Algorithmic Exclusion and Financial Gender Biases within Nigerian Fintech Micro-Lending Apps

African Journal of Gender and Women Studies | Vol. 11, No. 9, August 2026 | pp. 249–256

DOI: 10.46882/2026/AJGWS/001279

Article Type: Original Research Article

Title: Algorithmic Exclusion and Financial Gender Biases within Nigerian Fintech Micro-Lending Apps

Names of Authors: Dr. Olumide E. Balogun¹, Dr. Chidinma O. Nze²

Authors’ Affiliations: ¹Department of Computer Science, University of Ibadan, Ibadan, Nigeria; ²African Institute for Mathematical Sciences, Lagos, Nigeria

Abstract:
This paper investigates automated credit-scoring algorithms utilized by smartphone-based digital micro-lending platforms in urban Nigeria, focusing on algorithmic gender bias. Utilizing a mixed-methods code-audit framework, we evaluated three commercial lending models with an anonymized historical dataset of 35,000 credit applications. The quantitative modeling reveals that female applicants are assigned a 32% lower creditworthiness score on average compared to male peers with identical income levels (p < 0.01). This automated disparity stems from predictive variables that penalize fragmented retail transaction histories and a lack of formal utilities registration, patterns highly common among informal female traders. Furthermore, machine learning pipelines use smartphone metadata proxies, such as contact-list sizes and digital wallet variety, that inadvertently reinforce systemic patriarchal exclusions. Qualitative focus groups with female merchants showed that automated algorithmic rejections offer no transparency, preventing applicants from understanding or contesting errors. This digital barrier leaves women locked out of short-term business capital, widening the economic gender gap within Nigeria's expanding fintech ecosystem. The study concludes that technological neutrality is a myth when models are trained on structurally biased datasets. We recommend implementing mandatory algorithmic fairness audits, establishing independent financial technology oversight boards, and integrating alternative qualitative data points into digital credit assessments.

Keywords: Fintech Lending, Algorithmic Bias, Machine Learning, Financial Inclusion, Market Women, Nigeria

Manuscript Timeline: Received: April 02, 2026; Revised: June 10, 2026; Accepted: July 14, 2026; Published: August 31, 2026.

Citation: Balogun, O. E., & Nze, C. O. (2026). Algorithmic Exclusion and Financial Gender Biases within Nigerian Fintech Micro-Lending Apps. African Journal of Gender and Women Studies, 11(9), 249–256. DOI: 10.46882/2026/AJGWS/001279

Subscribe to read the full article: https://internationalscholarsjournals.org/subscribe-to-read