Wednesday, April 23, 2008

Bidding Guidance Variation Explained?

In my last post, I looked at the differences in loss figures shown in the Bidding Guidance and the Performance Page from which these figures are based on. (here) Andrew from Prosper dropped in to explain what might be happening.


“Andrew from Prosper here. What you're seeing is expected – on some of the credit grade segments, we manually adjust the default rate up or down. We do this in one of three cases: 1) if the segment is very thin, and the performance page doesn't provide any data at all, 2) if the segment is thin, and the performance page uses default roll rates, but we think the actual roll rates will be higher or lower, or 3) in the case of E and HR borrowers, where the credit grade definition changed in Feb 2007.”


But I have to say, I’m still a little confused by the process which Prosper derives its Bidding Guidance historical loss figures. I decided to take a closer look at non-auto funding AA grade loans.

In Andrew’s three cases, case 1 and case 3 do not apply (there is PP data and these are AA loans). So we’re left with case 2: “if the segment is thin, and the performance page uses default roll rates, but we think the actual roll rates will be higher or lower”

Here is a table of AA grade non-auto funding credit slices. It shows the number of historical loans in each slice, the number of late and defaulted loans, and the loss projection (net default + adjustment) shown on both the Performance Page and the Bidding Guidance.



As you can see, yes, these slices are very thin. So it appears that Prosper is tweaking the roll rate assumptions when calculating the figures used in the Bidding Guidance. What is the process for tweaking? I have no idea.



At the end of the day, what we are seeing is slightly higher (except a few cases where the increase is substantial) loss projections displayed in the Bidding Guidance. This difference falls in the lender’s favor, so I guess it’s a good thing.

The only problem I have is that I cannot recreate these figures even though the criteria are clearly spelled out at the bottom of the Bidding Guidance page. I recommend that Prosper revise the first footnote to more properly define where these numbers are coming from:

"1 Estimated average annualized loss rate based on the historical performance of Prosper loans for borrowers with similar characteristics, originated between Jun-01-2006 and Jan-31-2008, measured as of Mar-10-2008, and tweaked a little based on intuition, our magic 8 ball, and input from our credit consultants Carnac Magnificent Consulting, LLC. Actual performance may differ from estimated performance due to many reasons, for example, worsening economic conditions."

2 comments:

j9359 said...

One thing that bothers me about the bidding guidance and performance pages is that there is no information about loans that did not fund.
I haven't completely thought this through yet but...
How about loans that match the criteria but do not fund? Say we are looking at C listings that are 0-0-0 with DTI <= 40 and asking for $5,000 or less. Suppose the estimated loss is 4%, so if I targeted an ROI of 10% the bid would be 14%.
I'd like to know how many listings that matched this criteria did not fund. If it is 1 out of 100 then I wouldn't be worried, but 1 out of 10 or worse might indicate that there is something besides the numbers going on.
I'm looking for some sort of confidence indication that tells me whether the performance of a particular segment is due to the numbers or is it because of careful loan selection by lenders.
Hope this makes some sense, maybe I'll post more on my blog.
john.

hoopdeez said...

you coulda called me first.