Administrative Data from Federal Disaster Recovery Loan Applications

This dataset underlies the Econometrica paper by Benjamin Collier and Cameron Ellis (2024), which estimates a demand curve for disaster recovery loans using administrative records of over one million applicants. The data include household-level application records from a federal loan program offered after natural disasters, combined with exogenous variation in interest rates from 24 natural experiments. The dataset enables replication of results regarding credit demand sensitivity, interest rate elasticity, and the impact of credit scores on loan take-up.