Funding Decisions in Online Marketplace Lending
HKIMR Working Paper 01/2020
29 Pages Posted: 8 Jan 2020 Last revised: 5 Mar 2020
Date Written: January 9, 2020
This study analyzes more than 28 million recent loan listings on LendingClub, one of the world’s largest online marketplace lending platform. Using tree-based machine learning, we develop robust predictive representations of funding decisions on this fintech peer-to-peer lending platform. We find that a borrower's employment length is the main factor in the preference of lenders making funding decisions. The significant role of employment length is consistent with the widespread use of the lending platform to obtain better refinance for existing obligations. Requested amount and the existing leverage of a borrower are secondary in lenders' consideration. The credit pricing charged on a funded listing fully depends on the loan grade assigned by LendingClub. Monetary policy seems to have little impact on funding decisions on this platform.
Keywords: Financial Technology; Fintech Lending; LendingClub; P2P Lending; Peer-to-peer Lending; Shadow Banking
JEL Classification: G21; G23
Suggested Citation: Suggested Citation