Banking the Unbanked: Using Grocery Data for Credit Decisions
53 Pages Posted: 8 Jul 2021 Last revised: 3 Aug 2022
Date Written: August 2, 2022
Many consumers across the world do not have access to credit and one of the key reasons is that lenders do not have sufficient data to assess the creditworthiness of those consumers. This paper evaluates the potential of a new type of alternative data source to predict consumers' creditworthiness: grocery transaction data. Our analysis takes advantage of a unique, individual-level match of credit card data and supermarket loyalty card data, which allows us to build a credit scoring algorithm that incorporates grocery data. We find that grocery data can improve out- of-sample predictive accuracy of missing a credit card payment by up to 12.5%. We also find that consumers who do not have credit scores or are lower-income are more likely to benefit from the use of grocery data. Overall, our findings suggest that grocery data may serve as a channel through which traditionally under-served consumers in credit markets can signal their creditworthiness to lenders.
Keywords: Grocery data, Alternative data, Financial inclusion, Consumer finance, Credit cards market, Habits, Economics of data
JEL Classification: D12, D18, G40, G51
Suggested Citation: Suggested Citation