A stochastic meta-frontier approach to estimating the impact of cooperatives membership on rice farmers’ efficiency: Contrasting results from Senegal

Using cross-sectional data from 835 rice-farming households in Senegal, we investigated the extent to which membership in farmers’ cooperatives affects farm technical efficiency. To do so, we combine the propensity score matching method with the sample selection stochastic frontier model (Greene, 2010) and the stochastic meta-frontier approach (Huang et al., 2014). The propensity score matching helps in mitigating biases from observable variables. The sample selection stochastic frontier framework was used to control for biases arising from unobserved characteristics in the production frontier. Using the meta-frontier approach, farmers’ technical efficiency were estimated and compared. Results show that cooperative membership contributes significantly in improving rice production. However, when considering group-specific frontiers (farmers operating in their own benchmark: members vs non-members), cooperatives members do not technically perform better than non-members. Furthermore, when considering the meta-frontier estimates, significant differences in technical efficiency between members and non-members can still be observed in favour of non-members.