Lending in Need: Natural Disaster Shocks and Online Borrowing in Indonesia

Abstract

Natural disasters can generate urgent household liquidity needs, yet little is known about how borrowing through digital lending platforms responds to disaster exposure. Using approved-loan records from a major FinTech platform in Indonesia, this study examines the relationship between natural disasters and online borrowing through fixed-effects panel analysis and a supplementary matched difference-in-differences analysis. The results show that disaster exposure is positively associated with borrowing amounts at both the provincial and individual levels, while the relationships with new-borrower participation, loan duration, and overdue amounts are not statistically significant. Further analysis of the December 2021 and December 2022 Semeru eruptions reveals a positive short-term borrowing response, with a coefficient of 0.027 for log-transformed daily borrowing and larger estimates during the first six days following the eruptions. Although the matched analysis provides exploratory evidence within a short event window, its findings are consistent with the broader pattern observed in the panel analysis. Overall, the study highlights borrowing amounts as a key margin of adjustment in digital lending following natural disasters, offering new evidence on how FinTech borrowing responds to localized economic shocks.