The stochastic frontier production function is used to estimate agricultural efficiency index. Then, controlling for household characteristics and other exogenous variables, the efficiency index is regressed on the probability of being sick. Estimation is performed using the treatment effect model where the probability of being sidelined by sickness is modeled as a probit. This framework allows policy simulations that underscore the impact of farmers’ health status on both agricultural efficiency and poverty reduction. Overall, regression results confirm the negative impact of health impediment on farmers’ agricultural efficiency. Simulation results show that improving farmers’ agricultural efficiency by investing in farmers’ health may not necessarily lead to poverty reduction. Additional policy instruments may be needed to achieve simultaneous increase in agricultural productivity and reduction in poverty rate.
A stochastic production frontier approach
International Food Policy Research Institute (IFPRI)