Impact Evaluation of the SHARPE Programme in Ethiopia: Academic Report

Abstract

The Strengthening Host and Refugee Populations in Ethiopia (SHARPE) programme uses a market systems development approach to promote increased self-reliance and economic opportunities for refugees and host communities through the piloting and scaling of interventions across different sectors. This approach is based upon understanding the economic barriers that refugee and host communities face, and working with key stakeholders – including businesses, government, and service providers – to improve market function for people in these regions. This report focuses on evaluating the impacts of investments SHARPE has made in the financial market system, which have focused on developing markets for digital financial services in refugee hosting areas. A highlight of the evaluation are two co-developed randomized control trials, designed to help SHARPE and its partners overcome constraints found while implementing the programme. The report finds evidence that robust markets for digital financial services are emerging in refugee hosting areas near Jijiga and are a little farther behind in Dollo Ado. Enrolment in the mobile money product, HelloCash, has been quite robust, though lower among women and refugees. HelloCash users are more likely to report financial inclusion (beyond inclusion through Hello Cash); they are 8.8 percentage points more likely to report being self-employed; they are 6.1 percentage points more likely to report typically having enough income; and they appear less food insecure than non-users. To try to enrol more women and refugees and catalyse HelloCash use among those groups, we conducted two randomized trials, one which allowed high volume customers to refer customers and receive a small bonus for doing so, and one which provided inactive customers with small incentives to start using the system. The former trial led to increased enrolment, but the share of women and refugees enrolling did not change; the latter led to increased use among women, but not refugees. We conclude with some ideas about further experiments to catalyse more use among refugees.