|
Title:
|
FILTERING P2P LOANS BASED ON CO-UTILE REPUTATION |
|
Author(s):
|
Abeba Nigussie Turi, Josep Domingo-Ferrer, David Sánchez |
|
ISBN:
|
978-989-8533-56-2 |
|
Editors:
|
Hans Weghorn |
|
Year:
|
2016 |
|
Edition:
|
Single |
|
Keywords:
|
P2P Lending, Co-utility, Reputation, Trust |
|
Type:
|
Full Paper |
|
First Page:
|
139 |
|
Last Page:
|
146 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Peer-to-peer loans suffer from a mistrust effect, which is amplified by the information asymmetry underlying the P2P online lending market. In this work, we assume that a reputation capital can act as a trust-building intangible collateral in such a market. We leverage a distributed reputation protocol based on the co-utility principle, whereby peer interactions should be mutually beneficial and hence self-enforcing. In this co-utile protocol, rational peers cooperate to compute each others reputation scores. Reputation scores of borrowers are computed based on the outcome of direct transactions. Then, the reputation mechanism helps filter credible borrowers based on their respective reputation scores. By using an experiment on a simulated platform with a randomly selected sample of loans from LendingClub, we show that this protocol can improve the efficiency of the P2P online lending market by filtering out defaulting borrowers. Moreover, the results show that there is a negative correlation between the probability of default and the global reputation of the borrowers, which implies that the protocol helps accurately identify each borrowers type. |
|
|
|
|
|
|
