Description

Abstract: The effects of alternative methodological approaches on estimates of technical efficiency (TE) are investigated, using data for 501 groundnut farmers in Senegal (1982-2000). Meta-analysis of information of previous studies shows that models using the translog functional form produce higher average TE values than those using a Cobb-Douglas model. Also, models using data envelopment analysis (DEA) and stochastic frontiers tend to yield significantly higher average TEs than deterministic frontier models. It is demonstrated that the impact of distorted markets should be taken into account in the formulation of agricultural policies designed to increase agricultural productivity. It was also found that larger peanut farms are more efficient than smaller ones, regardless of optimizing behaviour. Farmers tend to employ more labour than needed and could benefit from having additional land to cultivate. They also use more inputs than the profit-maximizing farmers and receive prices for their peanut output that are lower than market prices. Therefore, successful policies toward improving groundnut production in Senegal should allocate more land to farmers and should promote practices that enhance, or at least maintain, soil fertility. KEYWORDS: TROPAG | Arachis hypogaea | farms | efficiency | production functions | economic models | farm size | Arachis hypogaea | soil management | farm inputs | West Africa | Senegal.