| Sumario: | The management of forest commons requires coordination within a community and between communities. This coordination is usually challenging given the incentives for free riding. The literature suggests that communities with heterogeneous members, particularly from different ethnic groups, are less effective in commons management and public goods provision. In this paper, we empirically test if forest degradation differs within and between ethnically diverse communities. Using data for the majority of villages in rural Gambia, our first set of results suggest that ethno-linguistic fractionalization within villages correlates with a reduction in forest cover and an increase in forest loss over a period of 15 years. Then, we propose a new approach to assess forest management between heterogeneous communities. Forest units within exclusively shared territory between pairs of neighboring villages are defined by dyadic polygons, which are used to apply dyadic regression methods. In our second set of results, we find evidence suggesting that spaces shared between ethnically distant communities exhibit higher forest cover and lower forest loss. Further analyses suggest that the effect of ethnic distance on reduced forest degradation is present mainly when villages are heterogeneous in other characteristics as well. Our study motivates the identification of relevant spaces for the analysis of competition over local commons, providing a novel approach of potential interest to applications beyond our area of study.
|
|---|