The Impact of Credit and Training on Farmers Efficiency: A Semi‐Parametric Meta‐Frontier Analysis

Smallholder farmers in developing countries face several constraints, which affect their productivity. To reduce these constraints and enhance productivity, government and non‐governmental agencies implement programmes that provide credit and training to farmers. In this study, we evaluate the impact of one of such programme on the efficiency of farmers. Specifically, we evaluate the impact of providing both credit and training on the efficiency of peanut farmers in Haiti. To achieve this goal, first, we use the Propensity Score Matching (PSM) technique to balance our dataset, thereby accounting for selection bias based on observable factors. Second, we correct for selection bias from non‐observable factors by using the Selectivity‐bias Correction Stochastic Production Frontier (SC‐SPF) to estimate the efficiency scores on the balanced sample. Third, given that there might be differences in the technology used by farmers, we estimate a meta‐frontier to compare the efficiency scores while addressing the incorrect skewness problem in frontier analysis. Our results show that farmers who received both credit and training had higher efficiency scores, suggesting that interventions that increase credit access and transfer knowledge can enhance productivity and help to address food security issues.