Does participation in public works encourage fertilizer use in rural Ethiopia?

Abstract

A sixth of the world's population receives inadequate nutrition.1 The problem is especially severe in Africa where agricultural sectors are dominated by subsistence farmers. African smallholder farmers could double yields by doubling their fertilizer use. 2 Yet, in many countries, subsistence farmers do not utilize technologically advanced agricultural technologies that are apparently available to them at subsidized prices. Extreme poverty and the lack of effective access to disposable income in the aftermath of shocks are likely to be a partial determinant of the low rates of technology adoption among smallholder farmers in Africa. One strategy to address this situation is the provision of economic assistance through food aid programs. This thesis evaluates the indirect impact of food aid programs on agricultural productivity via changes in participants' input decisions. This project also identifies the main determinants of participation in public works programs and fertilizer use and examines whether these decisions affect one another. Our contribution to the literature is twofold. First, we apply novel econometric techniques to correctly compute the marginal effects of a system of two simultaneous models with binary and censored latent dependent variables. Second, we use a unique cross-sectional sample of households from rural Ethiopia that permits us to examine the relation between fertilizer use and participation in a recently established safety net program. Our results show no evidence that participation in public works programs and other income support programs adversely affect the adoption of technology. In fact, the results suggest that the programs, to some extent, encourage adoption. We also show that previous choices of both fertilizer use and participation in public works, educational attainment, household characteristics, income-related variables, and some regional agro-ecological factors are among the main determinants of the participation and usage decisions. Although the empirical results support the argument that participation in safety net programs may help reduce poverty and the risks of food insecurity via changes in the participants' production habits, the results also indicate that there is room for improvement in the distribution of food aid.