1. Describe the customer value proposition, profit model, core capabilities, and dynamic capabilities that underlay Prosper's business model. Note that Prosper's 'matching' service involves two sets of customers: borrowers and lenders. Address both sets of customers.

Communications of the Association for Information Systems Volume 29 Article 13 10-1-2011 Prosper-The eBay for Money in Lending 2.0 Hui Wang Department of Information Technology, Radford University, hwang26@radford.edu Martina E. Greiner Department of Information Systems and Quantitative Analysis, University of Nebraska at Omaha Follow this and additional works at: http://aisel.aisnet.org/cais Recommended Citation Wang, Hui and Greiner, Martina E. (2011) "Prosper-The eBay for Money in Lending 2.0," Communications of the Association for Information Systems: Vol. 29, Article 13. Available at: http://aiselaisnet.org/cais/vol29/iss1/13 This material is brought to you by the Journals at AIS Electronic Library (AlSeL). It has been accepted for inclusion in Communications of the Association for Information Systems by an authorized administrator of AIS Electronic Library (AlSeL). For more information, please contact elibrary@aisnet.org.borrowers, so that loan defaults do not lead to over-proportional losses. In July 2009, the minimum bid was lowered from $50 to $25 to increase diversification. Leveraging Social Capital to Reduce Default Rates In the age of Web 2.0, Prosper embraced the idea of leveraging a borrower's social connections (i.e., social capital) to help funding a loan, lowering interest rates on loans, and decreasing default rates. This led to the introduction of several social features, including a group system, endorsements, and friend networks Providing Borrower Information Borrower information is important for lenders to assess a borrower's credit risk. Prosper disclosed that between September 1, 2007, and August 31, 2008, self-reported information on borrower employment and income was verified only for approximately 23 percent of all borrower listings that had bids totaling 70 percent or more of the requested loan amount (3,486 out of 15,436 listings). Only approximately 56 percent (1,966 out of 3,486) of these borrowers undergoing verification during this period provided satisfactory responses and received a loan [Prosper Prospectus 2009.7.13, p. 42]. Verification of all borrowers' information might lower default rates; however, it would increase Prosper's operation cost considerably. Regulatory Challenges-SEC Regulation and the Quiet Period Prosper entered a highly regulated industry with an innovative business model. Being the first company with a new business model can create many challenges, and a legal strategy as part of the business model is required. Overcoming regulatory problems emerged as one of the most influential challenges Prosper faced in its first four years. Lending activities are typically state regulated and require lending licenses. On Prosper, lenders do not directly lend money to borrowers. Instead, lenders purchase "promissory notes" issued by Prosper as portions of a loan. If these Notes are securities, then they need to be registered with and regulated by the U.S. Securities and Exchange Commission (SEC). Before launch, Prosper worked with counselors on this issue to bypass this regulatory hurdle. Prosper's position was that the Notes issued by Prosper were not securities and therefore did not need to be Communications of the Association for Information Systems registered with the SEC. In October 2007, Prosper applied with SEC to open a secondary market that would allow lenders to trade Notes with each other." During the approval process, the marketplace had to enter a "quiet period" during which all new lending operations were suspended and only existing loans were served. Prosper's quiet period started on October 15, 2008. However, on November 24, 2008 Prosper received a cease and desist letter from the SEC. It was found that Prosper "from approximately January 2006 through October 14, 2008, violated Sections 5(a) and (c) of the Securities Act, which prohibit the offer or sale of securities without an effective registration statement or a valid exemption from registration." The Notes that Prosper issued have to be regarded as securities and thus need to be overseen by the SEC. SEC prohibited Prosper from issuing Notes until it registered with the SEC. Except for a short re-open and re-close in the state of California, Prosper couldn't continue its operation for nine months until it received SEC approval in July 2009. Prosper grew fast and substantially over its first three years, but being forced to halt for nine months after three years' of operation was traumatizing. Uncertainties surrounding the legal status of Notes led to confusion of borrowers and lenders and reduced trust. What is more, just before Prosper started its nine-month long quiet period, its major competitor, Lending Club, received SEC approval and emerged from its own six-month quiet period. Lending Club, launched in May 2007, applied for SEC approval within a year before its business grew too big. The timing of its return to the market was superb, as Prosper could not offer any new loans at that time, leaving it the only service provider for nine months. The Challenge Designing Social Networks The Prosper Group System Chris Larsen stated in 2006: "Credit markets have destroyed the sense of commitment and shame if you don't pay. ... So we try to make sure buyers are tightly associated with a group whose reputation is directly impacted by one person not paying. That should dramatically lower default costs" [Hof, 2006]. *The Prospectus can be found at http://www.sec.gow/Archives/edgar/data/1418285/DD0110485807078072/807-27421 1s1.htm. "The Cease and Desist Letter from the SEC can be found at: http:/www.sec.gov/litigation/admin/2008/33-8084.pdf. Volume 29 Article 13 251Prosper experimented with different tools to allow borrowers and lenders to connect with each other and to allow Prosper to connect with its customers. Prosper introduced a group system for its borrowers and lenders. Each group was led by a group leader, usually its founder. When a borrower joined a group, the leader had the option to review the listing, receive more details about the borrower from Prosper, or confirm