Identifying performance benchmarks in Ghanaian agriculture through efficiency analysis

Agricultural production in Ghana is mainly carried out by smallholder farmers on a subsistence basis. Smallholders constitute about 95% of the farming population and produce 80% of the annual output. This study investigates the level of technical efficiency of a sample of 294 households from the Upper East region of Ghana. Technical efficiency and its determinants are investigated using parametric stochastic frontiers and nonparametric data envelopment analysis methods. This is the first study that uses bootstrap DEA and Bayesian frontier methods in efficiency analysis of agriculture in Ghana. Results from our analysis show that mean technical efficiency is low and there is significant variability in efficiency among the sample farms. There is significant positive correlation of efficiency estimates from nonparametric and parametric models which indicates the robustness of the results. The results imply that agricultural productivity can be increased substantially through improvement in technical efficiency. From a policy perspective, age (as a proxy for farming experience), educational status, use of hired labour and farm size, have been found to hold the greatest potential for improving technical efficiency in Ghanaian agriculture. Policy implications from the analysis of the determinants of technical efficiency are sensitive to the methods used to estimate technical efficiency.