Measuring the Welfare Cost of Asymmetric Information in Consumer Credit Markets

How to cite

DeFusco, Anthony; Tang, Huan; Yannelis, Constantine. 2021. "Measuring the Welfare Cost of Asymmetric Information in Consumer Credit Markets." AEA RCT Registry. AEARCTR-0008183. https://doi.org/10.1257/rct.8183-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Anthony DeFusco* , Huan Tang , Constantine Yannelis

* Primary investigator

Registry ID
AEARCTR-0008183
DOI
https://doi.org/10.1257/rct.8183-1.1
First registered
Last updated
Status
Completed
Trial dates
2018-01-01 – 2018-03-31
Countries
China ()
Registry record
http://www.socialscienceregistry.org/trials/8183

Abstract

Information asymmetries are known in theory to lead to inefficiently low credit provision, yet empirical estimates of the resulting welfare losses are scarce. This paper leverages a randomized experiment conducted by a large fintech lender to estimate welfare losses arising from asymmetric information in the market for online consumer credit. Building on methods from the insurance literature, we show how exogenous variation in interest rates can be used to estimate borrower demand and lender cost curves and recover implied welfare losses. While asymmetric information generates large equilibrium price distortions, we find only small overall welfare losses, particularly for high-credit-score borrowers.

Intervention

The experiment we analyze was designed independently by our data provider to examine loan applicants’ sensitivity to borrowing costs and was implemented from January to March, 2018. The platform randomly selected 11,180 potential borrowers into the experiment who had successfully applied and qualified for credit during this window. Selected applicants were then randomly divided into two equal-sized groups: one control group and one treatment group. Applicants in the control group were offered standard financing terms, while the treatment group received a 40 percent reduction in their overall cost of borrowing. Data was then collected for each of these loan applicants on whether they took up the loan and their subsequent repayment behavior.

Experimental design

The platform randomly selected 11,180 potential borrowers into the experiment who had successfully applied and qualified for credit during this window. Selected applicants were then randomly divided into two equal-sized groups: one control group and one treatment group. Applicants in the control group were offered standard financing terms, while the treatment group received a 40 percent reduction in their overall cost of borrowing. Data was then collected for each of these loan applicants on whether they took up the loan and their subsequent repayment behavior.

Design details

Randomization method
Randomization done electronically by the lender.
Randomization unit
Randomization was at the loan-applicant level.
Clusters
11,180 loan applicants
Observations
11,180 loan applicants
Arms
5590 loan applicants
Primary outcomes
Take up: An indicator for whether the loan application took up the loan Charge-offs: The fraction of promised loan payments never received by the lender.

Cited by

No citation data available for this registration yet.

Keywords