Collections. A subject close to everybody’s heart. Over the next two or three weeks, I intend to publish a number of posts discussing various aspects of collections – the status of the legal test and evolving legal strategy; debt sales and collections on charge-off accounts; additional plans for collections strategy and reporting. For the first post, I want to take a look at Prosper’s “primary collections” (Delinquency Days 31 to 121).
It is impossible to take a look at collections results without referencing the macro-economic environment in which they take place. Not to put too fine of a point on it, the current environment is the worst in decades.
In their most recent report, American Express announced an annualized charge-off greater than 10% (more than double what it was at year end 2007). In one of the lead presentations at last week’s convention of ACA Internation,
Kaulkin Ginsberg stated that the average liquidation of fresh delinquencies has declined by 55%-60% between 1Q07 and 1Q09 (for more information on Kaulkin Ginsberg see www.kaulkin.com – the power of their presentation can be found at http://kaulkin.com/files/KGC-Creative-Winning-Strategies_7-14-091.pdf).
With that as a backdrop, the following table takes a look at several statistics for Prosper accounts that reached 31 days past due. To explain why I picked the months shown – February 2007 was the first month where the number of new delinquencies exceed 200 accounts; September 2007 was my first full month at Prosper, March 2008 was the first full month of placements at AmSher and March 2009 is the last month for which we have complete primary data (an account reaching 31 days in March will have had time to charge off).
|
|
Placed |
% High CG |
% Low Bal |
% BK |
% Cured |
% Payer |
$/Acct |
|
Feb 2007 |
202 |
5% |
79% |
2% |
21% |
27% |
$ 142.39 |
|
Sep 2007 |
316 |
13% |
62% |
5% |
26% |
33% |
$ 201.43 |
|
Mar 2008 |
293 |
21% |
56% |
8% |
27% |
40% |
$ 251.79 |
|
Mar 2009 |
391 |
40% |
51% |
9% |
29% |
43% |
$ 205.97 |
Notes:
1. % High Credit Grade = % of Accounts of AA/A/B
2. % Low Balance = % of Accounts with Monthly Payment < $200
I include the mix variables of %High Credit Grade and % Low Balance to characterize differences in the placement populations. Not surprisingly, a low balance is easier to collect than a high balance. Conversely, it may be counterintuitive, but “good accounts” that go delinquent are more likely to go all the way bad than marginal accounts. This is why the AmEx charge-off rate has been harder hit in this recession than HSBC or WaMu (before they were bought by Chase).
I’m pleased with the fact that we’ve actually been able to improve the percentage of accounts on which we receive some collections payment. Truth be told, the “$/Acct” figure is the one that I focus on the most. We’re down 18% year over year – while much better than the 35-40% quoted in the Kaulin presentation – this is tied to people with payments in the $400-600 range seeming to “give up” sooner. We’re working on a test program to try to address this issue.
So what’s the bottom-line, am I satisfied with our collections? No.
Do I think AmSher doing a good job for us? Yes.
Where will we be a year from now? My goal is to have the $/Acct over $300 – if the economy really starts to improve, that shouldn’t too hard and I’ll up the goal. If the economy stays flat, it could be a challenge.
then I wait till I get it back since I don’t want to limit my losses if / when they come. 
] 
Prosper moderates all comments and will approve those that are directly relevant to the post. We do not publish comments that are spam, are offensive or appear to pass you off as another person.