Communities in the study area were heterogeneous concerning forest dependency, income and occupation.
The different forms of capital assets were closely linked to co-management, with social capital having the most significant influence while human and manufactured capital had no role in the governance of co-management.
Within social capital, local leaders played a dominant role over the other variables.
Poor dependent people were less likely to be selected as participants, which undermined the essence of co-management.
The impact of co-management on forest vegetation cover was found to be positive during the project period. However, a slight decrease was observed at the end of the project indicating a challenge of sustaining this approach.
Co-management in the Rema-Kalenga Wildlife Sanctuary was evaluated to assess how fairly capital assets were considered in the beneficiary selection, and to what extent it affected vegetation cover. A semi-structured questionnaire survey was employed to collect necessary information along with satellite images. The study revealed that many variables of social capital and a few variables of natural and financial capital played a significant role in the participant selection process. Analysis of dependency showed that the participants did not rely significantly on the forest both in terms of resource collection and their monetary value implying that the most dependent people were not adequately represented in the co-management team. The dominance of local leaders suppressed the voices of others in the management venture. An increase in forest vegetation cover was observed during the project period, although shortly after the end of co-management projects a slight deterioration of forest cover was noted. The findings of the study can serve as a guide in the future application of community forestry programmes in protected areas of Bangladesh and elsewhere in the world within similar socio-physiographic settings.