Working Paper: CEPR ID: DP14918
Authors: Christoph Basten; Steven Ongena
Abstract: How does banks’ geographical footprint change when a FinTech platform allows offering mortgages to regions without branch presence? Unique data on responses from different banks to each household yield three salient findings: First, banks offer 4% more often and 6 basis points cheaper when markets have high versus low concentration, implying more profitable follow-on business. Second, they offer 2% more often and 2 bps cheaper when unemployment or house price growth in the applicant’s state are one standard deviation less correlated with those at home, improving portfolio diversification. Third, these offers are increasingly automated, using available hard information more efficiently.
Keywords: mortgage lending; spatial competition; credit risk; diversification; banking; automation; fintech; online pricing
JEL Codes: G2; L1; R3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Reduction in lending by major Swiss banks (UBS and Credit Suisse) (F65) | Lower local market concentration (L49) |
| Lower local market concentration (L49) | Banks are more likely to make an offer (G21) |
| Lower local market concentration (L49) | Banks provide rates that are cheaper (G21) |
| Unemployment or house price growth correlation with bank's home state (R29) | Banks offer more often (G21) |
| Unemployment or house price growth correlation with bank's home state (R29) | Banks provide rates that are cheaper (G21) |
| Automation of offers (D26) | Banks utilize hard information more efficiently (G21) |