UC Berkeley

The number of voluntary interventions seeking to generate carbon offsets by reducing deforestation and forest degradation, generally known as REDD+ projects, has increased significantly over the past decade. Offsets are issued based on project performance in comparison to a baseline scenario representing the expected deforestation in a project area in the absence of REDD+. Baselines from most ongoing REDD+ projects were established following four methodologies approved by the largest voluntary carbon offset certification scheme worldwide, the Verified Carbon Standard (VCS) from Verra. These methodologies often rely on oversimplified assumptions about deforestation that remain overlooked by project developers, certification bodies, and buyers. Here, we explore what these methodological assumptions are and their implications. We then construct alternative deforestation baselines for four ongoing VCS-certified projects using the four VCS-REDD+ methodologies and examine how they differ. Overall, we observe large discrepancies among the project baselines. On average, the highest baseline value we calculate for each project is more than 14 times greater than the lowest value across the four projects studied. This illustrates the lack of robustness and consistency across the VCS-REDD+ methodologies. The results also call into question the additionality of carbon offsets issued based on these methodologies. New baseline methods need to be urgently developed if voluntary REDD+ projects are to reliably estimate their additional contribution to climate change mitigation. The incorporation of causal inference methods represents current best practices in measuring the efficacy of REDD+ interventions. Regrettably, these methods remain largely overlooked by project developers, certification standards, and governmental and international bodies. Dynamic baselines developed by independent analysts could potentially enable project developers to distinguish the impacts of the REDD+ intervention from confounding factors and properly estimate additionality.