Date of Award


Level of Access Assigned by Author

Campus-Only Thesis

Degree Name

Master of Science (MS)


Resource Economics and Policy


Timothy J. Dalton

Second Committee Member

Kathleen P. Bell

Third Committee Member

George K. Criner


The social and economic costs of malaria are substantial in Africa. At the root of the problem, malaria debilitates a large share of the labor force and causes the death of millions of children. As a secondary effect, malaria reduces GDP per capita and reduces economic growth (Gallup and Sachs, 1998). This current research assesses, in the first step, factors that affect illness (malaria and other short-term diseases) in an important rice producing area of Mali. A survival model is employed to identify the factors that impact the number of workdays lost per illness episode. The results of the survival model showed that illness length is shorter for individuals who received a malaria treatment, analgesic, or a traditional treatment than for individuals receiving a combination of traditional and medical treatments. Agricultural tasks were also found to have significant impacts on illness length. The number of workdays lost per illness episode were lower during the planting period and greater during field preparation as wells as during weeding and bio-chemical applications. These results suggest two findings. First, agricultural activities with an important timing dime~isions, uch as planting, can force i l l individuals to work even when they have not recovered from an illness. Second, agricultural activities that do not require to be performed at a specific time may be delayed by the presence of illness in the household. A Poisson Model is used to identify the factors that influence the number of illness episodes per household during the wet cropping season. This model is employed to explain the count of illness for family members who do not participate in rice cultivation. The Poisson model indicates that adults' literacy has a positive effect on the number of illness cases. This suggests that education improves the ability to recognize illness symptoms and hence, the likelihood to report sick family members. Results suggest that households where several family members received malaria treatment may signal the lack of prophylaxis use. Wealth, proxied by the value of agricultural equipment or the total value of farm assets, reduces illness count per household. These two models were employed to generate predictive values which were included in the second step of this thesis as direct and indirect measures of illness. The second step consists of estimating labor demand ~nodclsto determine if illness translates into labor loss. Illness was found to reduce family labor as well as total labor which is composed of hired and family labor. The results also indicate that is important to consider household colnposition when quantifying the impact of illness on labor. In a region where food security and dependence on rice import are important development issues, this thesis suggests, assuming a positive marginal productivity of labor, that improving health would increase rice production in Mali.