Total Factor Productivity of Agricultural Commodities in the Economic Community of West African States: 1961-2005

This study examines total factor productivity growth and its decomposition for rice, cotton and millet producing countries in ECOWAS. The productivity measures were estimated using Stochastic Frontier Analysis (SFA) and Data Envelopment Analysis (DEA). The data cover a 45-year period (1961-2005) separated into pre-ECOWAS (1961-1978) and ECOWAS (1979-2005) periods in order to study the effects of ECOWAS reforms on productivity growth of the selected crops. Calculations are based on data collected from FAOSTAT database, International Rice Research Institute (IRRI) world rice statistics, and International Cotton Advisory Committee database. The data include output of each crop (rice, cotton and millet) and seven input variables comprising land area, labour, seed, fertilizer, tractor use, irrigation and country dummies. The results for both SFA and DEA show that: (i) there are inefficiencies but productivity progress among ECOWAS member nations producing rice, cotton and millet; (ii) though magnitudes of the inefficiencies and productivity progress vary across models applied and by segmentation of the data set, there is little or no conflict in the overall results; (iii) technical change has had the greatest impact on productivity, indicating that producers have a tendency to catch up with the front runners; and (iv) the total factor productivity in ECOWAS and pre-ECOWAS sub-period differ across crops depending on the models applied. In general, policy makers should try not to be indifferent with respect to the approach used for efficiency and productivity measurement, as these may give different results.