Measuring the impacts of agricultural input subsidies on fertilizer use, land allocation and forest pressure: Evidence from Malawi's 2009 Farm Input Subsidy Program.

摘要

This thesis investigates the impacts of Malawi's Farm Input Subsidy Program (FISP) on smallholder farmers' behavior, decisions, and outcomes. Four phenomena are studied: (1) use of fertilizer for maize production; (2) maize yields; (3) land allocation; and (4) forest clearing. The study uses cross-sectional data from 380 farm households in Kasungu and Machinga districts of Malawi. The FISP was implemented through a voucher system that targeted deserving households. To study the impacts of the FISP, a two-stage regression approach is used. In the first stage, selection into the subsidy program is treated as endogenous and conditional on household- and village-specific factors. A multinomial logistic regression is used to predict the probability of participation. In the second stage, Tobit regressions are used to examine the impacts of the subsidy program on fertilizer use and forest clearing. Subsequent to this, a production function for maize is used to measure differences in maize yields between program participants and non-participants. To examine the impacts of the FISP on land allocation, a system of three land share regressions is estimated.;Results suggest that the most vulnerable people in the communities studied were not the main recipients of the coupons, contrary to program design. Nevertheless, the results suggest that the subsidy program increased fertilizer use among participating households. Fertilizer use was found to be positively correlated with maize yields. In addition, farmers who planted improved maize seeds, such as those subsidized by the FISP, obtained higher yields than those producing traditional maize. The results also show that households that received coupons for maize inputs allocated 20% more land to maize than those that received no coupon. The analysis suggests that the program may have promoted intensification rather than extensification of maize and tobacco production in the two study areas. Households that participated in the Farm Input Subsidy Program cleared less forest land for agricultural expansion in the study year than those that did not, although those who received subsidies related to tobacco production had a program-induced derived demand for trees, which were used to construct tobacco drying sheds.