Category: Credit Risk

On the Rise of the FinTechs—Credit Scoring using Digital Footprints

Read Paper

This paper evaluates users’ digital footprints or the information that people leave online by accessing or registering a website to predict consumers’ likelihood of default. Using more than 250,000 observations, the authors show that information gleaned from people’s digital footprint can be equal to or even exceed the information content of credit bureau scores.

FDIC Center for Financial Research Working Paper No. 2018-04

Tobias Berg, Valentin Burg, Ana Gombovic and Manju Puri

Data point: The geography of credit invisibility

Read Report

This paper examines geographic patterns to assess the extent to which where one resides is correlated with one’s likelihood of remaining credit invisible. The paper explores the following topics: credit deserts, credit invisibility in rural and urban areas, entry products by geography, and credit invisibility.

The Bureau of Consumer Financial Protection’s Office of Research

Kenneth Brevoort, Jasper Clarkberg, Michelle Kambara, and Benjamin Litwin.

What Do a Million Observations Have to Say About Loan Defaults? Opening the Black Box of Relationships

Read Paper

The paper evaluates the impact of aspects of customer-bank relationships on loan default rates. Using a dataset that consists of more than 1 million loans made by 296 German banks, the research finds that banks with relationship-specific information from customers establishing transaction accounts are less likely to experience loan defaults and act differently in screening and monitoring behaviors than banks with no information.

Manju Puri, Jörg Rocholl, Sascha Steffen

Predictably Unequal? The Effects of Machine Learning on Credit Markets

Read Paper

This paper uses various types of machine learning models to predict credit risk using historical mortgage data. It finds gains in predictiveness that would likely lead to an increase in approvals across all demographic groups, but that average prices could increase for African-American and Hispanic borrowers due to differences in risk calculations.

Andreas Fuster, Paul Goldsmith-Pinkham, Tarun Ramadorai, and Ansgar Walther.

Consumer Lending Discrimination in the FinTech Era

Read Paper

This paper evaluates differentials between borrowers of different races in loan approval rates and pricing between fintech and traditional mortgage lenders. The paper finds that unexplained pricing differentials are smaller among technology-heavy lenders and that such differentials overall have declined as the mortgage industry as increased reliance on algorithmic lending in recent years.

UC Berkeley Public Law Research Paper

Robert Bartlett, Adair Morse, Richard Stanton, Nancy Wallace.

Join our Newsletter

Translate »