Impact of Digital Credit on Household Consumption in Sub-Saharan Africa

Thesis Abstract

Digital credit has emerged as a transformative financial tool in Sub-Saharan Africa, with potential implications for household expenditure patterns, liquidity management, and welfare outcomes. The study addresses the problem of limited empirical evidence on how digital credit access influences household consumption decisions in diverse Sub-Saharan African contexts, considering heterogeneous effects across income groups, urban-rural settings, and credit maturities. The aim is to quantify the causal and correlational relationships between digital credit utilization and household consumption, while unpacking channels such as precautionary saving, credit-constrained demand, and liquidity effects. Specific objectives are to (i) assess the association between digital credit uptake and total and discretionary per capita consumption, (ii) examine heterogeneous effects by income deciles, gender of household head, and urbanicity, (iii) evaluate the role of repayment behavior and interest rates in moderating consumption changes, and (iv) identify policy-relevant mechanisms through which digital credit may influence resilience to income shocks. The methodological approach adopts a mixed-methods design anchored in a quasi-experimental framework. A panel dataset combining waves from the Demographic and Health Surveys and country-level digital credit provider statistics across five Sub-Saharan African countries (Nigeria, Kenya, Ghana, Tanzania, and Uganda) will be constructed for 2018–2023, comprising approximately 12,500 households. A propensity score matching technique will be employed to create comparable treatment and control groups based on digital credit exposure, followed by difference-in-differences estimation to identify causal effects on monthly household consumption, controlling for household characteristics, macroeconomic conditions, and credit-market factors. Complementary instrumental variable analyses will be applied where endogeneity concerns persist, using exogenous shocks to mobile money ecosystems and regulatory changes as instruments. To enrich interpretation, semi-structured interviews with 40 financial behavior respondents across three countries will be conducted, analyzed thematically to reveal micro-foundations of consumption adjustments and risk-management strategies. The analytical toolkit includes OLS and panel fixed-effects regressions, quantile regression to explore effects at different points of the consumption distribution, and robustness checks with placebo tests and alternative specifications. The theoretical lens integrates the credit constraint view, the precautionary saving channel, and the liquidity shock framework, drawing on the Life-Cycle Hypothesis and the Homo Economicus paradigm in dynamic contexts, complemented by information asymmetry and digital financial inclusion theories. Key expected findings anticipate that digital credit access is associated with a statistically significant increase in average monthly per capita consumption for lower-income households, driven by reduced liquidity constraints and improved ability to smooth consumption in the face of irregular income. The study expects heterogeneous effects, with stronger consumption amplification for urban households and male-headed households, and attenuated effects for higher-income groups where liquidity constraints are less binding. It is hypothesized that higher interest rates and shorter tenors may attenuate consumption gains due to repayment burdens, while flexible repayment schedules and transparent pricing will enhance the positive consumption impact. The research also expects that households with evidence of prudent repayment behavior exhibit larger consumption improvements, indicating that credit quality moderates welfare outcomes. The study contributes to knowledge by providing robust, cross-country evidence on the household-level welfare implications of digital credit, delineating the conditions under which digital credit serves as a consumption-smoothing mechanism versus a debt-induced expenditure risk. It advances methodological practice in development economics by integrating quasi-experimental causal inference with qualitative insights to map channels from digital credit to consumption. Policy implications include guidance for designing user-centered digital credit products with transparent terms, responsible lending practices, and regulatory frameworks that support financial inclusion while mitigating over-indebtedness. The main conclusion anticipates that responsibly managed digital credit can enhance household consumption smoothing and welfare for the most financially vulnerable, provided there is adequate consumer protection, appropriate repayment structures, and strong financial literacy. Recommendations emphasize targeted outreach to low-income and rural households, standardization of pricing disclosure, and collaboration between regulators, fintechs, and social protection programs to maximize positive consumption and resilience outcomes.