Structural estimation of spatial spillover effects of cash transfers

Enhancing evidence transferability

Douglas Gollin

University of Oxford

Institute of Anthropology, Gender and African Studies, University of Nairobi

June 2020 to December 2021

Secondary data analysis

Kenya, Uganda, Lesotho, Tanzania

This evaluation will examine the causal spillover effects experienced by non-recipients of cash transfer programmes in rural, developing country contexts. It will do this by developing and applying middle-range theory to address a method gap for the estimation of spillover effects in the presence of partial economic integration of local markets.

The study will model and estimate the extent of economic integration of local markets and how this affects variation in spillover effects over space and across different groups of non-recipients. It will use structural model-ling to estimate the general equilibrium effects within a network.

The data falls into three categories:

1. Secondary quantitative data from the Aspirations Project, a recently completed randomised controlled trial of GiveDirectly’s unconditional cash transfer program in Western Kenya;
2. Primary qualitative data to be collected in Kenya; and
3. Publicly available, fully anonymised secondary data from impact evaluations in other settings and countries.

The study will enable decision-makers to make better-informed policy choices based on an understanding of the causal impact of cash transfer programmes on people other than direct recipients.