Verified hot air: how a popular carbon offset project in Peru has taken tens of millions of dollars from hundreds of companies and individuals but done nothing to prevent climate change
By Simon Counsell
A new investigation into a REDD+ project in the rainforests of Peru, which has sold a total of nearly ten million carbon credits to hundreds of companies and individuals around the world, has revealed that the scheme generated no real carbon savings, and that deforestation in the area actually increased dramatically after the project started. Detailed analysis of how the project’s carbon credits were generated has found that there was no basis for ‘additionality’, the deforestation ‘baseline’ was grossly exaggerated to create the appearance of reduced deforestation and thus emissions reductions, and that loopholes in the methodology allowed for huge amounts of actual emissions to go unaccounted for. Monitoring on behalf of the verifier, Verra, belatedly spotted serious problems, but millions of ‘verified carbon units’ (VCUs, i.e. carbon credits) were approved nonetheless.
Whilst the Lima-based company that developed the project may have pocketed many millions of dollars in VCU sales, and Verra received around a million dollars in commissions for verifying the fraudulent credits, the supposed project beneficiaries in the forest received almost nothing. Buyers of the credits have included BP Gas, the German ‘green energy’ supplier Entega, and industrial conglomerate Swire. Various carbon offset brokerage outfits such as Helsinki-based Compensation Foundation have also bought significant quantities of the project’s credits. The investigation was undertaken for Foodwatch e.v., which is exposing how food retailers such as Rewe and Aldi are greenwashing their products with these phony offsets.
The full report of the investigation into the project is available here.
The project, called ‘REDD Project in Brazil Nut concessions in Madre de Dios’, was developed by the private Lima-based offset project developer Bosques Amazonicos SAC (BAM). Officially started on January 1st, 2010, it created a partnership with an organisation, FEPROCAMD, which represents hundreds of Brazil nut collectors in the Madre de Dios region of the Peruvian Amazon rainforest.
Under the contract, in exchange for carbon rights to the forest concessions used by 405 Brazil nut collectors, which together originally covered more than 300,000 hectares, BAM was to provide technical and financial support to FEPROCAMD’s members, and a share of the carbon credits generated by the project. The project claimed it would help protect the rainforest where the Brazil nuts were harvested by giving the nut collectors a better income. A new nut processing plant was to be important, as it would increase the value of the product.
In 2012, the project was validated under the Voluntary Carbon Standard (VCS) and in 2014 it was validated (by Scientific Certification Services Inc, SCS) under the Community, Climate and Biodiversity Standard (CCBS). It has since been verified twice more by Verra, most recently in 2020, and continues to operate, selling carbon credits as recently as 5th November 2021.
Verified hot air: No real basis for additionality
From the outset, the project lacked any real basis of additionality. It did not bring about any new legal designation or protection for the forest in the area concerned, because the ~400 Brazil nut harvesters (BNHs) on whom the project would primarily rely already had their forest harvesting concessions legally designated by the Peruvian government before the project started. Instead, the project relied on an economic claim to additionality. The project developer BAM claimed that, without the project, and the additional income it would supposedly bring, the harvesters would not have sufficient financial incentive to protect the forest and would clear it for farmland. This argument was flawed, because the harvesters were very unlikely ever to be allowed to deforest all their concessions (averaging around 800 hectares in size). They are required by law to keep the forest from which they harvest nuts standing, apart from two hectares which could be used for farming. But, more importantly for at least the first ten years, the project delivered very little or no income or benefits to the Brazil nut harvesters anyway.
In order to calculate how many credits the project could generate, BAM used its own ‘modelling’ system to create a counterfactual baseline of how much deforestation might occur in the area in the future without the project. However, analysis using Global Forest Watch’s database (which is based on the same satellite images supposedly employed by the project) has found that this baseline was greatly inflated, apparently by a factor of 8-10. The project claimed that deforestation would occur at a rate of around 1.2 percent per year across the region, whereas at the time the project started the real rate was really only about 0.15 percent per year. This meant that the project could appear to prevent deforestation even if it made no actual difference at all (or deforestation increased greatly) – and thus generate many credits merely through contrived carbon accounting.
In actual fact, deforestation in the project area more than doubled after the project started. Many of the Brazil nut harvesters legally felled forest for farmland in their concessions, but more significantly many also started extracting commercial quantities of timber, thus causing widespread forest degradation. Because they were involved in logging the forest rather than preserving it, resulting in a significant increase in actual carbon emissions, more than half of the Brazil nut harvesters were eventually simply excised from the project for carbon accounting purposes.
Other means were used to conceal actual emissions from the project which, if included, would have greatly reduced the carbon credits available for sale. Under Verra’s carbon calculation rules, emissions of less than 5% of the total baseline figures can simply be ignored as ‘negligible’. Under a more realistic baseline for the project, however, emissions from various sources would have been much greater than five percent of the total, and would have had to be deducted from the claimed reductions, thus reducing the claimed emissions reductions. The emissions from forest degradation due to logging within the BNH concessions were similarly ‘disappeared’ because they were technically shifted from the project area to the ‘leakage belt’, where the baseline was equally hugely inflated, and hence could be disregarded as ‘negligible’.
