According to a latest research conducted by Joonhyuk Yang, Assistant Professor of Marketing, Mendoza College of Business, University of Notre Dame and Jung Youn Lee, Assistant Professor of Marketing, Jones Graduate School of Business at Rice University, how people shop and what they buy at the grocery stores can predict whether they pay their credit card bills on time.
“As marketing professors, we wanted to learn about alternatives to traditional credit scores. So, we teamed up with a multinational conglomerate that, among other things, runs a large supermarket chain and a credit card issuer. By analysing consumer-level data from those two business units, we could see how 30,089 individuals shop and manage their finances,” the duo stated.
The experts, during their research, found that people with more consistent grocery shopping habits are more likely to pay their credit card bills on time. These are the kind of consumers who prefer shopping on the same day of the week, while spending about the same amount monthly, as they buy similar items across trips and take advantage of deals regularly. The study also established a link between people’s shopping behaviour and their financial health. For example, shoppers who frequently purchase cigarettes or energy drinks are likely to miss credit card payments. Those who often buy fresh milk or salad dressing tend to be more diligent about paying their bills.
“In general, buying healthier but less convenient food predicted responsible payment behaviours. This was true even when we held consumer characteristics such as income, occupation, credit score and family size constant,” said Yang and Lee, based upon their findings, the duo has now developed a credit scoring algorithm that scores people based on their grocery shopping habits along with traditional credit risk indicators. When the algorithm was simulated with approval decisions, it was discovered that using grocery data could help lenders predict defaults more accurately while boosting their per-customer profits.
According to the World Bank, more than 1 billion people worldwide lack access to formal financial systems and, as a result, have no credit scores. In the United States alone, about 45 million adults have no credit history or not enough of one to generate a score. This factor makes it hard for these individuals to access credit, even if they are responsible borrowers. Without credit, it’s harder to get a car, a job or even a house.
“It’s a problem that disproportionately affects underprivileged groups, including people of colour and women. In response, policymakers and researchers are increasingly interested in using alternative data sources to assess creditworthiness. For instance, Fannie Mae (housing financing agency) now considers mortgage applicants’ rent payment histories, allowing those without traditional credit histories to demonstrate their creditworthiness,” the experts noted further, while adding, “Grocery data is especially promising because there’s so much of it. Pretty much everybody buys groceries, and not just once. Information about consumer preferences is continuously being generated in every aisle of grocery stores around the globe. Our study shows that this data has value far beyond the grocery industry.”
As per Yang and Lee, the study will now serve as a proof of concept, offering insights for the design and implementation of future research. However, several key questions remain. For example, what if this approach affects different groups unequally? And what about privacy concerns?
“Our follow-up research aims to address these issues. We’re collaborating with a conglomerate in Peru, a cash-reliant country with a significant unbanked population. Building upon our current findings, we’re working closely with that company to test the impact of our approach on low-income populations. We’ll be helping to evaluate credit applicants using retail transaction data, aiming not just to improve profitability but also to boost social inclusion in the region,” the duo concluded.