From the CDM to Article 6 – are we learning from or repeating past carbon market mistakes?
As the world transitions from the Kyoto Protocol’s Clean Development Mechanism (CDM) towards the Paris Agreement’s Article 6.2 and 6.4 mechanisms, carbon markets are once again being promoted as a key tool for climate mitigation and sustainable development. But are the same governance failures, land conflicts, and social injustices that characterised earlier offset mechanisms now just being reproduced under new institutional labels?
Evidence from projects implemented under the CDM and the voluntary carbon market (VCM), together with analysis of emerging Article 6 regulations – which are unlikely to fundamentally alter current implementation practices – suggests that the risks facing future projects are very real. While climate mitigation projects can generate important environmental benefits, they also require vast amounts of land. Crucially, in many low- and middle-income countries, these lands are not empty or unused – as they are often purported to be – but are in fact already governed through customary tenure systems that remain poorly recognised in official records.
A growing concern is therefore not only whether carbon markets fulfil their claims of reducing emissions, but also whether they perpetuate patterns of exclusion, weak participation, and land dispossession that have long accompanied large-scale land investments.
Lessons from the CDM and the VCM: Formal compliance, real-world conflict
The CDM was the first large-scale global offset mechanism under the Kyoto Protocol. It enabled projects in developing countries to generate tradable carbon credits that could be used by industrialised countries to meet mitigation targets. However, many of these projects operated in land-intensive sectors, such as afforestation, reforestation, hydropower, and other renewable energy sectors, and used formal land classifications that failed to capture how land was actually used and governed on the ground.
Evidence from the Land Matrix Initiative, for example, shows that projects were often implemented through concessions, leases, or land acquisitions in areas officially labelled as “state land”, “degraded land”, or “vacant land” – yet these same lands were frequently used by Indigenous Peoples and local communities for farming, grazing, firewood collection, or cultural practices under customary systems not recognised in cadastral records.
In Kenya’s Mau Forest, for instance, a CDM project has been implemented, even though the area is marked by longstanding land conflicts and contested claims. Likewise in Brazil, the Plantar forestry CDM project relied on eucalyptus plantations established on so-called terras devolutas (“vacant lands”), despite longstanding collective use by geraizeiro communities.
These cases reveal a broader structural problem: CDM reporting systems relied heavily on what could be formally documented, audited, and verified. So while stakeholder consultations, compensation schemes, and legal land status could all be recorded within project design documents, procedural compliance often obscured unresolved land conflicts, unequal power relations, and weak community participation.
Despite the decline of the CDM, rather than these problems disappearing in its wake, they evolved within the rapidly expanding VCM. The Land Matrix analytical report on land-based offset projects shows that carbon offsetting through the VCM has in fact become a major driver of large-scale land acquisitions (LSLAs) across the Global South. Of 217 documented land deals linked to offset projects, 183 are already registered in carbon credit registries and together cover approximately 8.8 million hectares – an area nearly the size of Austria. Moreover, the report demonstrates that while these projects are usually promoted through the language of “nature-based solutions” and “co-benefits”, in reality their expansion has intensified pressure on land rights and in some cases resulted in severe land conflicts, particularly in regions characterised by weak governance systems and customary tenure arrangements.
The report further highlights that dominant certification standards, particularly Verra’s Verified Carbon Standard (VCS), frequently contain only vague safeguards regarding benefit-sharing, Free, Prior and Informed Consent (FPIC), and community participation – and even then, although the standards mention consultations and safeguards, implementation often depends on the discretion of project developers. Notwithstanding that current revisions of the standards are headed in the right direction, many of the older projects still work under versions that only weakly protect local livelihoods and land rights.