borrower information by requesting more information directly from the borrower, a process called vetting. Lenders were also able to join groups to show their support for the group. Each group received a star rating based on the collective loan performance of the group members. The star rating would increase if the group beat a particular default rate and decrease if the group underperformed. The star rating allowed differentiation between good and bad groups. Prosper expected the group system to help by: . Lowering late payments and default rates as o Borrowers feel obligated to the group and repay loan in order to keep up the reputation of a group [Hogg, 1993]. O Group members discipline members who cheat, do not pay a loan, or harm the group in any other ways. o Group leaders filter potential borrowers, select high-quality borrowers, and confirm borrower information by requesting additional information from borrowers. Signaling credibility, increasing funding chances, and lowering interest rates as o Lenders prefer borrowers who belong to a group, because they expect group members to be less delinquent due to group pressure. More bids lead to increased funding chances (see Figure 4) and lower interest rates. o A bid from the group leader is expected to further signal credibility. Encouraging borrowers to access the collective knowledge of a group as Group members consult and help each other, for example, by reviewing the listing and giving tips on how to improve it (see Jan's story at the beginning of the case) . Attracting new borrowers and lenders as Group leaders receive a group leader reward (0.5 percent to 4 percent) on each successful loan and are motivated to bring in borrowers from outside and build strong groups. Strong groups with low default rates would bring in lenders. 100% 90% 80% 70% Communications of the Association for Information Systems 60% 50% 30% 20% 10% 0% 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 2006 2007 2008 of funded Listings with Group Membership of funded Listings without Group Membership Figure 4. Funding Rates for Listings with and without Group Membership by Month Problems with the Prosper Group System Unfortunately, Prosper discovered that the benefits of Grameen Bank's solidarity lending using off-line-often closely knitted-groups was difficult to transfer to an electronic marketplace First, only in few instances did the groups have the spirit that Chris Larsen envisioned. A Prosper lender writes in an open letter to Prosper: "First, while there may be 1 or 2 groups out there somewhere in the 3600 Prosper groups that actually have the dynamic [Chris Larsen, Prosper's CEO] envisioned, 99.9% of them do not. Most are groups of people who don't know each other at all. No commitment. No shame" [Fred93, 2007]. Many borrowers never spoke to their fellow group members but communicated only with the group leader. Without knowing each other, it was difficult for the borrowers to develop a sense of community and responsibility toward their group. One borrower Volume 29 257 Article 13commented, "I'm not connected all the time.... It was a onetime connection. Once I got the loan ... I made my payments [and did not stay connected]." Second, the group leader reward was created as compensation for vetting borrowers (performing the background check) and recruiting group members. Instead, some group leaders tried to maximize their personal financial gains by accepting and promoting borrowers regardless of their credibility. Some big groups (e.g., TwoMillionare debate) allowed up to sixty listings per day-filtering, selecting, and reviewing all borrowers is next to impossible in these cases. In addition, some groups used community payments (payments by the group in case a borrower does not pay) to keep their group's rating artificially high to keep attracting lenders. This could skew the group rating system. Lenders criticized these practices fiercely on Prosper's forum and on external forums and blogs such as www.prosper.org. The delinquency rates of loans associated with groups support this resentment (see Figure 5). Some group leaders also used the system for their personal gain. Lenders' portfolio plans might trigger automatic bidding if a listing is funded to a certain percentage. Group leaders would place multiple bids on their group members' listings to trigger lenders' portfolio plans, and they might even get outbid by subsequent bids. This practice soured some lenders. The hope that Prosper had for groups acquiring outside borrowers and lenders was not fulfilled. Many of the big groups were not bringing in new borrowers from outside but were recruiting borrowers who already came to Prosper. 50% 50% 40%% 30% 20% Communications of the Association for Information Systems 10% 0% 2 3 4 5 6 7 8 9 10 11 12 1 345 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 2006 2007 2008 " Delinquency Rate of Loans with Groupmembership "Delinquency Rate of Loans without Groupmembership Figure 5. Delinquency Rates for Loans by Group Membership* The Decline of Prosper's Group System Prosper implemented the group system from the start and tweaked it during the next one and a half years (e.g., introducing community payments in August 2006, adding group leader endorsements in September 2006). Groups experienced their greatest success from mid-2006 to early 2007, with July 2006 having 65 percent of new listings belonging to a group (see Figure 6). Groups became less and less attractive after the first quarter of September 2007. The decline of Prosper's group continued in September 2007 with the discontinuing of the group leader reward. Many group leaders did not continue with the groups because of the lack of incentive. The group system was not revived after Prosper's quiet period and had not played a major role as of February 2010. Instead, the endorsement and friend system grew significantly. A Prosper executive evaluates the success of the groups: "Early on we were really excited about groups. It felt like the right way to bring in the social bonding, and get better payback rates and things like that. I think due to some implementation details, it didn't work out quite as well as we had hoped. ... I'd say we are trying to figure out what's the next step with those. How to get them to function the way we want. It's very hard to pull