verification in the VCM.” This response is more revealing than Verra likely intended. Strikingly, for all its umbrage at our work, Verra does not dispute our central point that there is inherent conflict and bias created by allowing developers to select and pay their own auditors.

In our work, we cite a broad array of academic studies from a variety of contexts showing that third-party auditors selected and paid by the company they are auditing — the system used in the offset market — face an economic conflict of interest that skews their reviews. Auditors are also subject — as everyone is — to a largely unconscious cognitive phenomenon of self-serving bias that causes them to interpret evidence in favor of their clients.

Third-party auditors in the carbon offset marketplace have not ensured credit integrity over the last 20 years. Despite the requirements for auditing that have been in place since the first offset projects, there have been an ongoing series of scandals and research reports of serious credit overclaiming. In every case, third-party auditors signed off. That real-world result aligns with the academic research on conflict of interest and bias of auditors selected and paid by the audited firm.

The upshot of what we have shown is that assertions along the lines of “auditors ensure that a project’s climate impacts are real” — claims Verra itself has made—are simply not warranted by the evidence. Offset market proponents keep saying that auditors will guarantee integrity, while an extensive body of research, combined with 20 years of experience with carbon offsets, plainly shows they won’t.

Although Verra’s response fails to dispute our main point — and thereby implicitly affirms it — we offer the following comments in reply:

• Verra defends the current developer-pays auditing system in passing by saying, essentially, “but everyone does it.” This feeble defense does nothing to erase the existence of conflicts of interest or self-serving biases in the offset market. More importantly, it overlooks how, as we explain in our work, carbon offset auditors are unusually susceptible to the effects of a self-serving bias because offsets present the very conditions that research has shown make this kind of bias most pronounced.

• Our longer report includes results from our look at 95 projects known to have been over-credited, in which we found that 21 of 33 Verra-approved auditors were involved in one or more of 305 audits of the 95 projects. This finding simply indicates that the problem is linked to more than just a few auditors failing to do a good job. We are not blaming the individual auditors for projects that overstate emissions benefits. Our data are consistent, though, with a structural problem in the system in which auditors are selected and paid by the audited firm. In its response, Verra focuses on what went wrong with these projects. But why the projects were over-credited isn’t the point. Criminal behavior, mistakes, and unwarranted assumptions can all play a role. The point is simply this: the credits weren’t deserved, but many auditors nevertheless signed off.

• Verra points to new methodologies that it says will improve the credibility of carbon claims. These may be a step in the right direction for credit quality. But these new methodologies do not address the foundational problem we cite, which is having auditors selected and paid by the firms being audited. So long as that bias-inducing structure exists, auditors cannot play the integrity-guaranteeing role they have been claimed to fulfill.

• Verra also states that it has in place robust oversight of auditors. Left unacknowledged is the intrinsic conflict and inherent bias Verra itself faces in pushing auditors to be tougher due to its fee structure tied to the number of credits claimed.

Research and experience both show that carbon offset auditors selected and paid by project developers cannot guarantee the integrity of climate benefit claims. Auditors face economic pressures and cognitive biases that weigh against them fulfilling that role, especially if overclaiming of credits occurs at anything close to the rate of 80% or more, as research has indicated that it does.

07.26.2025

Response to IAVVB

In its response to Coglianese and Giles, IAVVB argues that “the conclusions drawn in the report are based on oversimplified assumptions, limited evidence, and a misunderstanding of the VVB function within the system”. IAVVB states that the role of validation and verification bodies is to assess whether projects conform to methodologies and registry rules and that the issuance of carbon credits is “the responsibility of the program operator”.

Here is Coglianese and Giles’ response to IAVVB, posted on LinkedIn:

Response to the IAVVB’s “Position Paper”

Cary Coglianese & Cynthia Giles

Peer-reviewed research indicates that 80 percent or more of the credits awarded to climate offset projects may not reflect real emission reductions. We published an editorial this summer in the journal Science, supported by a longer report released at the same time, explaining why auditors paid by project developers are not the guarantors of credit integrity they are claimed to be. We showed that auditors paid by firms being audited face both economic conflicts of interest and a well-documented self-serving bias that together can lead even well-intentioned auditors to interpret evidence in favor of their clients. Yet offset proponents repeatedly make broad declarations that auditors “ensure that a project’s climate impacts are real and permanent.” There is a certain irony that offset market advocates too often overclaim what auditing can do by way of fixing a market plagued with widespread credit overclaiming.

Following the publication of our editorial and report, the International Association of Validation and Verification Bodies (IAVVB) — an association that represents carbon offset auditors — posted on LinkedIn a response in the form of a self-styled “position paper” that criticizes our work. Although IAVVB seems, on the one hand, to have joined with others in asserting that auditors serve as “a foundational component of environmental market credibility,” the organization, on the other hand, disavows auditors’ responsibility for ensuring that carbon credits reflect true climate benefits. IAVVB states that credit integrity “remains the responsibility of the program operator.” Perhaps it will be disheartening for prospective purchasers of credits to read that, contrary to all the claims about auditors serving as the bulwark of credit integrity, auditors themselves apparently do not see it as their responsibility to guarantee that credits are real. Caveat emptor.

As we noted repeatedly in our work, this is not how the market players themselves characterize auditors’ function. Auditors have been characterized as the “backbone” and “cornerstone” of credit integrity. Principal architects of the carbon market have claimed that auditors “provide independent confirmation that the mitigation activity achieves the claimed GHG emissions reductions or removals.”

IAVVB’s response to our work essentially concedes our core point when it acknowledges what it terms “subjective bias” in the work of auditors. It offers no rebuttal to our central finding about structural impediments to auditor independence. It even implicitly agrees, landing on the dismal note that this foundational problem will remain because the practical challenges of addressing it are “unresolved.”

IAVVB misses the mark in critiquing our study of the extent of auditor involvement in questionable offset projects. In drawing attention to our examination of auditor involvement in 95 projects with serious credit overstatements, IAVVB wrongly accuses us of“misrepresentation through selective sampling.” It mistakenly claims that we implied that the 95 projects we looked at represent the “norm” within the voluntary credit market. To the contrary, we explicitly made clear, in both our editorial and accompanying report, that our aim was to see whether projects that admittedly fell at the tail end of the distribution had been audited by a small number of audit firms or whether — as would be consistent with a more systematic phenomenon — these seriously problematic projects had been reviewed by numerous different auditors. It turned out that nearly two-thirds of the auditors then registered with Verra had been involved in these plainly overstated projects across three major categories (forest protection, rice cultivation, and cookstove distribution). As we stated in our report, “[o]ur sample is not intended to be representative of all projects within these categories or all projects registered by Verra.” Rather, what our sample showed was that auditors’ failure to prevent serious documented overclaiming stemmed from more than just a few isolated auditors, i.e., it was consistent with structural limitations in the auditing system.

Ultimately, IAVVB’s response affirms our principal point, which is that the claims that offset proponents make about auditing’s curative properties are not supported by science or experience. The prevailing offset system—with auditors selected and paid by project developers — remains vulnerable to well-documented conflicts of interest and self-serving bias. The entire system is one in which basically everyone stands to gain from more carbon credits: project developers, purchasers, registries—and, yes, auditors. A system with these kinds of incentives prevents auditors from providing the assurance of credit integrity claimed by numerous market players.