Why one developer’s collapse moved voluntary carbon prices and sentiment globally
KOKO mattered because it looked like a systemic supplier in clean cooking, not a single project. It operated a network of roughly 3,000 fuel points and served more than 1.5 million households in Kenya, so many market participants treated it as a proxy for scale and bankability in household energy credits. That perception amplified the impact when operations shut down in late January and early February 2026.
The trigger was not only operational. The shock was tied to a regulatory gate: the reported refusal of a Letter of Authorization in Kenya, which meant KOKO could not sell credits with authorization under Article 6 and lost the upside associated with “compliance-like” demand that some buyers were pricing in.
Pricing reacted in a two-tier way because the market had been valuing two different products. In the CORSIA Phase 1 channel, spot prices showed volatility as narratives of scarcity collided with new supply entering the market. In the pure voluntary channel, traders estimated that KOKO credits without a LoA could fall toward about $3 to $4 per tCO2e, which is a fast repricing of regulatory delivery risk rather than a slow reassessment of climate impact.
Sentiment spread because KOKO represented a financial model as much as a project type. A carbon-centric revenue model, combined with cookstove methodology risk, sovereign authorization risk, and operational continuity risk, is a cluster that procurement teams can easily generalise to the whole cookstove vertical when they need to make quick portfolio decisions.
Portfolio behaviour followed. Buyers and intermediaries with exposure to household devices, including cookstoves and household water, started to revisit risk haircuts on forward delivery and on credits tied to contested narratives. Preference increased for credits perceived as more defensible, including CCP-eligible approaches and projects with measured usage.
The uncomfortable question is why the shock was so believable. The answer sits in the technical vulnerabilities of cookstove crediting, and in how quickly those vulnerabilities can turn into reputational and contractual risk.
Cookstove credits under the microscope: baseline risk, usage uncertainty, and over-crediting narratives
Cookstove crediting is highly sensitive to a few inputs that are hard to observe at scale. Baseline selection matters because the emissions you claim to avoid depend on what fuel and stove you assume households would have used. Stacking matters because households often use multiple cooking options in parallel. Usage rate matters because “distributed” is not the same as “used”, and small errors in usage assumptions can translate into large errors in credited tCO2e.
Buyers got a clear signal from academic scrutiny. A UC Berkeley study published in Nature Sustainability (January 2024) reviewed five dominant methodologies and a sample of 51 projects across 25 countries. The sample was described as representative of about 40% of credits issued at the time of selection (May 2023). The study found pervasive over-crediting linked to simplified accounting practices.
Public narratives then compressed nuance into headline risk. Media coverage framed the issue as overstatement on the order of about 10x, which created a reputational hazard for corporate buyers and for funds using Article 6 or voluntary claims, even though individual projects can differ materially in design and monitoring.
Methodology owners pushed back and also pointed to transition already underway. Gold Standard stated it had updated rules in 2021 with more conservative defaults and safeguards, including caps on key inputs such as baseline fuel and usage. That matters because the market is not static. Some supply is legacy and survey-based, while newer supply is moving toward measured approaches.
Procurement implications are practical and immediate. Buyers increasingly need to separate legacy credits that rely heavily on surveys and assumptions from credits supported by metering or measured usage and tighter controls, especially where CCP eligibility and “high integrity” claims are part of the value proposition.
The next step is to understand what standards and registries are actually tightening, because those changes determine which vintages remain claimable, which credits become harder to sell, and which projects can still finance growth.
What standards and registries may tighten next: monitoring, digital MRV, and claims controls for household energy projects
CCP eligibility is becoming a segmentation tool, not a blanket endorsement of cookstoves. ICVCM granted the Core Carbon Principles label to a limited set of cookstove methodologies, including Gold Standard Metered & Measured, GS TPDDTEC, and Verra VM0050, with additional conditions. For buyers, the key point is that CCP eligibility applies to a subset of methodologies and requirements, not to “cookstoves” as a category.
Most existing supply does not automatically clear that bar. Reporting cited that by late 2024, around 64% of cookstove offsets available were based on methodologies judged insufficiently rigorous for CCP, while credits from approved methodologies were a small share. That imbalance is why methodology transition can move both prices and liquidity.
Contract details now depend on methodology conditions and dates. For VM0050 CCP, Verra’s guidance includes conditions such as using fNRB values from CDM Tool 33 (latest version) for emission reductions up to 31 December 2025. That kind of cut-off is not academic. It affects which monitoring periods and vintages can credibly be represented as meeting a given label expectation.
Digitalisation is also changing the operational risk profile. Verra has been digitalising methodologies for submission through its Project Hub and Digital Project Submission Tool, with announcements in January 2025 aimed at standardising data and reporting. Gold Standard has also moved on digital MRV, including pilots approved in late 2024 and early 2025, with the direction of travel pointing toward more structured data capture, including usage or sales-linked evidence.
Claims controls are tightening alongside MRV. Verra released updated CCP label guidance, effective 15 July 2025 for verification requests, which increases the need to track exactly which tranche of credits is CCP-labelled, under what conditions, and what must be disclosed to end clients.
With segmented supply and stricter claims rules, buyers now need a due diligence approach that captures delivery risk, reputational exposure, and vintage-by-vintage quality, rather than treating cookstoves as a single interchangeable bucket.
