Metin Cakir

Date of Award


Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)


Resource Economics and Policy


Timothy J. Dalton

Second Committee Member

George K. Criner

Third Committee Member

Hsiang Tai Cheng


Increasing rice production in West Africa has been one of the most important policy objectives of West African governments. This is because enhancements in production are crucial for improving the economies of the region as well as fighting against poverty and malnutrition. This research contributes to this development objective by conducting efficiency analyses of irrigated rice production. The general objective is to examine the efficiency, and the factors that influence the efficiency of irrigated rice farmers, in Mali. A special focus is devoted to quantifying the effects of irrigation scheme rehabilitation. To reach this goal I conduct a stochastic parametric frontier analysis using a cross-sectional data set collected by the Human Health Consortium associated with West Africa Rice Development Association, (WARDA) in 1996- 1997. Data from 147 households from 5 different villages located in the 'Office du Niger' cultivating 398 rice plots are examined. I employed four different functional forms along with the three different specifications of the error distribution to estimate the best-practice production frontier. The results show that the estimates are not considerably different across the different specifications of the error distributions. However, they are particularly sensitive to the choice functional form. I also conduct statistical tests to choose the most appropriate functional form that fits the data. It was found that the mean technical efficiency of farmers located in villages with rehabilitated irrigation schemes is around 4.5 percentage points lower than their counterparts in villages with non-rehabilitated irrigation infrastructure. It was also found that the technical improvements in the irrigation scheme help to intensify production and provide diverse opportunities of cultivation. Lastly, I analyzed the correlates of technical inefficiency. I employed two different approaches: the one-step approach and the two-step approach. I found that the inefficiency of farmers is not related to managerial factors but rather they are related to the technological factors, such as inefficient quantity composition of the input set, or insufficiency of infrastructure. These analyses also demonstrated that the two-step approach, although commonly used in the literature, provides biased estimates that can potentially mislead policy. A clear message of this research is that extension services should be directed separately for heterogeneous areas, i.e. rehabilitated and nonrehabilitated areas. It was found that there is more need for technology improvements in non-rehabilitated areas that can enhance the productive use of seed and nitrogen. As this research shows, this can be done by improvements in infrastructure. On the other hand, for rehabilitated areas there is more room to increase rice production by making a better use of the existing technology. For these villages policies should be directed to increase the productivity of manuallabor through extension education or labor-saving technologies.