Calling so-called raw delinquency statistics "meaningless and misleading," IndyMac chairman and CEO Mike Perry said late last week
that the thrift -- which lost $509 million in the fourth quarter and suspended its dividend amid increasing borrower delinquencies -- would cease reporting them, as it looked to shift its reporting to so-called static-pool data. Perry said the company was "working feverishly" provide static-pool data to investors by the end of the second quarter, if not sooner, but would stop publishing raw data by product and channel, effective immediately.
Some background may be in order: Static-pool reporting is the practice of benchmarking performance data against total production in some given "static pool," whereas the raw deliquency statistics Perry is mentioning benchmark against current unpaid principal balance.
Perry's argument is illustrated simply. If a particular group of loans prepays to the tune of 80 percent, so that only 20 percent of the original group remains, benchmarking delinquencies against the remaining 20 percent would make delinquency numbers look a whole heck of a lot higher to investors than benchmarking against the original 100 percent. So the bottom line here -- according to Perry -- is that the performance of loans originated depends on what you're comparing them against.
Perry said in his remarks that "sophisiticated" investors only rely on static-pool analysis, and that "the public, press and even some equity analysts are obsessed with raw servicing delinquency data."
Despite Perry's ranting on the subject, I have plenty of questions that I'd love to see answered.
First, it appears to me that Perry is confusing "static-pool analysis" with "vintage analysis."
The example he gives in his explanation -- loans originated in 2002 -- would be considered a vintage analysis
by those so-called "sophisticated investors" he refers to in order to butress his point. A static-pool analysis, by way of contrast, refers to an analysis of pooled loans within a particular securitized trust, which are usually some subset of a particular vintage and are likely also more stratified that a true vintage analysis would be.
Look at it this way. "All Alt-A loans originated in 2007" are a vintage. "Alt-A loans, originated between January and March of 2007, with a weighted FICO of 700, originated by Countrywide and IndyMac, comprised mostly of no-doc and doc-lite underwriting" would comprise a static-pool, generally speaking. True static-pool analysis involves the performance tracking of a pool of investments with similar risk-return profiles, typically loans that have similar vintages and underwriting criteria. Not just all loans in a particular vintage.
So you'll have to excuse me for being thrown off when Perry launches into the benefits of static-pool analysis and then proceeds to use a vintage analysis to justify his argument. The two serve very different purposes for that all-important "sophisticated investor," and I'd expect the CEO of one of the largest independent mortgage lenders to know the difference.
Perry does bring up a very useful point, in a general sense: delinquencies can be tough to read, especially since no company handles reporting consistently. Some report on UPB including foreclosures, others on UPB excluding foreclosures, and still others on actual units (rather than on a principal balance basis). It's enough to make an analyst cringe and decide that an orange is, for all intents and purposes, an apple.
Which is part of the reason that a little thing called RegAB
was tossed into the ring during 2004 by the Securities and Exchange Commission; organizations were required to be compliant fully by 2006 with the items defined by the regulation. One of those items requires static-pool disclosures by all issuers for any securitization, so that "sophisticated investors" could have a decent shot at making an apples-to-apples comparison.
And that lands me at my second question: what does Perry mean by saying that IndyMac wants to provide "static-pool" data to investors?
For one thing, it's already been doing just that for any loans it services that are part of a securitized trust, since RegAB requires it; so providing that data shouldn't pose much in the way of "working feverishly" to prepare it.
For those whole loans it's holding in portfolio (and that aren't securitized), the question becomes what the definition of a "static-pool" really is in IndyMac's eyes. We don't yet have the answer to that. But in the meantime, why not pump out the static-pool data it does have on hand? If Perry is interested in fostering transparency with investors, that would seem to be a very good place to start.
On to question three: when are market participants going to recognize the effects of credit quality relative to, and directly on, prepayments?
Perry points to prepayments as the chief reason raw servicing data is so skewed, because they reduce the denominator used in raw deliquency calculations. That's only correct so far as it goes, and it likely signals a realization on his part that Alt-A and subprime aren't coming back anytime soon -- making now a "good" time to stop reporting the raw data by servicing category, since DQ percentages are only going to go up for each as a result.
But if we've learned anything in this credit mess, it's that all prepayments are not created equal -- and that prepayments aren't the only reason loans in a portfolio will run off.
First off, there are prepayments that are voluntary, and those that aren't. Think of it this way: a borrower that would have defaulted in 2006 refis into a new loan in 2006 and now defaults in 2008. That's very different sort of prepayment than a creditworthy borrower deciding to refinance because they simply want a lower payment. The real
problem with the 2006 and 2007 vintages, at the core, isn't prepayments per se; it's that the game of musical chairs finally stopped for those borrowers whose previous defaults had essentially been "revintaged."
So the 2003-5 vintages end up looking great from a credit perspective, even if prepayment velocity is off the charts; analysts start making complex models that only look at the effect of prepayments in whatever static pool they've got, and everyone declares credit risk mostly irrelevant. In contrast, the 2006-7 vintages look horrible from a credit perspective, prepayments slow and become much more volatile, Wall Street takes a look at its models and realizes some important data was missing -- and, of course, lender CEOs have to pen very public explanations explaining that prepayments are "screwing everything up."
With all of this talk of Federal regulation, and with consumers and industry types all pointing at delinquency data to butress their own set of arguments, perhaps some of our collective energy would be better spent figuring out how to get delinquency data right. I mean, really
right. Like maybe developing a national standard for loan portfolio reporting that matches what's done on a securitized basis, and then forcing anyone holding a loan to adhere to that standard. Maybe that's what IndyMac and its fearless leader are yet out to do.
But short of that, we'll always end up with companies cherry picking a reporting standard that works well at the time.
Editor's note: The author is indebted to numerous market participants, including Calculated Risk's Tanta, for comments on an earlier draft of this story.