Résumé : There has been growing interest lately in the role of microfinance to support environmental management of micro-enterprises and poor households. Worldwide, the number of green microfinance projects increases, yet there seems to be little discussion on how effective green microfinance is in achieving its environmental goals. This paper aims to position itself in this debate. We look at the first large-scale green microfinance programme for biodiversity conservation: Proyecto CAMBio. It consists of a combination of credits, technical assistance and conditional payments for environmentally friendly agricultural activities (PES). We focus on its implementation in Nicaragua by the microfinance institution FDL and the NGO Nitlapan. We perform an in-depth econometric analysis of a survey we conducted on a sample of 128 rural producers. We assess the clients’ characteristics that influenced the evolution of the environmental value of their farm –as defined by the indicators we used– on a span of five years, and we assess Proyecto CAMBio’s possible role in this evolution. Moreover, we further look into the effectiveness of PES in rewarding environmental betterment. Factors such as the decision to change the main economic activities, or clients’ strategies or opportunities in land accumulationappear to have the strongest influence on the evolution of the environmental value of the clients’ farm. While the project per se, even if carefully implemented in agreement with its guidelines and well performing at financial level, does not appear to have significantly influenced the evolution of the environmental value of the clients’ farm. Moreover, the PES does not seem to reward environmental improvement while instead it rewards the more credit-worthy activities, producers with more access to land and credit and in addition producers that plant fewer trees per hectare. With these results, we underline the importance of the local territorial dynamics and the complexity of the socio-environmental systems against a vision based simply on single economic actors. From our results it appears that green microfinance, without strategic articulation with local actors and broader territorial dynamics, would tend to (indirectly) support preexisting socioeconomic structures and the possibly related environmental degradation processes. We hence call for a more proactive engagement of green microfinance in the territorial dynamics and with local actors with the aim to support more sustainable livelihood trajectories and development pathways.