Comparing Returns on Prosper Before and After the Auction Model Ended

Back in 2010 the management of Prosper took a gamble. They decided to do away with the auction model that had been the basis of their investing platform since Prosper.com launched in 2006. The change was made 18 months ago on December 19, 2010 so I thought now would be a good time to check in and see what impact this change has had on investors.

The Old Auction Model at Prosper

For all you newer investors let me explain how the auction model worked. A borrower would apply for a loan on Prosper and state a maximum interest rate that they would be willing to pay for that loan. Prosper assigned a rating to the loan and then investors would “bid” on the loan.

Let’s take an example to demonstrate how this worked. Betty wants to borrow $5,000 and is willing to pay up to 15% interest for than loan. Prosper gives Betty a B rating and investors start bidding. Many investors bid on the 15% rate, and soon she has enough bids to meet her request of $5,000.  Still some investors are interested in lending to Betty, and they are willing to go down to 14% or 13%.  The greater the investor demand, the lower Betty’s rate goes.  Betty’s loan is ultimately priced based on the pool of bids at the lowest rates totaling $5,000.  For example, if Betty had $5,000 in bids at a mix of 14% and 13%, the whole loan would be funded at 14%.

This worked ok in theory but it was a hassle for investors. The loan would stay on the platform for the full 14 days, which gave investors time to bid the loan lower even if it was fully invested.  That tended to slow down investors’ deployment of money on the platform, and some large investors did not like the uncertainty in how a particular type of listing would ultimately be priced.  Then there was the problem of constantly receiving emails saying you were outbid on a loan.

What Did This Change Mean for Investors

Regardless of which model we preferred I was curious to see how this change impacted returns. The table shows some statistics from all loans issued on Prosper in 2010 and 2011 (all the data is from Lendstats.com).

Summary returns for 2010 and 2011 at Prosper

So from this simple analysis it looks like the auction model was a better deal for investors because in 2011 returns went down. It was not a large drop: 8.22% compared with 9.52% but it was a drop nonetheless. I didn’t want to jump to any conclusions so I contacted Jim Catlin, the head of Risk Management at Prosper to get his opinion on this data.

When I asked him to explain this difference he said that while pricing and rating mix were affected by the change to fixed pricing, one of the biggest drivers of the high returns in 2010 was their evolving underwriting strategy. Here is what Catlin said:

Coming out of the quiet period in 2009 and 2010 our top priority was delivering great returns in order to build momentum with investors. In 2010 in particular we were very conservative with our underwriting in order to be sure investor returns came in better than the expectations we set with them when they invested in the loans. In that period there were some pockets of loans – for example HR loans – where actual loss performance came in at 60-70% of our expectation.  While that was great for lenders it meant that some borrowers were not getting the best possible price from us.

Over the course of late 2010 and 2011 we gradually adjusted our underwriting. We wanted to continue giving investors great, predictable returns, but we also wanted to increase the number of borrowers on our platform. The adjustments enabled more borrowers to get lower rates, and investors have still enjoyed returns consistent with the expectations we set when they invested.

Going forward we will continue to build on our impressive track record of delivering consistent, strong returns to diversified investors.  Over the last 3 years we have earned the trust of our investors by delivering returns and loss rates consistent with the expectations we set at the time of investment, and we look forward to continuing to serve our investors in that way.

In other words returns were slightly down for 2011 vintage loans over 2010 because of underwriting not because of the end of the auction model. Investors that put money to work in 2010 just got an exceptional deal – this vintage of Prosper loans may well end up providing the best returns of any year in p2p lending. Lucky for you if you were an investor during this period.

Overall I would say investors are better off with the fixed price system. I certainly prefer it over the auction model so I am glad Prosper took that gamble back in 2010. I think it has positioned them well for the future.

What do you think? If you are a long time Prosper investor did you like the auction model? As always I am interested to hear your comments.

  • Peter Renton

    Peter Renton is the chairman and co-founder of Fintech Nexus, the world’s largest digital media company focused on fintech. Peter has been writing about fintech since 2010 and he is the author and creator of the Fintech One-on-One Podcast, the first and longest-running fintech interview series.