A Simulation Impact Evaluation of Rural Income Transfers in Malawi and Ghana

Abstract

A considerable body of experimental economics research examines the impacts of cash transfers (a treatment) on recipient households (the treatment group). In many developing countries, though, cash transfers are insignificant compared to other transfer mechanisms in terms of their claim on public resources. For example, in Malawi, fertilizer subsidies dwarf cash transfers, while next door in Zambia, the government pays farmers prices well above market levels for their maize. Yet no study to our knowledge has attempted to compare the full impact of cash payments and other kinds of transfers on rural incomes and welfare in low income countries. Economic theory is unclear on the effectiveness of alternative transfer schemes in a context of imperfect markets; thus, empirical impact analysis is required to analyze, select and design income-transfer mechanisms. It is not clear how to design a feasible randomized experiment or econometric model to compare the efficiency of a variety of alternative transfer schemes with both direct and indirect impacts on a heterogeneous rural population. This paper employs a simulation model of heterogeneous, interacting agents to compare the impacts of direct payments and alternative transfer mechanisms on production, incomes and welfare in rural Malawi and Ghana. We calibrate our simulations to existing fertilizer subsidy schemes in the two countries: the Malawi Agricultural Inputs Subsidy Program (MAISP) and Ghana‘s temporary input subsidy program, initiated in 2008 and continued every year since. In each country, we then compare the input subsidy to two other transfer schemes: a market price support for staples, similar to what historically has been implemented in both countries, and cash transfers (the Malawi Social Cash Transfer Scheme, SCTS, and Ghana‘s Livelihood Empowerment Against Poverty, LEAP)