Applying the same means used to calculate the project’s claimed VCUs, but with a baseline reflecting the actual rate of deforestation in the area prior to the project, it has been calculated that even the theoretical amount of VCUs which the project could have generated was only around eight per cent of those which were actually claimed to have been generated. Between 2000 and 2016, the project generated 10.25 million VCUs, 9.3 million of which have been sold to date. Each of these could have been sold for US$4.50-US$5.00 each, or even more. However, taking into account all the emissions which were carefully excluded from the project, there is a strong case that none of the VCUs generated by the project represent any real emissions reductions. From 2010 onwards, when the project started, the area was actually a net emitter of greenhouse gases above what would have been a reasonable baseline.
The project should never have been verified
These gross problems with the project and the way it falsely generated credits point to clear failings on the part of the certifier, Verra. Apart from the dodgy carbon accounting, there are many reasons why the project should never have been verified in the first place.
First, there was a profound misunderstanding or misrepresentation of the supposed role of the Brazil nut harvesting concessionaires. They have had neither the incentives nor the rights which would be required for them to fulfil their role in protecting the forest. There were serious problems of lack of clarity over the tenure of the forest. None of the Brazil nut harvesters actually lived on their concessions, and all are relatively recent migrants from the Andean uplands rather than being indigenous or local people. Some were involved in illegal land deals with other migrants, or illegal gold mining. Many of them were in fact causing additional deforestation.
Second, the benefits from the project were heavily skewed towards the project developer, BAM, which under the terms of the project agreement received 70% of the proceeds of the sale of carbon credits. Investment which was deemed essential to the success of the project – especially for the construction of a Brazil nut processing plant in the project area, but also other aspects of the project – never actually happened. At least up until 2014, none of the Brazil nut concessionaires themselves had received any actual payments and, with one known exception, there is evidence that very little or none may have happened ever since for most of the concessionaires. The incentive being provided to the concessionaires to prevent deforestation was also therefore very little or nil. Whilst the Brazil nut harvesters’ association FEPROCAMD was supposed to receive 30% of the benefit of the projects, it did not even have a project bank account up until at least January 2021 (i.e., eleven years after the project started). Documents show that FEPROCAMD had repeatedly requested the project developer BAM for more financial accounting information, copies of contracts, verification reports etc, but had failed to receive them.
After 2012, monitoring of the project by BAM, and verification of it by the verification company, Verra, was very fragmented and often lagged a very long way behind the closing of carbon verification and accounting periods. Verification was supposed to take place after the first three years, then after each two years thereafter. Each verification would calculate how many credits could be issued for the period just completed, each running to several million. But the verification reports from 2014 onwards, were not completed until 2019 or 2020, years after the carbon accounting period ended. It becomes increasingly clear from the verification reports that concerns about some fundamental problems with the project have been mounting. According to the latest of these reports, relating to 2015-2016, many serious problems remained to be addressed by the project developer – including about the appropriateness of the baseline scenario under which all the supposed emissions reductions are calculated. Despite these growing concerns, which should have had a material impact on the issuance of VCUs, even halting them altogether, the verifiers nevertheless continued to verify that the project had generated many millions of VCUs.
The financial structuring of the Verra verification process represents a very clear and significant conflict of interest. In addition to the very substantial (six-figure) fees which are paid by the project developer to the verifier for each verification exercise, on top of the initial fee of US$115,000 for opening a project account, a fee of US$0.10 is also payable on each issued VCU. This means that, in the case of this project, Verra would have so far received commissions of around $1 million for issuing the VCUs under the inflated baseline, rather than very little or nothing had a realistic baseline been applied.
Alarm bells should be ringing
Under the agreement reached at UNFCC climate COP26 of some of the basic mechanisms for a global carbon market, verified emissions reductions from private offset schemes are likely to get a significant boost, even being used to count against countries’ Nationally Determined Contributions. The problems shown with the carbon accounting for this particular project in Peru are certainly not unique – lack of additionality and inflated baselines have been shown elsewhere, and are probably endemic in similar kinds of projects. The fundamental problem of conflict of interest on the part of offset verifiers also applies to schemes such as the Gold Standard, which receives an even higher commission for every credit issued.
The consequences of failing to address these problems – and especially to definitively remove all fraudulent offset credits from the markets, and all conflicts of interest in offset verification – could be catastrophic, as private and ‘nature-based’ offsets schemes start to fill the gaps left by government inaction. For companies buying such credits, either as a well-meaning attempt to help the planet, or more cynically for mere greenwashing purposes, very loud alarm bells should be ringing; regardless of whether they are verified or not, many carbon credits are going to prove as worthless as those from the REDD Project in Brazil Nut concessions in Madre de Dios.
