On Viability
Another thing that came out at Prosper Days 2008 was some more insight on the pace of growth on Prosper and the burn rate of it’s Venture Capital.
The viability of Prosper should be a huge concern for Prosper lenders. If Prosper goes out of business one day, someone needs to service our loans – that is, receive payments, distribute them to the multiple lenders on each loan, collect from delinquent borrowers, etc… How can lenders be sure this will go well? I can only imagine the transfer of servicing from Prosper to whoever will be a big clusterschmuck as well.
The last session was a Town Hall type luncheon with Chris Larsen and John Witchel from Prosper and Bob Kagle (Director) from Benchmark Capital. During that session, the most critical of all questions was asked by Leporello. I don’t remember the exact quote, so I’ll paraphrase as best as I remember:
“Chris, you mentioned at the Hawaii Meet & Greet in summer 2007 that Prosper needs about 4-5X growth in loan originations to break-even. Loan originations peaked in April 2007, and have decreased in volume and growth has remained flat since. How much longer can Prosper sustain a platform that is not making any money?”Or paraphrased again more simply,
“How long before Prosper runs out of money?”Chris Larsen made some interesting comments in response. I don’t remember them all, but what I do remember (if correctly) is that he claimed to still have $20 million of Venture Capital still in the bank. Based on some
rough calculations by Trav (registration and verification required) on the prospers.org forum, that gives them about 4 more years of VC to play around with.
Bob Kagle also made the point that as a representative of Benchmark Capital, he is actually glad to see where Prosper is right now. That is, getting the gremlins out of the system before really revving it up. He also mentioned that his firm is ok with letting Prosper develop slowly, and most of the companies BC invests in are usually long term plays in the realm of 4-5 years for profitability.
Whether or not these statements were entirely truthful, they certainly make sense. Prosper still has at least 2 years to make this thing fly, or at least give the impression that it can by then. Considering how far they’ve come and how much time they really have left (as opposed to 6 months, as some have opined), things look promising for Prosper.
But what about growth? How have things been going? Can Prosper really work? Why are loan originations so much lower now than in early 2007?This got me thinking. (don’t laugh)
I’ve already commented on the impact of some of the recent changes a while ago, so I won’t repeat those arguments, but here are the basics:
Prosper’s early growth and large originations were not what one would call “good growth”. Lenders funded a bunch of crap, and the late/default rates show how dumb we were.
Fred updates his blog periodically with
late loan statistics.
As you can see, the results haven’t been too pleasant.
What is Prosper doing right now to combat lender stupidity? There are 2 major things that have changed in the last 6 months that are taking Prosper in a more focused direction.
The first is the Bidding Guidance, which I’ve already written enough about on my blog. By using their own data, Prosper is helping lenders stay away from total crap by showing lenders how loans with certain credit details have actually performed in the marketplace.
The other thing that Prosper is doing is coaxing lenders toward more reasonable loans with the Portfolio Plans. I’ve written a lot about those too, but they sure help lenders avoid crap as well.
How important are these recent changes? In my humble opinion, they are critically important. While still supporting the “open market” by letting almost anyone apply for a loan, Prosper is subtly shifting a majority of loans originated to a higher quality.
How should we define “a higher quality”? We all have our own lending standard, but since I used Prosper’s Select Index in an earlier PD08 recap, I guess well just use that.
Again, Prosper’s Select Index are loans with the following qualities: AA-E credit grade, DTI less than 40%, 0 Current DQ’s and 0-3 Recent Inquiries.
Let’s take a look at the numbers from the Performance page.
This first chart looks at the ratio of loans made that fall within Prosper’s Select Index vs. All Loans Made on a month-by-month basis. (click for larger version)
Thanks to the Bidding Guidance and Portfolio Plans, Prosper Select Index loans make up more than 50% of all the loans starting in January 2008. A marked shift to quality.
And what about total loan originations? Again, here is a month-by-month look at Prosper Select Index originations and Total originations. (click for larger version)
When looking at Total originations, Prosper’s volume peaked in April 2007. But when looking at the growth of Prosper Select Index loans, the last 3 months were larger than April 2007, with a recent peak in January 2008.
Why do I think this is so important?Remember Fred’s chart earlier of loan performance? Pretty scary, huh?
Let’s take a look at how Prosper Select Index loans have fared (first using the same scale):
And again, but with a better scale to see more detail:
Well, performance hasn’t been perfect, but it’s a much better view compared to Fred’s. Because Fred’s graph looks at the performance of all loans, there is quite undeniably a large overweighting of, well, crap that makes things appear so bad.
One more chart. This time I just shifted all the lines to the left so that they all start at the same point. This is one way to look at all of the Prosper Select Index loans at a given number of months past origination:
Not so bad.
So what am I trying to say? Hearing just these few things at Prosper Days 2008 made me feel much better about Prosper’s survival. Digging into the data just reinforced it.
* just to cross the t’s here, there are currently 68 loans in Prosper’s collections “Test” group classified as repurchased, but are really either defaulted or in collections. (the status is hard to define) Because of the repurchased classification, they do not appear on the late charts I created above. I do not know how many of these 68 loans fall in the Prosper Select Index, when they went late, or when they originated. If, for example, all 68 loans fall in the Index, this would push up the monthly late %’s by about 1%, but that’s the worse case. In any case, the performance of the Index is much better than the entire Prosper loan portfolio…
Part 8 - Collections
You see, Loan Sharking is when you lend money to people, and if they get behind on their payments you start breaking legs.
Comparing these results with the wide open Query 5 table, there doesn’t seem to be much value in an endorsement without a bid.
