The Impact of Malaria on Land Use -- Empirical Evidence from Amazon

Presenter: Shufang Zhang, Harvard University

Abstract

Malaria, as a major source of disease burden in the developing world, presents a big barrier for development. Besides a few studies documented malaria’s impact on agricultural outputs, little is known about the other pathways that the disease hinders development and poverty reduction especially in rural areas.

Our study explores the impact of malaria on land use as a result of impaired labor productivity due to the disease. Using four waves of household survey data and remote sensing data of land use in a resettlement project area in Brazilian Amazon between 1985 and 1995, the study investigates how the burden of malaria impacts on the probability of a plot being occupied and on the land use type once a plot was occupied. To address the measurement error and potential endogeneity of reported malaria rate, the variable is instrumented by the distance to the major river in the area and the distance to the forest reserves, given the fact that water helps the breeding of mosquitoes and the aboriginal residents lived close to the forest reserve were the reservoir of parasites.

Higher malaria is found to lower the probability of a plot being occupied or lived on. More specifically, the probability of a plot to be occupied if free of malaria is 0.37 higher than a plot with malaria rate of 50%. In addition, higher malaria rate in the previous year significantly lower the probability of a plot to be occupied in the future.

Malaria is also found to hider the productive use of the land. Nested logistic model results show that higher malaria rate lowers the probability of a plot to be cleared and further lowers the probability of the cleared area to be used for agricultural production. This effect translates into a 13% less of a plot to be cleared and planted for a plot with malaria rate of 50% than one that is free of malaria.