in the right group leaders. We phased out the group leader rewards because we felt it was driving the wrong behavior. So all along it has been a challenge to find out what's going to drive the right behavior and make them work correctly. ... We are still working at it. It hasn't been an unqualified success, nor has it been a complete failure" (Prosper executive). Loans originated from 2/2008 to 10/2008; delinquency status as of Feb 2010; month shows the month the loan originated in. Volume 29 Article 13 25370% 60% 50% 40% 30% 20% 10% 0% 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 2006 2007 2008 -% Listings with Group membership of Total Listings Figure 6. Percentage of Listings with Group Memberships Endorsement and Friends Endorsements were introduced in September 2006 to allow group leaders to endorse borrowers and write recommendations. In February 2007, this feature was expanded to allow other Prosper members to endorse borrowers. Furthermore, borrowers are able to add other Prosper members as "friends" to show that they have a connection within the Prosper community. Prosper also displays whether endorsement givers or friends have placed a bid on the related listing. Both endorsements and friends are expected to help lenders identify high quality listings as well as give borrowers higher social pressure and incentive to pay back their loans. Endorsements and friends are shown to have positive influences on funding chances and loan payment, although the results are mixed [Greiner and Wang, 2009; Lin et al., 2009]. Prosper Forums, Blog, and Websites Borrowers and lenders had the opportunity to communicate with each other in forums organized by Prosper. These Communications of the Association for Information Systems forums included a special corner for lenders, borrowers, and group leaders where topics interesting to groups were discussed. For example, in the borrower forums, new and potential borrowers could ask for help and information about how to write a listing, introduce themselves to lenders, and promote their own listings. These thriving forums were an important source of information for Prosper members. However, the forums were also a platform for criticism of individual borrowers, lenders, groups, or Prosper. Some topics were heavily discussed and both positive and negative opinions were voiced. In March 2007, Prosper locked several threads (many of them critical of a large group) and banned from the forums users who were identified as being in violation of forum policies. In November 2007 the forums were deleted and replaced with a forum where contributions were moderated by Prosper employees." As a reaction, many former Prosper forum participants switched to third-party forums (e.g., www.prosper.org) and blogs in order to continue non-moderated discussion about Prosper. The official Prosper forum was discontinued in December 2009 due to lack of activity. Prosper has used other venues for social networking. Prosper started a blog (http://blog.prosper.com/) in December 2007 that gives advice to borrowers and lenders, reports on important changes at Prosper, and discusses important topics in P2P lending. In March 2010, Prosper started a separate website called Talk Taboo where people can share their financial stories and learn about debt consolidation. So far this website is targeted toward borrowers and encourages P2P lending A Little More Safety, a Little Less Freedom Initially, Prosper started with a wide open approach "being a pure web 2.0 company" (Prosper executive). Later on, the model shifted to include "a little more safety, a little less freedom, than where Web 2.0 started in [20104/[20105. ... You need as much freedom as you can get, but you also need to have control" (Prosper executive). The shift in strategy required changes and support on the technology side. Most of the information systems were developed in- house, and only few third-party software packages were used. The system introduced at launch in February 2006 "An archive of the official forum before November 2007 can be viewed at http://www.prosperreport.com/- Volume 29 254 Article 13has been tweaked and changed repeatedly. New features were added and existing features were improved or changed as a result of feedback from users and requirements from the business environment. Many changes were the results of abuse of the system, and unintended consequences. For example, initially, members were allowed to add as many bank accounts as they wanted. This was thought a nice service for lenders and borrowers, to be able to pay or receive money from different sources. However, some members were using Prosper as a cheap way to move money between accounts. On realizing this system abuse, Prosper limited the default number of bank accounts to one primary bank account. Some members were abusing the e-mail system and sending unsolicited e- mails that were not Prosper related to other users. Prosper's system had to change to add more control to stop this from happening. In a marketplace where participants remained largely anonymous, fraud detection was a technological challenge as well. To facilitate fraud investigation, IP addresses were recorded, and the Prosper system was integrated with a third-party system to do external checks on members. VII. THE COMPETITIVE ENVIRONMENT AND THE FUTURE OF PROSPER Prosper entered an established financial industry with its idea of circumventing the middleman-an idea that could threaten established players if the idea proved successful. Competitive forces for Prosper come from two different directions: traditional financial institutions such as banks and credit companies, and other existing and future P2P lending companies Competition from Traditional Banks and Credit Companies Prosper sees credit companies as their main current competitor. "The competitors have changed so radically with the financial collapse, and it hasn't returned anything close to normal. But still, we would say, the traditional credit card companies are really our biggest competitor. This is really the way small business lending has happened in main street America" (Prosper executive). Prosper differentiates itself from traditional financial institutions by not being the grantor of credit but rather providing the infrastructure to individuals who will then price