Your Movement in a City Reveals Your Credit: Credit Default Prediction Based on Geolocation Information

Do individuals at high risk of credit default frequent different locations within a city than those at low risk? Leveraging large-scale geolocation data, we propose that geosimilarity risk and geolocation network size serve as novel and informative classifiers for predicting individual credit default. We define two individuals as geosimilarity network (GSN) neighbors if they share at least one visited location during a given period. Using consumer location traces combined with loan repayment histories from a leading FinTech company, we find that the GSN neighbors of a defaulter are approximately three times more likely to default than the average borrower and about 4.5 times more likely to default than the GSN neighbors of a non-defaulter. Moreover, geosimilarity risk and geolocation network size significantly explain default outcomes after controlling for traditional factors such as demographics, financial capacity, and loan characteristics. Incorporating these geolocation-based measures into standard credit risk models improves predictive accuracy by approximately 9 percent. These findings highlight the value of spatial mobility data as a complementary source of information for credit risk assessment.