P2P-Banking.com

The study “What’s in a Picture? Evidence of Discrimination from Prosper.com” by the economic professors Devin Pope and Justin Sydnor finds:

We analyze discrimination in a new type of credit market known as peer-to-peer lending. Specifically, we examine how lenders in this online market respond to signals of characteristics such as race, age, and gender that are conveyed via pictures and text. We find evidence of significant racial disparities; loan listings with blacks in the attached picture are 25 to 35 percent less likely to receive funding than those of whites with similar credit profiles.  conditional on receiving a loan, the interest rate paid by blacks is 60 to 80 basis points higher than that paid by comparable whites. Though less significant than the effects for race, we find
that the market also discriminates somewhat against the elderly and the overweight, but in favor of women and those that signal military involvement.
Despite the higher average interest rates charged to blacks, lenders making such loans earn a lower net return compared to loans made to whites with similar credit profiles because blacks have higher relative default rates. This pattern of net returns is inconsistent with theories of accurate statistical discrimination (equal net returns) or costly taste-based preferences against
loaning money to black borrowers (higher net returns for blacks). It is instead consistent with partial tastebased preferences by lenders in favor of blacks over whites or with systematic underestimation by lenders of relative default rates between blacks and whites.

Their conclusion:

Yet the data tell a very different story that suggests that this peer-to-peer lending market actually treats the races more equally than would be expected in a market with accurate statistical discrimination.

I would interpret this conclusion as a negation of p2p lending leading to racial discrimination. However Ron Shevlin at MarketingROI comes to different conclusions.

Short news update on Prosper.com developments:

• Changes at the Prosper management team: John Witchel (former CTO) and Tom Pigoski are no longer listed as part of the management team. A comment on this blog post indicates that John Witchel has left the company while Tom Pigoski is on family leave. Chris Denend is now CTO.
• Prosper says it will enable faster listings for qualified borrowers. In a streamlined process some borrowers may post loan listings without adding a title, description or picture.
• Reading statistics published by Prosper? Have a good look on the definitions! In the market survey results, published on Aug. 12th, it looks like on some parameters the year to date figures for 2008 are rising compared to 2007. However looking in the definitions, Prosper compares 7 months in 2008 to 6 months in 2007. (quote:
  2008 Year-to-Date: January 1, 2008 through July 31, 2008.
  2007 Year-to-Date: January 1, 2007 through June 30, 2007.)

The Prosper study I featured yesterday lists the Prosper fees. While it does not attempt to calculate the revenues of Prosper.com it gives some indications regarding the operating costs:

According to Mendelson (2006), the primary costs of Prosper consist of …, (2) a $4 fee for identity authentication, credit pulling, and bank-account setup per active borrower, (3) customer service at the average rate of four interactions per loan and $2 per interaction, and (4) a fixed overhead cost of approximately $3 million per year. …

Given these estimates, it is difficult to measure Prosper accounting in precision. However, there is no doubt that Prosper’s revenue does not cover its full cost (as of February 2008). The difference is met by a large stock of venture capital.

Regarding the Prosper revenues, Mike did a calculation estimating the November 2007 revenues at 114,000 US$. If this is correct, it did not even cover fixed overhead.

Prosper changed fees since that last calculation.

The new study "Dynamic Learning and Selection: The Early Years of Prosper.com" by Seth Freedman and Ginger Zhe Jin, both at the Department of Economics, University of Maryland analyses Prosper data in a time frame from April 19th 2006 to December 31st 2007.

The study analyses the development of the Prosper.com marketplace and how lenders refined their strategies as a result to own experiences and changed settings.

They write:

Overall, we conclude that Prosper is evolving from a comprehensive market to a market that primarily serves the borrowers who have access to traditional credit. This implies that Prosper will compete head-to-head with the traditional banks rather than pick up a missing market. Assuming away any cost in information processing, we estimate that the average rate of return of a Prosper loan is 6% if Prosper loans continue to perform according to what we have predicted from their existing performance. From the lenders point of view, this number compares favorably to 6-month certificate of deposit and 3-year Treasury bill, but less favorably to the rate of return implied by the S&P 500 in the same time period.