and grant credit. A Prosper executive draws the analogy between a retailer and eBay: "The way we work is that we just give the tool to people who can decide on their own who is going to get the credit at what price. So that's what we define as infrastructure; there is no central grantor of a price Communications of the Association for Information Systems or credit. It's a broad distributed market that uses our infrastructure. ... Think about eBay. They don't decide how much their items will be sold for, they just provide the tools to allow them to be sold." Banks operate on a spread base. They pool money from individuals (lenders) who deposit money for a certain percentage rate and then grant credit to other individuals (borrowers) for a higher percentage rate. Banks earn money based on this spread. A Prosper executive expects P2P lending to fundamentally change the financial industry because it gives lenders and borrowers the power to eliminate banks as the middleman: "with P2P lending, [individuals] can be the banker themselves. So, spread modeling banking should be dead soon." Competition from Other P2P Lending Companies In the U.S., Prosper and Lending Club are the major players in the for-profit P2P lending area. Whereas Prosper deploys an auction-style lending process and lets the lenders decide the price of the credit, Lending Club sets a fixed rate on loans. Prosper CEO identifies how Prosper differentiates itself from Lending Club: "They have taken a much more dictatoral view. We put them in the category of a credit company. Their management are the ones who decide who gets credit at what price, rather than let the community do that. We don't think that's the right way to go. ... So, we are unique in our approach." In addition to the auction-based style, Prosper claims that capturing and monetizing the social capital of borrowers are the major differences from Lending Club. A Prosper executive comments: "[Prosper's] model uniquely has the ability to capture and more important monetize social [element]. If you look at a bank, or even Lending Club, [they] are dictating credit and price. There is no way to value that social [element]. Prosper is the first model where you can see the social, which could be community, friends, family, ... that can now be valued by individuals who control the granting of credits and the price of the credit. So that's really the breakout here. Looking Forward: The Future of P2P Lending and Prosper "I think [P2P lending] will have its place next to lending from banks because people are always anxious for new avenues for credit. Ideas like the open market that we have could be really interesting as far as credit moves in the U.S. or even around the world and level of involvement of individual investors. And on the borrowing side ... that Oh, I need to borrow money, well, one of my options is P2P lending' will expand. How rapidly is the big question" (Prosper executive). Volume 29 Article 13 255Prosper is still the largest P2P lending marketplace in the U.S. with almost $191 million of loans generated in its first four years. As of March 2010, Lending Club almost reached the $100 million threshold. Prosper's advantage was its early entry into the market in February 2006, one and half years before Lending Club. Lending activities prospered, and the majority of the loans were made in its first three years of existence. Lending Club launched in May 2007 and closed for SEC registration from April 2008 to October 2008. After Lending Club emerged from its quiet period, Prosper was closed down for nine months for SEC registration. During this time, Prosper was not able to create new loans. Lending Club grew and surpassed Prosper in terms of monthly loan origination amount (see Figure 7) and reduced the gap in overall loan origination (see Figure 8). As of February 2010, Prosper has not been able to repeat its success before the quiet period. Loan originations by month are still a fraction of what they were before the quiet period and considerably lower than the loan originations of Lending Club. Currently, it is unclear whether Prosper can recover from the quiet period and which of the two P2P lending marketplaces will play the bigger role. Another round of competition for dominance in the United States' P2P lending market has begun. Lending 12 -Club - -ProsperQulet - Quiet Period 10 Period Millions Dollars A 2 4 6 8 10 12 2 4 6 8 10 12 2 4 6 8 10 12 2 4 6 8 10 12 2 4 2006 2007 2008 2009 2010 Prosper Lending Club Figure 7. Loan Origination Prosper and Lending Club by Month and Dollar Amount Communications of the Association for Information Systems 250 Lending Club Prosper Quiet 200 Quiet- Period Millions Dollars Period 150 100 2 4 6 2 10 12 2 4 6 2 10 12 2 4 6 8 10 12 2 4 6 8 10 12 2 4 2006 2007 2008 2009 2010 Prosper Lending dub Figure 8. Cumulative Loan Amounts by Month and Dollar Amount Volume 29 256 Article 13REFERENCES Editor's Note: The following reference list contains hyperlinks to World Wide Web pages. Readers who have the ability to access the Web directly from their word processor or are reading the article on the Web, can gain direct access to these linked references. Readers are warned, however, that: . These links existed as of the date of publication but are not guaranteed to be working thereafter. 2. The contents of Web pages may change over time. Where version information is provided in the References, different versions may not contain the information or the conclusions referenced. 3. The author(s) of the Web pages, not AIS, is (are) responsible for the accuracy of their content. 