Words Matter: The Role of Texts in Online Credit Markets

Journal of Financial and Quantitative Analysis, forthcoming

58 Pages Posted: 6 Jun 2014 Last revised: 14 Aug 2021

See all articles by Qiang Gao

Qiang Gao

Zicklin School of Business, City University of New York

Mingfeng Lin

Scheller College of Business, Georgia Institute of Technology

Richard W. Sias

University of Arizona - Department of Finance

Date Written: September 24, 2018

Abstract

We use debt crowdfunding data to examine how borrowers’ writing style is associated with lender and borrower behavior. Controlling for credit and auction characteristics, lenders bid more aggressively, are more likely to fund, and charge lower rates to borrowers whose writing is more readable, more positive, and contains fewer deception cues. Consistent with information garnered from writing driving the relation, controlling for credit and auction characteristics, borrowers whose writing is more readable, more positive, and has fewer deception cues are less likely to default. Investors, however, fail to fully account for the information contained in borrowers’ writing—especially deception cues.

Keywords: texts, peer-to-peer lending, crowdfunding, predictive analysis, sentiment analysis, subjectivity analysis, readability analysis, deception detection, machine learning

Suggested Citation

Gao, Qiang and Lin, Mingfeng and Sias, Richard W., Words Matter: The Role of Texts in Online Credit Markets (September 24, 2018). Journal of Financial and Quantitative Analysis, forthcoming, Available at SSRN: https://ssrn.com/abstract=2446114 or http://dx.doi.org/10.2139/ssrn.2446114

Qiang Gao (Contact Author)

Zicklin School of Business, City University of New York ( email )

17 Lexington Avenue
New York, NY 10010
United States
6463123192 (Phone)
6463123192 (Fax)

Mingfeng Lin

Scheller College of Business, Georgia Institute of Technology ( email )

United States

Richard W. Sias

University of Arizona - Department of Finance ( email )

McClelland Hall
P.O. Box 210108
Tucson, AZ 85721-0108
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
993
Abstract Views
8,812
Rank
42,450
PlumX Metrics