The global banking industry has undergone a wave of mergers over the last two decades and, as a result of which, mega banks have arisen in many countries around the world including Korea. Korean banks respond to this recent trend of bank consolidation by lowering the deposit interest rate and raising the loan interest rate (Park, 2004). Another major change had taken place even before the megabank trend. After the government allowed financial intermediation services by non-bank financial institutions in the late 1990s, commercial banks had started to reduce making loans to individual consumers since the monitoring costs are relatively higher than the interest income. As a result, banks have become more conservative in making loans to borrowers with low credit. In contrast to banks that strictly adhere to credit evaluation models, investors of online P2P (Peer-to-Peer) lending use much more flexible methods to judge borrowers' qualifications for a loan, which include ‘hard information factors,' ‘social interaction factors,' and ‘soft information factors.' Naturally, borrowers with low credit ratings use online P2P lending as one of the available funding sources. In this paper, we explore factors of success in online P2P lending repayment. In particular, this study examines how each of these three most common factors- ‘hard information factors', ‘social interaction factors', and ‘soft information factors'-- affect the loan repayment. This paper uses data from 542 loans from a leading crowd funding company in Korea from June 2008 through November 2010. We use a number of questions and answers, pre-poll results, and a number of attempts for making loans as the proxy variables of social interaction factors. Unlike foreign online P2P lending, Korea's internet lending takes place through the social relationship between borrowers and investors, formed by communicating via the online bulletin board. We obtained the following empirical results. Firstly, borrowers' personal bankruptcy experience has a significantly positive impact on the success of loan repayment. This is because borrowers with experience in personal bankruptcy have an incentive to improve their credit ratings through sincere loan repayment because a typical online P2P lending system uses a savings bank account as an escrow account. Therefore, they can improve their credit ratings for loans from commercial or savings banks by maintaining good credit standing online. These results imply that people with poor credit can use online P2P lending not only for raising money but also for improving their credit quality. Secondly, the purposes of loans for housing, living, and medical expenses have a significantly positive relationship with the success of loan repayment. This means that if they get loans for essential living expenses, they would exert greater efforts to repay the loans. Thirdly, debtors, who actively share personal information and interact with lenders online through P2P auction site message boards, are more likely to repay their loans than borrowers who have less social interaction. It appears that lenders can get more access to borrowers' personal information and thus effectively monitor their loans. Borrowers can also establish trust with lenders and make a stronger effort to repay loans. The implications of our analysis are summarized as follows: Firstly, although earlier studies on personal loans have focused on personal bankruptcy decisions, this paper examines the effects of individual peculiarities on loan repayment. In particular, we find that bankrupt borrowers make greater efforts to repay the loans through online P2P lending. This is because if borrowers repay their loans in time, they can improve their overall credit ratings for loans from commercial banks in the future. Therefore, policy makers can consider online P2P lending as one of policy tools to encourage financial consumers with low credits for credit recovery. Secondly, we show that borrowers who raise money for essential living expenses make stronger efforts to repay their loans. Thirdly, we find that close interaction between potential borrowers and lenders is an important success factor in loan repayment. We conclude that such interaction can help reduce the information asymmetry, establish trust between borrowers and lenders, and raise the possibility of loan repayment. Our paper also has an important policy implication for consumer banking and small and medium enterprise finance. Domestic online P2P lending intermediaries work in partnership with savings banks and use their bank accounts as escrow accounts. However, due to the recent insolvencies of a number of large project finance operations, many savings banks are unable to play the roles of online P2P lending firms' escrow accounts. For example, ‘Pop Funding', one of the biggest online P2P lending intermediaries in Korea, had been in partnership with ‘Jeil Savings Bank'. But this bank was forced to shut down the operation because of its insolvency. Therefore, in order to firmly establish online P2P lending as a way to provide financing and credit enhancement for low credit borrowers, commercial banks should also play the role of providing escrow accounts for online P2P lending companies. In fact, European banks have recently been trying to enter the online P2P lending market more aggressively.