4. The author(s) of this article, not AIS, is (are) responsible for the accuracy of the URL and version information. Anonymous (2006) "America's First People-to-People Lending Marketplace: Prosper", paymentsnews.com, http://www.paymentsnews.com/2006/02/americas first .html (current Mar. 20, 2011). Anonymous (2008) "Lessons from the P2P Community", Point for Credit Union Research & Advice (11)11. Fred93 (2007) "Open Letter #1 to Prosper.com: Don't Mislead People", prospers.org, http://fred93.com/fbi/open- etter-number-1.pdf (current May 9, 2011). Greiner, M.E. and H. Wang (2009) "The Role of Social Capital in People-to-People Lending Marketplaces", ICIS 2009 Proceedings, Paper 29. Hof R. (2006) "Prosper: The eBay of Loans?" Business Week, http://www.businessweek.com/technology/content/feb2006/tc20060213 147523.htm (current Mar, 20, 2011). Hogg, M.A. (1993) "Group Cohesiveness: A Critical Review and Some New Directions", European Review of Social Psychology (4), pp. 85-111. Kim, J.J. (2008) "Where Either a Borrower Or a Lender Can Be", The Wall Street Journal, http://online.wsj.com/article/SB 120526439925827991.html (current Mar. 20, 2011). Communications of the Association for Information Systems Lin, M., N.R. Prabhala, and S. Viswanathan (2009) "Can Social Networks Help Mitigate Information Asymmetry in Online Markets?" ICIS 2009 Proceedings, Paper 202. Tedeschi, B. (2006) "It's Like Lending to a Friend, Except You'll Get Interest", The New York Times, http://www.nytimes.com/2006/02/13/technology/13ecom.html (current Mar. 20, 2011). Wang, H., M. Greiner, and J. Aronson (2009) "People-to-People Lending: The Emerging E-Commerce Transformation of a Financial Market" in Nelson, M., M. Shaw, and T. Strader (eds.) Value Creation in E- Business Management, Berlin Heidelberg, Germany: Springer, pp. 182-195. ABOUT THE AUTHORS Hui Wang is an assistant professor in the Department of Information Technology at Radford University. She holds a Ph.D. in Management Information Systems from The University of Georgia. She also holds Masters' degrees in MIS and Economics from the University of Nebraska at Omaha, and a Bachelor's degree in Economics from Nankai University, China. Her current research interests include C2C e-commerce, social networks as a business platform, and knowledge management systems. She has published her research in the International Journal of Electronic Commerce, Value Creation in E-Business Management, International Journal of Behavioural and Healthcare Research, and IS conferences such as the International Conference on Information Systems (ICIS). Martina Greiner is an assistant professor at the Department of Information Systems and Quantitative Analysis at he University of Nebraska at Omaha. She received her Ph.D. in Management Information Systems from the University of Georgia and holds a Master's of Business Administration and Economics from the University of Hohenheim, Germany. Her research interests include e-commerce and digital communities, trust in the online environment, and applying organizational theories to IS issues. She has published her research in IS journals such as the International Journal of Electronic Commerce, Communications of the ACM, and Communications of the Association for Information Systems, and presented her research at MIS conferences such as the International Conference on Information Systems (ICIS). Volume 29 Article 13 257Copyright @ 2011 by the Association for Information Systems. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and full citation on the first page. Copyright for components of this work owned by others than the Association for Information Systems must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists requires prior specific permission and/or fee. Request permission to publish from: AIS Administrative Office, P.O. Box 2712 Atlanta, GA, 30301-2712, Attn: Reprints; or via e-mail from ais@aisnet.org. Communications of the Association for Information Systems Volume 29 258 Article 13HIS A ommunications of the ssociation for Information S, ystems ISSN: 1529-3181 EDITOR-IN-CHIEF lize Zigurs University of Nebraska at Omaha AIS SENIOR EDITORIAL BOARD Guy Fitzgerald lize Zigurs Kalle Lyytinen Vice President Publications Editor, CAIS Editor, JA/S Brunel University University of Nebraska at Omaha Case Western Reserve University Edward A. Stohr Blake Ives Paul Gray Editor-at-Large Editor, Electronic Publications Founding Editor, CAIS Stevens Institute of Technology University of Houston Claremont Graduate University CAIS ADVISORY BOARD Gordon Davis Ken Kraemer M. Lynne Markus Richard Mason University of Minnesota University of California at Irvine Bentley University Southern Methodist University Jay Nunamaker Henk So Ralph Sprague Hugh J. Watson University of Arizona University of Groningen University of Hawaii University of Georgia CAIS SENIOR EDITORS Steve Alter Jane Fedorowicz Jerry Luftman University of San Francisco Bentley University Stevens Institute of Technology CAIS EDITORIAL BOARD Communications of the Association for Information Systems Monica Adya Michel Avital Dinesh Batra Indranil Bose Marquette University University of Amsterdam Florida International University of Hong Kong University Thomas Case Evan Duggar Mary Granger Ake Gronlund Georgia Southern University of the West George Washington University of Umea University Indies University Douglas Havelka K.D. Josh Michel Kalika Karlheinz Kautz Miami University Washington State University of Paris Copenhagen Business University Dauphine School Julie Kendall Nancy Lankton Claudia Loebbecke Paul Benjamin Lowry Rutgers University Marshall University University of Cologne Brigham Young University Sal March Don McCubbrey Fred Niederman Shan Ling Pan Vanderbilt University University of Denver St. Louis University National University of Singapore Katia Passerini Jan Recker Jackie Rees Raj Sharman New Jersey Institute of Queensland University of Purdue University State University of New Technology Technology York at Buffalo Mikko Siponen Thompson Teo Chelley Vician Padmal Vitharana University of Oulu National University of University of St. Thomas Syracuse University Singapore Rolf Wigand Fons Wijnhoven Vance Wilson Yajiong Xue University of Arkansas, University of Twente Worcester Polytechnic East Carolina University Little Rock Institute DEPARTMENTS Information Systems and Healthcare Information Technology and Systems Papers in French Editor: Vance Wilson Editors: Sal March and Dinesh Batra Editor: Michel Kalika ADMINISTRATIVE PERSONNEL James P. Tinsley ipin Arora Sheri Hronek Copyediting by AIS Executive Director CAIS Managing Editor CAIS Publications Editor S4Carlisle Publishing University of Nebraska at Omaha Hronek Associates, Inc. Services Volume 29 Article 13Communications of the Association for Information Systems CAIST Prosper-The eBay for Money in Lending 2.0 Hui Wang Department of Information Technology, Radford University hwang26@radford.edu Martina E. Greiner Department of Information Systems and Quantitative Analysis, University of Nebraska at Omaha