Buyer due diligence after Koko: how to assess delivery risk, reputational exposure, and vintage quality in cookstove portfolios
Authorization status is now a hard gate, not a footnote. Buyers targeting Article 6 outcomes should verify the host-country LoA status and conditions, and align contract language on corresponding adjustments. Kenya’s Climate Change (Carbon Markets) Regulations 2024 formalise requirements around authorisation, consistency with national processes, and interaction with national infrastructure, which is exactly the kind of framework that can turn into a delivery constraint if not met.
Pricing and contract design should reflect two different assets. Buyers can no longer treat “CORSIA-tagged potential” and “VCM-only” as a minor attribute difference. The KOKO case shows how quickly the market can reclassify a credit from premium expectations to voluntary levels around $3 to $4 per tCO2e, with liquidity risk on top. Practical protections include milestone-based payments, termination rights tied to loss of authorisation or label, escrow structures, and clear remedies if verification or authorisation gates are not achieved.
Vintage and methodology gating should be explicit and documented. Buyers should build an eligibility matrix by methodology and version, monitoring period, and any date-bound conditions such as tool versions and cut-offs. This reduces the risk of buying a vintage that cannot support the intended claim, whether that claim is CCP labelling or a broader “high integrity” positioning.
Reputational exposure needs a policy, not ad hoc debate. The Berkeley findings and the broader public narrative mean risk committees will ask why a portfolio includes credits from methodologies associated with over-crediting concerns. A defensible approach is to set minimum evidence thresholds, such as metering or robust dMRV, and to exclude or cap exposure to higher-risk methodologies unless there is strong project-specific justification.
Operational continuity is part of climate integrity in household energy. Usage drives issuance, and usage depends on distribution, maintenance, customer support, and the ability to keep the service running. Buyers should diligence implementer resilience, including financial stress signals, dependence on a single monetisation channel, and regulatory concentration that can create a single point of failure.
Once buyers demand covenants and protections, the developer-side question becomes unavoidable. How do you finance working capital and receivables when carbon revenue is volatile, label-dependent, and sometimes gated by sovereign authorisation?
The finance lesson for developers: working capital, receivables, and the danger of carbon revenue as the core business model
Carbon revenue growth does not equal resilience when it depends on a sovereign gate. TechCabal reported that KOKO’s UK carbon arm grew rapidly, with revenue of £39.8m in 2024 up from £1.8m in 2023, and then entered administration in February 2026 after the LoA shock. The lesson is not about scale. It is about fragility when monetisation depends on authorisation status and market labels.
Working capital is structurally hard in household energy projects. Costs are front-loaded, including hardware, last-mile distribution, PAYG systems, and field operations. Cash-in from carbon is delayed by the monitoring period, verification, issuance, sale, and retirement cycle. If the market demands CCP-aligned approaches and more rigorous dMRV, MRV costs and timelines can increase, which tightens the working capital squeeze.
Receivables can behave like options if contracts are not designed carefully. Offtake agreements can become conditional if buyers can walk away when authorisation fails or labels are lost. Developers then face the risk of holding inventory that must be discounted or cannot be sold at all. More robust structures include limited prepayment, volume flexibility, clear representations on registry and label status, and risk-sharing mechanisms that do not leave one side carrying all the downside.
Regulatory concentration risk is now a finance risk. Kenya’s 2024 regulations formalise LoA and national registry requirements. If a developer’s business case depends on “authorised” value, execution risk becomes legal and institutional, including process compliance, reporting, and other conditions that can delay or block monetisation.
Investors and operators are likely to push for a different model. Clean cooking can still be financeable, but it becomes more robust when unit economics of the underlying service stand on their own, with carbon as upside rather than the core business model. Otherwise, a methodology downgrade or a tightening of claims rules can cascade into covenant breaches and insolvency.
If the sector wants to keep funding clean cooking at scale, the closing question is what happens next to household energy credits. The answer will likely involve consolidation, methodology transition, and a more explicit pricing of authorisation and measurement quality.
What comes next for household energy credits: consolidation, new methodologies, and pathways to regain market confidence
Methodology transition is likely to accelerate toward measured and higher-scrutiny approaches. VM0050 and Gold Standard’s metered pathways are part of that shift. CLEAR, linked to the Clean Cooking Alliance and 4C, was described as under review as of December 2025 for UNFCCC Article 6.4 and for voluntary standards, which signals a real pipeline but not immediate market-wide replacement.
Supply-side consolidation is a plausible outcome of higher MRV and governance costs. Developers with weaker MRV and limited capital may exit or be acquired, while buyers concentrate volumes with operators that can sustain metering, dMRV, QA, and stronger governance. That concentration can reduce the risk of a KOKO-like single point failure, but it can also reduce short-term supply.
Repricing is likely to become more structural. The market is already separating legacy commodity-like credits from CCP-labelled or metered credits, with spreads driven by integrity, delivery risk, and authorisation status. The KOKO episode made one point hard to ignore: authorisation is a pricing attribute, not a legal detail.
Data transparency will be a key trust lever. Digital submission tools and dMRV can reduce documentation errors and increase auditability. Buyers should expect more requests for device-level or statistically robust usage evidence, stronger controls on stacking, and more public disclosure that helps defend credits against over-crediting narratives.
A practical playbook is emerging for market confidence. Buyers can segment portfolios by methodology, vintage, and label; use contracts with milestones and remedies; maintain claims-ready reporting aligned with CCP guidance; and engage early on LoA, registry processes, and benefit-sharing expectations. That combination lets companies keep financing clean cooking while avoiding risks that the market is now pricing in immediately.
