This paper investigates links between the strength and type of social networks, as a collective or community characteristic, and private forest conservation activity in rural Nepal.

Using rural household data from the Nepal Living Standards Survey, this paper tests the effects of alternative social network (SN) indices on the number of trees planted on private land. To handle excess zeros and overdispersion in the data, the zero-inflated negative binomial (ZINB) model is used while testing and correcting for endogeneity in the SN indices.

Findings are consistent with the theory that social networks, such as community forest user groups, can have positive effects not just on community forest management but also spillovers or externality effects that enhance private conservation effort. This generalization requires careful qualification.

Social networks directly related to conservation activities increase tree planting on private land. Whereas, social networks not related to conservation activities exhibit rivalry. While there has been considerable discussion in the planning and development literature about building and enhancing social networks, the policy implication is that not all social networks are created equal when it comes to enhancing critical conservation activities.