Communications of the Association for Information Systems Abstract: When a bank or a credit union turns you down for a loan because your debt-to-income ratio is too high, can you turn to total strangers to get the money? Yes, you can, and we are not talking about criminal acts. It's called peer-to-peer (P2P) lending or social lending. Prosper was the first company to take the vision of social lending and convert it into practice on the Internet in the United States, and it enjoyed fast growth after launch. Four years later, however, it is facing old and new challenges, and its survival is on the line. This case depicts the opportunities and pressures Prosper faced, its actions and reactions, and its future. Prosper has made and will make many important decisions, and Prosper's successes and challenges are rich material to study. Keywords: P2P lending. www.prosper.com, financial business model, financial industry Editor's Note: A teaching note for this case can be obtained from hwang26@radford.edu. Only active MIS faculty who are currently listed in the AIS Faculty Directory are eligible to receive the teaching note. Volume 29, Article 13, pp. 243-258, October 2011 The manuscript was received 8/19/2010 and was with the authors 4 months for 1 revision. Volume 29 Article 13Prosper-The eBay for Money in Lending 2.0 I. INTRODUCTION: A BORROWER STORY Like many others, Lara Miller is a talented young woman with a passion for fashion. She had a plan to turn that passion into a successful business. Her talent was recognized by the press and seventeen retail outlets, but Lara's business plan faced an obstacle. Without a credit history or sufficient collateral, traditional financing through a bank was difficult. The young designer and her small start-up faced a difficult challenge. "Fortunately, a friend told me about Prosper," Lara says. Within minutes of visiting Prosper's peer-to-peer lending site, Lara posted her loan request and, "[it was amazing. I posted the loan, and almost immediately, I had 8 bids" (from stories on Prosper.com). Total strangers bid on Lara's loan listing, and Lara's business was fueled to move forward. Over time, Lara took out three loans with Prosper. As of this writing, Lara's design career and her store are prospering. Her story is featured on Prosper's website, and Lara has a big smile on her face. Jan (not the real name), another Prosper borrower, was introduced to a Prosper borrower group by her daughter. The group leader gave Jan the advice she needed for her first loan request. She kept the listing simple and honest, telling people who she was and what she needed the money for, and she requested $5000 to improve her financial situation. She was contacted by a few potential lenders, who inquired about how she got into the financial difficulty she was in and whether she was able to repay the money. As Jan recollected, the whole process was very "personal" and she felt "connected," very different from her experience with a financial institution, where "all they look at is my credit score, how much I owe, and how much they are making. It's very cold that way, whereas [on] Prosper... [I tell people] this is what I do, I am a single professional woman, and I have some grandkids, I want to have a nice yard for them to play in." As Jan watched on Prosper's website, strangers started to "chip in" on her listing, and then her loan was fully funded! With the money she got on Prosper, she paid off her high interest charge cards, and the monthly payment she was carrying was lowered from over $500 to less than $300. With Prosper, Jan said "[It's] a lot more personal, rather than cold, institutional. ... It's very easy. I feel more connected, these people are willing to lend a hand to me, taking a chance on people, very philanthropic, I like that idea.... I felt more committed, people had taken a chance on me, loaned me money, in order for me to do this." Jan currently has two Prosper loans. She has paid off her high interest debts and put up a basketball court in her yard for her grandchildren. She is on her way to pay back both loans, and she has less than $1000 left on her first loan. Communications of the Association for Information Systems "Prosper helped me to reach my dreams," says Jan. This happy ending might lead to many more happy endings; Jan said, "I would like to be in a position eventually to be a lender, because I know how valuable that has been to me. ... I like the idea of philanthropy-somebody reaching out to help somebody else. ... [A lender] might have been in that [similar] position, or a relative in that position, as long as they feel secure that this person is not trying to get away with something, [they would like to help.]" That might be Prosper's dream-for people to reach out to help others and to live their dreams and prosper! II. P2P LENDING: A MARKETPLACE FOR LENDERS AND BORROWERS Peer-to-peer lending, also referred to as people-to-people lending, person-to-person lending, or social lending, is your lending and borrowing 2.0. Instead of applying for a loan from a bank or taking a cash advance from a credit card company at their rate, on a P2P lending website you can post a loan listing with the interest rate you are comfortable with, watch strangers declare the amount of money they'd like to fund your loan, and even watch the interest rate bid down. When the loan is fully funded, the money gets transferred to your bank account, and you start paying the loan back at the interest rate determined by the bids. P2P lending is not new. The practice of lending money to family members or within communities dates back to ancient times, long before the rise of any financial institutions. The emergence of the Internet, however, has revived the concept; it allows P2P lending to go beyond the circle of friends, family, and community to reach a much larger scale over the largest network in the world. Individuals register to become a lender, a borrower, or both. As eBay and Amazon have fundamentally changed the retail industry by connecting buyers and sellers directly, P2P lending marketplaces create an online platform to connect borrowers on the demand side and lenders on the supply side. Traditional loans involving a financial institution, such as a bank or a credit union, may be referred to as institution-to-people lending. In