Unlocking credit access: Using non-CDR mobile data to enhance credit scoring for financial inclusion

Rouzbeh Razavi and Nasr G. Elbahnasawy

Finance Research Letters, 2025, vol. 73, issue C

Abstract: A significant portion of the global adult population, particularly in developing markets, lacks access to formal credit due to the absence of traditional credit histories. This presents a major challenge for financial institutions, FinTech companies, and policymakers aiming to promote financial inclusion. While conventional credit scoring models are built on established financial data, the growing penetration of mobile phones offers an alternative means to assess credit risk. Unlike prior research focused on Call Detail Records (CDRs)—data generated by telecommunication providers capturing users' call and message activities, such as duration, frequency, and timing—this study investigates the predictive power of a broader spectrum of mobile usage data, including non-CDR attributes like social media engagement and web browsing habits, in assessing credit risk. Using a broad range of machine learning algorithms on actual mobile usage data from over 1,500 demographically diverse individuals over a two-week period, we find that while these mobile usage attributes alone cannot fully replace FICO scores in regression models (R²=0.30), they significantly enhance the accuracy of classification models, especially when combined with CDR data (Accuracy=0.89). These findings have important implications for credit markets in emerging economies, pathways for financial institutions and FinTech companies to engage with unbanked populations and support the growth of alternative credit assessment tools.

Keywords: Credit risk; Non-CDR data; Mobile data; Financial inclusion; Fintech; Machine learning; Unbanked populations (search for similar items in EconPapers)
JEL-codes: G21 Q01 R51 R58 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1544612324017112
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:finlet:v:73:y:2025:i:c:s1544612324017112

DOI: 10.1016/j.frl.2024.106682

Access Statistics for this article

Finance Research Letters is currently edited by R. Gençay

More articles in Finance Research Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().