The State of Safeguards in the VCM: Are We Protecting What We Should? 

When integrity is discussed in the voluntary carbon market (VCM), it is still largely defined by carbon metrics. Even the controversies that have drawn the most scrutiny — from over-crediting to overstated climate claims — are typically framed in terms of carbon performance. Ask what makes a “high-quality” carbon credit, and you will likely hear terms such as additionality, baselines, permanence, or leakage. 

Far less attention is paid to how projects affect people and ecosystems. Yet harms to communities, weak consent processes, inadequate grievance mechanisms and/or unclear land tenure arrangements have also shaped the market’s credibility. 

These concerns point to a deeper structural issue: safeguards exist, but they often do not function as intended.  

Safeguards are meant to prevent social and environmental harm, protecting communities and ecosystems from impacts such as loss of livelihoods, health risks, or environmental damage. To date, Calyx Global has assessed environmental and social risks across more than 100 carbon projects, spanning different technologies, geographies, and standards. So, what have we learned about how safeguards are actually applied in practice? Here are key takeaways. 

Safeguards are still treated as a checklist, not core to project design 

Across many projects, safeguards are included in project documentation primarily to pass validation rather than to genuinely shape project design. When approached this way, the focus shifts away from outcomes on the ground to whether the right boxes have been ticked. 

In practice, this often means stakeholder engagement is reduced to a one-off activity, or that grievance mechanisms exist on paper but are difficult for affected communities to access or trust. Safeguards may be documented, but they seldom change how projects distribute risks, benefits, and decision-making power. 

Verification bodies often face constraints in assessing safeguard implementation, including limited field time, uneven social expertise, and reliance on project-provided documentation. As a result, even material safeguard risks may go unflagged. 

Timing compounds the problem. Safeguards are frequently introduced late in project development, once technical and financial decisions are already locked in. At that point, even well-identified risks can be difficult to meaningfully address. Consequently, safeguards are often an afterthought, rarely serving as a condition for determining whether a project should go ahead at all. Even where safeguards are documented at validation, ongoing monitoring of social and environmental performance is typically less rigorous than carbon monitoring, limiting visibility into how risks evolve over time. 

From protecting people to protecting credits 

A more concerning shift is emerging in how safeguards are understood. Increasingly, they are framed as tools to manage reputational exposure, liability, or delivery risk. In other words, to protect the carbon credit itself. This focus is not surprising given recent scrutiny of the VCM. But it fundamentally alters the purpose of safeguards. In practice, meaningful consultation is not only a requirement but also a condition for a successful project. Projects with trusted, sustained participation will build local buy-in and support long-term project outcomes.  

Safeguards were designed to address power imbalances and prevent irreversible harm. When they are reduced only to risk-management instruments, their ability to identify and prevent deeper structural issues weakens. Based on our experience, reputational risk is rarely just an “optics” problem. Rather, the issues are often rooted in deeper structural problems, such as unresolved grievances, weak participation, inadequate consent processes, or underestimated environmental impacts. These are precisely the issues safeguards are meant to identify early in the project design process. When these structural weaknesses go unaddressed, they undermine not only the social legitimacy of the individual project but can undermine the long-term credibility and resilience of the market itself. 

Signs of improvement, but still some way to go 

There are signs that the market’s approach to safeguards is beginning to shift. Buyers and offtakers are paying closer attention to safeguard-related risks, and some standards have strengthened their requirements. These developments matter. But they do not yet address the core challenges. 

Strengthened requirements do not automatically translate into stronger implementation. Safeguards remain inconsistently applied, introduced too late, and treated as secondary to carbon performance. Auditors and validation bodies do not always have the tools or expertise to rigorously assess safeguards performance, even when they are requirements of the standard. As the VCM continues to mature, safeguards cannot be relegated to one-time checklists. If the market is to deliver on its integrity objectives, safeguards must be treated not as supporting documentation, but as substantive conditions that protect communities and ecosystems.