this model, financial institutions pool the money on the supply side (e.g., savings accounts and other investment instruments) and lend it to a pool of borrowers on the demand side, and earn a profit by charging fees and interest. In this traditional model, lenders do not have much say in the interest rate, nor do they Volume 29 244 Article 13have much control over who the money goes to or for what purpose. For instance, there is no way to know how a bank uses the money deposited in a money market account or a certificate of deposit. Borrowers, on the other hand, have little if any influence on the interest rate as evaluation instruments from traditional institutions often evaluate an individual using few hard credit criteria (e.g., credit score, existing debts, income, job) to evaluate credit risk. Soft information (such as relationships, emotional appeal, social connections) are rarely used by traditional financial institutions. Also, borrowers with an unfavorable credit history often have a hard time getting a good interest rate or even a loan at all from banks, and have to turn to payday loans or high interest credit cards. With P2P lending, the traditional financial services middlemen are bypassed. A lender can choose the borrowers to finance directly, and a borrower can affect lenders by leveraging his/her social capital.' These social features are attractive to both borrowers and lenders, and many of them have been well received. Zopa, the oldest online P2P lending marketplace, was launched in the UK in December 2005, and Prosper, the oldest online P2P lending marketplace in the U.S., in February 2006. Awareness of P2P lending has continued to grow, and P2P lending marketplaces now exist in many different countries, with a few variations in form. P2P lending marketplaces differ in several ways. Some focus on certain interest groups and the mechanism for interest rate determination; others focus on student loans, loans to disabled people, or business loans for developing countries. The two primary mechanisms for determining interest rates are auction style and credit-based fixed interest rates. P2P lending marketplaces also differ by lenders' motivation to lend and the degree of separation between lenders and borrowers [Wang et al., 2009]. A lender's motivation may range from profit to pure philanthropy. Some lenders find P2P lending attractive because they can support the people in need while making some money. The degree of separation between lenders and borrowers ranges from close relationship, such as family, to stranger. Wang et al. [2009] provide an overview of P2P lending in the U.S., considering the motive for lending and degree of separation among participants (see Figure 1). Stranger Lending Club $93.8M Kiva Communications of the Association for Information Systems $121.4M Prosper The $190.4M Philanthropic Model Lend4Health The $ 1M Profit-Seeking Eynanz. Model Degree of Separation $<. zimplemoney greennote sunknown s im lendingkarma the tupperware family friend virgin money kinship party model economic philanthropic motive of lending figure p2p marketplaces in u.s. social capital refers to value an individual connections. a context this could include group affiliations endorsements by other people and emotional appeal through listing description. size oval representing each company reflects its cumulative loan volume. which cannot be compared directly as time spans for these companies differ due differences their launch dates. source information with timeframe parentheses: prosper club kiva discontinued lend4health volume unknown fynanz from http: article prosper: america first online marketplace chris larsen vision who co-founded e-loan inc. had early using internet distributed platform access. idea was put control hands individuals allow them play much more active roles whole process borrowing rather than accepting terms dictated banks. always need banking but you don banks said larsen. whereas credit unions dictate e.g. gets loans at what price tools everyone ordinary americans supplement own bank. result is positive experience on both borrower side lender side. explains advantages borrowers: such radical change way interact have fundamental benefit controlling maximum rate want pay. passively accept you. can try it your own. say look i good story business going if does get funded ok not another one. here percent whatever is. sides. lenders experiences are equally rewarding. besides making return they invest also feel about reasons example help down street. beside direct participation anticipated that would financially borrowers. gap between interest rates paid charged massive. cds ranged while cards were easily or higher. payday charge higher upwards percent. comparison average cd listed table borrowers because leaner operation traditional financial institutions there no physical branches related infrastructure service receive pay fee equal specified percentage principal. grade year term aa b communications association systems c d e hr nc source: data category score official after two-month testing period launched february headquarters san francisco. brought his reality has been ceo since then. one week attracted total lend attention national media. month featured news stories top media including new york times two sections wall street journal economist march wcbs tv yahoo finance cbs news. three years members generated million loans. growths displayed posted requests received bids per realized founded provide customers access mortgage over broker. listings growth how works only registered borrow prosper. although member identity remains anonymous public authenticates registration checking security number. bank account number home address driver license state identification card required registration. obtains potential third bureau posts grades historical listing. makes specifies amount requested willing up subject minimum based rating must annual income occupation employment status. option write description typically states purpose situation. determines categorical label called corresponds risk proprietary evaluate quickly level choose bid attractive lend. even compete supplied outruns amount. fully before ends transferred account. all fixed unsecured amortizing simple rate. person institution. maintain funds sufficient make