From CDM and VCM to Article 6.2 and 6.4
The shift towards the Paris Agreement’s Article 6 mechanisms creates a critical governance challenge in light of these earlier carbon offset projects being absorbed, rather than replaced, by new institutional structures. Many existing CDM projects have already requested transition into Article 6.4, referred to as the Paris Agreement Crediting Mechanism (PACM), while numerous VCM projects could be authorised or linked through Article 6.2 arrangements as carbon markets become increasingly interconnected. This poses an uncomfortable possibility: unresolved governance failures from both the CDM and VCM may simply migrate into the new Paris Agreement architecture.
For example, a report by Carbon Market Watch (CMW) and the Land Matrix, which raises serious concerns regarding the safeguards embedded in Article 6.4, demonstrates that while the Article’s Sustainable Development Tool includes mandatory environmental and human-rights safeguards and grievance procedures, it remains largely procedural in many instances. According to the report, key weaknesses include weak recognition of customary rights and informal tenure systems; insufficient safeguards where Indigenous Peoples lack formal recognition; narrow interpretations of FPIC; and unclear hierarchy between international human rights standards and weak domestic legal systems.
These weaknesses matter because many projects in the Article 6.4 pipeline potentially overlap with lands used or claimed by Indigenous Peoples and local communities. This includes 309 CDM projects that applied for transition and 117 projects that submitted a prior consideration notification (PCN) for inclusion in the mechanism. Importantly, the report shows that these are not hypothetical concerns: many of these projects are already associated with deficient consultation, involuntary resettlement, inadequate compensation, and unresolved land conflicts.
While Article 6.4 has at least developed some centralised safeguard structures, the risks under Article 6.2 may be even greater. Article 6.2 operates through bilateral or multilateral agreements between countries and relies heavily on domestic legal systems, reporting structures, and nationally determined procedures. In addition, the system is primarily designed around accounting of carbon emissions, transparency, and verifiable reporting of mitigation outcomes. This creates a structural limitation.
Moreover, unlike Article 6.4, Article 6.2 does not include a centralised UN mechanism with harmonised safeguard procedures. Instead, participating countries have substantial flexibility in how cooperative approaches are designed and implemented. In practice, this means that critical issues such as land rights, consultation processes, benefit-sharing, and grievance mechanisms are largely delegated to domestic institutions and bilateral agreements.
As previous experiences from the CDM and VCM demonstrate, formal compliance with reporting requirements does not necessarily reflect how projects are experienced on the ground. Projects may document consultations and compensation plans while underlying issues – including customary land claims, unequal power relations, or unresolved conflicts – remain insufficiently addressed.
If Article 6.2 mechanisms rely primarily on domestic administrative systems and formal reporting categories, there is a serious risk that many underlying conflicts and inequalities will remain invisible within official reporting structures. This is especially concerning in countries with weak land governance systems or limited recognition of customary tenure rights.
In addition, because Article 6.2 governance is highly decentralised, safeguards may vary substantially across countries. In contexts with weak governance, politically marginalised communities may have limited opportunities to challenge projects or seek redress. There is therefore a risk that Article 6.2 could reproduce – and potentially even scale up – many of the governance problems already observed under the CDM and VCM.
A critical juncture for carbon markets
The transition from the CDM to Article 6 represents a pivotal moment in global climate governance. As carbon markets expand rapidly in the coming years, so too will land-intensive mitigation projects in all likelihood. But the evidence from the CDM and VCM is clear: climate mitigation projects implemented without strong land governance safeguards can deepen existing inequalities, intensify conflicts, and undermine the rights of Indigenous Peoples and local communities.
The key lesson is not that all carbon projects are inherently harmful. Rather, it is that procedural compliance alone is insufficient. Consultation without genuine participation, formal land records without recognition of customary rights, and benefit-sharing without fairness or accountability do not prevent injustice. Policymakers need to address the risks rooted in Article 6.4 in its current form. In addition, policymakers should be vigilant with respect to the even less regulated Article 6.2 mechanism.
If Article 6 mechanisms fail to address these structural weaknesses, the Paris Agreement risks repeating many of the mistakes of the past – and at a potentially much larger scale.