What the latest investigations reveal and why retirements are now a reputational liability
Registry actions and law enforcement attention have made one point hard to ignore: offset risk is not only about “credit quality”. Reporting on investigations in Brazil has linked some large forest and REDD+ projects to alleged illegality risks such as land grabbing and illegal logging, followed by registry suspensions and extraordinary checks. For buyers, that translates into operational, legal, and human-rights exposure across the supply chain, not just a technical debate about methodologies.
Retirement has become the reputational trigger. When a company retires credits to support a public claim, that action becomes traceable, easy to scrutinise, and easy to challenge. At the same time, litigation and enforcement around environmental claims and market instruments is rising, which raises the stakes for any claim that relies on offsets.
Procurement teams should treat retirement data as a risk signal, not a comfort blanket. Analysis of credits retired in Brazil between January 2024 and June 2025 has argued that a very large share of retirements from major projects could be “problematic” on issues such as methodologies, baselines, governance, and rights. Even if you disagree with the framing, the practical takeaway is clear: “registry plus price” is not a due diligence strategy.
Buyer behaviour is shifting, but not fast enough to eliminate headline risk. Market commentary in 2025 points to overall retirements staying broadly stable while demand rotates toward higher integrity and durability, with a reduced share of retirements on Verra compared with historical peaks. That is consistent with buyers learning that reputational risk can dominate the economics of a cheap credit.
Venezuela is a governance lesson, not a one-off story. In contexts with weak institutions, opaque land and carbon rights, and limited independent verifiability, the risk of disputes over additionality, title, and benefit sharing increases. For global buyers, “jurisdiction risk” belongs on the same checklist as project type and standard.
The uncomfortable pattern is that problems often surface downstream. If controversies emerge after credits are already bought or retired, the real question becomes how low-integrity credits enter the sourcing and procurement process even at sophisticated organisations.
How low-quality forest and land-use credits slip through due diligence and procurement processes
Baseline risk slips through when procurement buys a serial number and a co-benefit narrative. A common failure mode in project-based REDD+ is insufficient stress-testing of baseline plausibility and additionality. Public investigations and methodological revisions have repeatedly focused on inflated baselines relative to real deforestation trends, which can lead to over-crediting.
Rights and tenure risk slips through when teams treat community issues as “soft”. Weak verification of land tenure and FPIC can become a hard legal and reputational problem later. Litigation tracking has highlighted disputes and legal actions connected to REDD+ projects, including issues around consultation, legitimacy, and Indigenous rights in project areas.
Visibility breaks when credits pass through multiple hands. Spot purchases and layered intermediaries can reduce transparency on project history, holds or suspensions, changes in project proponents, and material events that affect eligibility. Registry status can change, and buyers often learn late if they do not actively monitor.
Incentives push buyers toward the wrong risk profile. Annual offset targets and limited budgets tend to favour low-priced avoidance credits, often in AFOLU, rather than credits with higher durability. When a controversy hits, the ex post cost is rarely the credit price. It is the internal investigation, the communications response, and the potential restatement of claims.
Label shopping creates false comfort. “Compliant with a standard” is not the same as meeting an integrity threshold. Buyers increasingly use integrity initiatives and independent ratings, but market research suggests adoption is uneven and often happens after purchase, when optionality is already gone.
These failure modes cluster around the same technical and governance issues. If additionality, baseline, leakage, permanence, and rights are where projects break, buyers need a checklist that procurement and risk can apply before signing.
A buyer’s verification checklist: additionality, baselines, leakage, permanence, and governance red flags
Additionality should be treated as a document test, not a marketing claim. Red flags include projects in areas already protected by law or public finance, activities that look business-as-usual, and weak economic justification with no credible investment or barrier analysis. Many buyers use integrity principles as internal references, including approaches aligned with VCMI guidance, to define what “good enough to claim” means for their organisation.
Baselines need sensitivity checks, not just a methodology citation. Buyers should ask for evidence behind deforestation models, including historical data, assumptions, and sensitivity analysis. Baseline outputs should be compared with regional trends and independent indicators such as satellite and remote sensing, with clear explanations for deviations. Brazil-linked controversies have repeatedly centred on baseline setting and over-crediting in large forest projects.
Leakage must be evaluated against real-world drivers. Buyers should confirm whether the standard and methodology apply robust leakage assessment and whether leakage deductions are meaningful. They should also ask for evidence of local enforcement capacity and an explanation of commodity-driven deforestation dynamics, because macro drivers can overwhelm project boundaries. Research on corruption risks in voluntary carbon markets reinforces why jurisdictional context matters when incentives are high and oversight is weak.
Permanence needs both a buffer view and an events view. Buyers should review buffer pool rules, insurance mechanisms, and monitoring governance, then stress-test material risks such as fire, illegal logging, and other reversal pathways. Reporting on large Amazon-linked deals and the operational realities of forest protection underscores that reversal risk is not theoretical. Portfolio concentration limits should reflect both physical risk and reputational risk, because investigations can function like a “reversal” for claims even when tonnes remain on a registry.
Governance and rights should be treated as eligibility criteria. Buyers should check FPIC documentation, benefit sharing terms, grievance mechanisms, and active disputes or litigation. Litigation updates have highlighted how disputes around consultation and Indigenous rights can become central to project credibility. In opaque institutional contexts, including those discussed in relation to Venezuela, uncertainty around title and benefit sharing can amplify every other risk.
A checklist only works if it is backed by hard controls. Buyers still need to verify registry status, audit trails, and the underlying MRV documents so they do not discover a hold, a review, or a mismatch after retirement.