fdic-insured non-interest bearing wells fargo payments payable deposited funding legal regulations made webbank fdic- insured utah-chartered industrial lender. issues sells assigns promissory notes evidenceng lenders. ways-bidding manually bidding automatically setting portfolio purchasing existing held less current auction yield usually portion reduce risk. should totaling full won obligated purchase principal bid. second set plan fit pre- defined criteria. under ratings estimated loss many plans desired place any them. offers ready-made different targets used modified. save searching management diversification. sell hold buy trading platform. increases liquidity do until mature investment appealing. en route big changes innovative peers implemented technology brings eminent benefits. holds promises allowing easy offering alternative satisfying human embededness networking. able paperwork lock lower rates-often percent-than otherwise allows reach out millions dispersed country give self-service cutting intermediaries system lowers cost benefiting consumers obtain conditions relationships gives chance adopt very hands-on micro-management strategy. small each. strategy gain where goes critical feature separating markets. whether whom actively doing so. collected lent average. invested thirty-three aspect alike. seeing hand cause so dear encouraging personal. moreover getting e-mail praising paying back something warm missing interactions unions. someone know decision that. saying purely think typical unlike when we talk mostly benefit. motivations well financial. carries those times. just nature work right spend most working same will stop somebody help. thing here. executive v. built like e-commerce businesses technology. capabilities differentiate itself institutional lending. provides market maker community builder. function secure efficient trade. technologies create transaction informing support. series transactions authentication verification reporting processing transfer settlement systems. addition providing efficiency routinize processes increase standardization improve cultivate trust long-term success marketplace. uses support instance automate rich histories see delinquency history last seven open lines range status length employment. database advanced search sift thousands profiles find meet preferences. preserve watch list forecast chart predicts likely promising listings. december released mobile version website exclusively ubiquity ask questions smart phone. builder beginning embraced leveraging connections participants. believe groups heart fostering culture responsible quoted paymentsnews.com simple: connect bonds form encourage behavior developing countries microcredit programs utilize shown grameen bangladesh poor lives. solidarity approach relies tightly knitted repay loan. connection wants leverage well. join established beginning. additional features added tweak system. forums networks blogs offer multiple venues other. some successful others learned lessons upsides downsides design effectiveness within years. vi. challenges reactions challenges. prominent faces default regulatory requirements capital. adjust policies strategies respond. challenge lending-default problem complete strangers back. offered words lose money. may part pays amortized payment period. means take long recover nominal opportunity excluded. takes monthly respectively delinquent debt buyer experiencing expected estimate order select retum-on-investment initially experian calculate roi. however soon indicator f>539 769 639 599 Credit Score Loans made from 2/13/2006 to 2/12/2007 661 609 792 1119 1222 1456 1633 141 Lender Rate (Annualized; before service fees) 9.1% 11.0% 13.7% 16.5% 19.8% 23.5% 23.6% 21.6% Experian: Historical Average Default Rate * * 0.2% 0.9% 1.8% 3.3% 6.2% 9.1% 13.9% N/A Experian: Range (Default rate) 0.0% 0.7%- 1.6% 2.9% 5.4% 9.1%- 15.1%- 0.4% 1.1% 2.1% 3.7% 7.2% 11.8% 28.2% N/A Prosper: Default Rat * * 9.7% 18.6% 24.5% 32.7% 36.3% 48.9% 64.4% 69.5% Prosper: Default Rate (Annualized) 3.1% 5.8% 7.6% 9.9% 10.9% 14.2% 18.0% 19.2% * In effect from 2/13/2006 to 2/12/2007; Credit Scores are based on Experian Scorex PLUSSM. *Experian historical average default rates by credit grade for borrowers with less than 20 percent debt-to-income ratio (Source: Prosper cited in http://www.bankrate.com/brmews/investing/20061006a3.asp.) *** Default rates as of Feb 2010 for loans made between 2/13/2006 and 2/12/2007 " Grameen bank: http://www.grameen-info.org/ Volume 29 Article 13 24930% 70% 60% 50% 40% 30% 20% 10% 0% 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 Loan Length in Months AA -A B C -D -E. E +HR -NC Figure 3. Delinquency Rates by Credit Grade and Loan Lengths Controlling Default Rates "Ultimately," said a Prosper executive, "part of the challenge of our business is that certain borrowers are going to default. This is just a fact of life. They get into trouble, whatever it is. It doesn't mean they intended to do this from the get-go, but it is just a fact of life. And so what we want to do is help our lenders price for that risk and price the risk in correctly" (Prosper executive). High default rates are bad for lenders, borrowers, and Prosper alike because they decrease lender's ROI. Prosper's selling point to lenders is that P2P loans are an alternative investment Communications of the Association for Information Systems instrument. So, the main challenge for lenders is to select listings with a high expectancy of repayment and for Prosper to help lenders to do so. This leads to several policy and strategy changes to decrease default rates and help lenders assess risks better. Changes in Risk Ratings and Minimum Credit Score Requirements for Borrowers At launch, Prosper assigned a credit grade to listings to help lenders evaluate the credit risk based on the past performance of the borrower. This credit grade was calculated based on the credit score obtained from Experian's credit report. The credit categories and corresponding credit score range are shown in Table 2. In February 2007, Prosper changed its policy and excluded borrowers in the NC (No Credit) category and allowed only borrowers with a credit score of 520 or higher to apply for a loan. In July 2009, Prosper raised he minimum credit score further from 520 to 640. In addition, Pro