Other findings are that high interest loans yield lower returns due to high default rates and that the probability for defaults of Prosper loans peak at month 10 and the edge down.

The main uses of Prosper loans are:

33% of all previous Prosper listings have mentioned credit card consolidation, which is higher than the mention of business (23%), mortgage (15%), education (22%), and family purposes (20%) such as weddings.  

Cited from the conclusion chapter of the study:

The first two years of Prosper has enlivened the concept of P2P lending, but the road towards success is full of challenge. While it is tempting to expect P2P lending to alleviate credit rationing for near- or sub-prime risks, we find Prosper evolving from a comprehensive market toward a market that primarily serves borrowers who have access to traditional credit. This implies that Prosper will compete head-to-head with the traditional banks, rather than pick up a missing market. This pattern is not unique to Prosper. …

How can Prosper compete with traditional banks? Our study suggests that the microfinance approach, as implemented through Prosper groups, has failed to select good risks or enhance loan performance. But on the up side, lenders are learning fast about the pitfalls of P2P lending thanks to the transparency of Prosper. Our calculation suggests that, if the loans continue to perform as what we have predicted from the market performance, Prosper loans could yield an average return of 6%.

See related post on the Prosper blog.

A recent Bankrate.com article gives an update on the development of p2p lending in the US.

Chris Larsen of Prosper.com sees the current financial situation as a chance for p2p lenders:

Home equity used to be the cash management tool for the credit-worthy borrower, and that has really, really dried up. In many ways, Prosper's three-year, 25,000 US$ loan is a pretty good proxy for what people were using home equity for — improving their home, starting a sole proprietorship, college costs and certainly for replacing credit card debt.

Javelin Strategy & Research is quoted that credit card debt is the main reason people want to use p2p lending:

We're forecasting that P2P lending specifically for credit card balances will grow from 38 billion US$ in 2007 to 159 billion US$ by 2012

The final advice of the article is:

Prospective borrowers and lenders would do well to thoroughly research P2P companies before jumping at the chance for a lower rate on a loan or a higher return on an investment. …

Prosper.com announced yesterday that the offers received for late loans to be sold off as defaults were to low to be acceotable. Doug Fuller, Vice President of Operations says, that with offers at 1.5 cents for the dollar, Prosper believes "the prudent course of business is NOT to sell them at this time". More on the Prosper blog post.

Prosper will run television ads (preview them online). The TV spots will be aired in test markets starting this week.

Zopa plans to launch a student loan feature in July. Students can already use peer-to-peer lending sites to apply for a personal loan. At Prosper.com, education loans account for less then 2% of its 135 million US$ in loans. (Sources: Smartmoney,  WSJ)

Prosper.com made the videos of the Prosper Days 2008 available. Thanks Prosper!

I will just feature one session here. In "Managing large portfolios" Adam Weyeneth, President of Fair Deal Credit talks about the issues for institutional lenders that want to invest into the asset class of p2p loans.

Link to this video and the related videos of the other sessions.

Announced today Prosper.com has achieved nationwide lending (exceptions South Dakota and Texas) with an interest rate maximum of 36 percent. Previously maximum interest rates varied on a state by state basis depending by the licenses Prosper had acquired.

Prosper chose the same construct to go national as did Lendingclub in December - both partnered with WebBank, Utah.

All loans originated through the Prosper marketplace are made by WebBank, a Utah-chartered Industrial Bank. Prosper provides services to WebBank in connection with the origination of such loans and Prosper services loans made to Prosper borrowers on behalf of registered Prosper lenders who purchase such loans.

This step has good potential to multiply the monthly loan volume originated by Prosper, as chances for obtaining a loan were in the past harmed in some states by low state interest rate caps (especially for lower credit grades).