Registry and documentation checks that matter: project history, monitoring reports, audits, and retirement trails
Project history should be reconstructed as a timeline before purchase. Buyers should document registration date, methodology versions, issuance batches, any holds or suspensions, changes in project proponents, and material land tenure events. Market reporting has shown registries placing projects on hold or pausing processes when events appear inconsistent with the project design, such as harvesting. That makes “check for holds and material events” a pre-contract requirement, not a post-issue clean-up step.
Monitoring and verification documents should be mandatory reading for material volumes. Buyers should require a complete MRV pack, including monitoring reports and verification body documentation, and then review baseline assumptions, sampling choices, uncertainty treatment, leakage deductions, and non-permanence risk assessment. Guidance on integrity and corruption risks supports adding a “red-team review” for large lots, because the cost of a second set of eyes is typically lower than the cost of a public dispute.
Audit quality should be assessed, not assumed. Buyers should look at who the validation and verification body is, the pattern of findings, and whether there were escalations or independent reviews. Registry actions cancelling overissued credits show that “verified” does not mean “immune”, and that overissuance can be corrected after the fact. That reality should feed directly into contract recourse and replacement clauses.
Retirement trails must reconcile serial numbers with claims. Buyers should ensure retired serials match what is used in reporting, whether in sustainability reports, product claims, or public webpages. They should also check for mismatches between purchased, retired, and claimed volumes, and maintain an evidence pack suitable for external assurance. Policy and legal commentary on integrity and transparency trends supports treating this as a governance control, not a communications preference.
Counterparty and contract terms should assume disputes will happen. Buyers should include representations and warranties on title, absence of undisclosed investigations, and mandatory disclosure of holds and suspensions. They should also negotiate make-good replacement, escrow, or other remedies if credits become publicly contested after retirement, because litigation and enforcement around offsetting claims is increasing.
Even strong documentation cannot eliminate downstream controversy. Buyers need a playbook for updating claims and communications quickly, so questioned credits do not turn into avoidable greenwashing risk.
How to update corporate claims and communications to reduce greenwashing risk when credits are questioned
Claims should be reclassified before you are forced to do it. When a project enters review or a registry hold, absolute claims like “carbon neutral” create a clean target for challenge. A more defensible approach is time-bound, conditional language that focuses on financing verified mitigation or removals, with explicit reference to registry status and MRV updates, and clear boundaries on what is being claimed.
Uncertainty should be disclosed alongside governance actions. Buyers should state what happened, such as an investigation, suspension, or methodology update, and what they are doing in response, such as freezing use of the credits for claims, commissioning an independent review, or preparing a replacement strategy. Market integrity research supports the idea that omission risk is a major driver of reputational damage, especially when stakeholders can see retirements on registries.
Materiality thresholds should be defined for restatements. If credits support KPIs or public targets, buyers should predefine triggers for restatement, such as registry suspension, cancellation for overissuance, or legal injunctions affecting the project. Registry actions advancing independent reviews and cancelling excess credits provide real-world examples of why “wait and see” can become a governance failure.
Greenwashing controls should be operational, not just legal review. Marketing and legal teams should align on wording and proof requirements, and the organisation should maintain an “evidence room” with serials, retirement certificates, MRV documents, and due diligence memos ready for audit or challenge. Legal and policy commentary on market integrity and transparency trends supports expecting more scrutiny, not less.
Stakeholder remediation should be part of the response when rights issues surface. If FPIC or community benefit sharing is questioned, buyers should engage with affected communities and credible third parties, and pause new purchases from the developer until issues are clarified. Litigation and disputes around consultation make this step reputationally decisive.
Communications is defensive by nature. The next step is to redesign the portfolio so “pulled credits” become less likely, through higher-integrity sourcing, more durable credit types, and structures that reduce leakage and governance risk.
What to do next: portfolio strategy shifts toward higher-integrity credits, removals, and jurisdictional approaches
A quality segmentation policy should be written down and enforced. Buyers should define which credit types are acceptable for each use case, such as residual emissions, product claims, or contribution claims, and set minimum filters such as integrity labels, rating thresholds, and MRV completeness. Concentration limits by jurisdiction, developer, and methodology should be explicit. Market commentary in 2025 about buyers sorting by quality and durability supports treating this as a mainstream shift, not a niche preference.
Durability should be increased where claims are strongest. Buyers can move up the durability curve by increasing the share of removals, including engineered removals or nature-based removals with robust permanence management, and by using higher-risk avoidance credits more cautiously. Integrity research supports separating “neutralisation-style” claims from “contribution” financing when permanence and baseline uncertainty are higher.
Jurisdictional and nested approaches should be considered when drivers are systemic. When deforestation drivers are economy-wide, jurisdictional crediting can reduce leakage and align incentives with public policy, benefit sharing, and enforcement capacity. Reporting on alliances and initiatives to raise forest finance through jurisdictional REDD+ signals continued market development in this direction, even if implementation quality still varies.
Contracting should be designed for integrity, not just delivery. Buyers should prefer longer-term offtakes with MRV milestones, replacement and true-up clauses, mandatory disclosure of material events, and audit rights. Corruption risk research supports adding controls that reduce information asymmetry and increase consequences for non-disclosure. Buyers should also include claims-use restrictions, such as prohibiting “carbon neutral” use if a rating falls or a registry suspension occurs.
Continuous monitoring should be operationalised. Buyers should integrate remote sensing, rating updates, and watchlists for developers and jurisdictions, including investigations, litigation, and registry holds, into a shared risk dashboard for procurement, legal, and communications. Reporting on investigations in Brazil shows why this matters: it is the practical way to catch “toxic retirements” before they become headlines.
Tokenisation should be treated as plumbing, not quality. Digitising assets can improve audit trails and settlement workflows, but it does not fix additionality, baseline setting, or rights. Buyers should make it explicit internally that the risk remains in the underlying project, MRV, and governance, regardless of whether the credit is wrapped in a token.
