I provide below an overview of the results of the three outputs in the research agenda.
PROJECT: Screening on Loan Terms: Evidence from Maturity Choice in Consumer Credit.
1. Developed a theoretical model that rationalizes the use of credit maturity as a screening device. Maturity serves as a screening device because long maturity reduces the need to roll over debt at a higher price in the future. Higher risk borrowers, with an uncertain future, are willing to pay higher interest rates to secure this insurance.
2. Developed a methodology to detect adverse or advantageous selection in online credit platforms. The method is based on a difference-in-difference estimation that compares changes in the selection status of applicants that arise due to changes in the contract menu offerings of the platform, relative to always selected or always unselected samples.
4. Results: We find that the average default rate of short-maturity loans decreases by 0.8 percentage points when a long-maturity loan is available at origination relative to when it is not. This implies that borrowers who look identical ex ante from the investors’ perspective but who have a higher default risk self-select out of short-term loans and into long-term ones. We use testable implications of the model to determine the underlying nature of the demand for maturity.
Project: High-Cost Debt and Perceived Creditworthiness: Evidence from the U.K.
1. Developed a rational framework in which households’ future financial health is negatively affected by high-cost credit stigma, which arises because lenders pool together in the same high-risk bucket all borrowers that use high-cost credit.
2. Developed and applied a research design to identify the effect on future financial health of taking up a high-cost loans. We first document that taking up a high-cost loan has a causal negative effect on borrowers’ credit score, our measure of borrower’s creditworthiness as perceived by lenders, and access to credit, but does not affect borrowers’ repayment behaviour for up to a year after take-up. The use of high-cost credit can thus lead to a decoupling between perceived and actual borrower risk, which is a necessary condition for the risk-pooling mechanism to independently affect access to credit. Consistent with stigma, we show that take up of high-cost credit has an effect on access to credit when it leads to an update in the borrowers’ perceived creditworthiness, but it has no effect when no such update occurs (e.g. when the borrower’s perceived risk is already high).
Project: Measuring Bias in Consumer Lending
1. Developed new incentive-based principal-agent model of bias. In the model, bias arises because loan examiners are encouraged to maximize a short-term outcome, not long-term profits, due to principal-agent concerns. The incentivized focus on short-run outcomes leads to bias in lending against illiquid subpopulations for whom short-term and long-term outcomes diverge.
2. Develop an empirical test for bias in credit, building on Becker’s outcome test. The outcome test is based on the idea that long-run profits should be identical for marginal applicants from all groups if loan examiners are unbiased and the disparities across groups are solely due to omitted variables or statistical discrimination.
3. Design an empirical strategy to estimate marginal profits by group. We demonstrate that differences in profitability at the margin required for the Becker outcome test can be identified using variation in the approval tendencies of the quasi-randomly assigned loan examiners.
4. Results: we find significant bias against both immigrant and older loan applicants when using a measure of long-term profits that includes all the future cash flows from the borrower as an outcome. Additional results support our proposed incentive-based model of bias. For example, there is no evidence of bias against immigrant or older applicants when we implement the Becker outcome test using the short-run default outcome used to evaluate examiner performance.
