According to figures from the World Bank, approximately 1.4 billion people worldwide remain unbanked, meaning they have minimal or no access to formal credit. A key reason for this exclusion is the lack of a formal credit history, which remains a cornerstone of decision-making for most traditional lenders. Without this financial footprint, individuals find themselves unable to access loans or credit cards, which in turn prevents them from ever establishing the credit history that financial institutions demand. This vicious cycle effectively locks vast populations out of economic participation and limits their ability to build stable financial futures.
New research from the University of Notre Dame offers a potential solution by demonstrating how alternative data — specifically retail transaction data — can be used to construct reliable credit scores for individuals who lack formal credit records. The study, “Who Benefits from Alternative Data for Credit Scoring? Evidence from Peru,” which is set to appear in the Journal of Marketing Research, presents a compelling case for using consumer behaviour data as a viable substitute for traditional credit metrics. The authors argue that such an approach provides a scalable and inclusive path toward financial participation, particularly for those long overlooked by conventional credit systems.
The research was led by Joonhyuk Yang, Assistant Professor of Marketing at Notre Dame’s Mendoza College of Business, alongside co-authors Jung Youn Lee from Rice University and Eric T. Anderson from Northwestern University. Their findings reveal that retail purchase data — such as shopping frequency, product choices, and responses to promotions — can significantly increase credit approval rates for individuals with no formal credit history. In their Peruvian case study, approval rates for these “no-history” applicants rose dramatically, from just 16 per cent to between 31 and 48 per cent, when such data was factored into the credit scoring process.
This work builds upon the team’s earlier research, which showed that grocery shopping behaviours could predict credit card repayment reliability among those who already had credit histories. For example, consumers who regularly purchased unhealthy or convenience foods, like cigarettes and ready-to-eat meals, had a higher risk of defaulting, whereas those who bought fresh ingredients and demonstrated budget-conscious shopping patterns were more likely to make payments on time. Although this previous study was limited to consumers already within the credit system, the current research significantly expands the scope by focusing on those entirely excluded from it.
To conduct the new study, the researchers partnered with a large Peruvian retailer that operates across multiple sectors. Using loyalty card data and sales records, they tracked the shopping behaviour of over 45,000 customers who made at least one purchase during two years. This behavioural data was merged with other conventional metrics, such as utility bill payment records and entries in Peru’s national credit registry. The combined dataset allowed the researchers to create credit profiles for individuals with and without credit histories and to simulate approval decisions across different lending scenarios, including those focused on risk reduction and credit expansion.
Perhaps most notably, the study found that retail data had minimal impact on approval rates for individuals who already had strong credit histories — those approval rates stayed at around 88 per cent. However, for applicants without formal histories, the impact was profound. As Yang noted, “Retail data barely moves the needle for people who already have credit scores, but it’s a game changer for those who don’t. That’s where inclusion happens.” This research supports the idea that alternative data can effectively bridge the gap between unbanked individuals and the credit system, allowing lenders to make informed, low-risk decisions while enabling more people to build financial identities. In doing so, it lays vital groundwork for a more equitable and inclusive financial future.
More information: Joonhyuk Yang et al, Who Benefits from Alternative Data for Credit Scoring? Evidence from Peru, Journal of Marketing Research. DOI: 10.2139/ssrn.4852032
Journal information: Journal of Marketing Research Provided by University of Notre Dame