Rethinking the design and implementation of financial services for poverty reduction: A case of Northern Ghana

Abstract

The thesis empirically examines how microfinance products are designed and implemented, and the implications for clients’ households and sources of livelihood. The study argues that the design of products and implementation that reflect the livelihood needs and poverty context of clients is one of the effective ways to reduce poverty. It investigates the microfinance operations of three financial institutions: Sinapi Aba Trust (SAT), St Joseph’s Cooperative Credit Union (CCU) and Sonzele Rural Bank (SRB) in Jirapa, a municipality in Northern Ghana. The study deployed a mixed-methods approach to collect data from six rural and urban communities. Data was sought from secondary sources, 20 interviews, 10 focus group discussions and 120 questionnaires. The research adopted the Sustainable Livelihoods and the Making Markets Work for the Poor approaches as a guide in the framework of analysis. The study, using qualitative and quantitative analytical tools found that product designs of SAT and SRB did not reflect the needs and poverty context of the majority of their clients. Clients of SAT and SRB were found to be less involved in the product design processes, suggesting a top-down institutional approach that seldom incorporated the needs of the poor. The method of group formation has a substantial implication on members’ poverty outcomes. Groups involving only females had a significant and positive relationship with members’ household and business outcomes, while members of male-only groups had a negative relationship with their household outcomes. The thesis concludes that accessible interest on loans and incentives to encourage savings would make microfinance markets work more sustainably for the rural poor. The findings challenge a reconsideration of the design of microfinance products to integrate financial technology as an efficient approach to deliver financial services, especially in rural areas.