Maybe The SEC Decided To Throw A Trial So It Can Go Back To Settling Everything

Author:
Publish date:
Updated on

I try to be honest when telling you that a court complaint or SEC filing or research paper is a fun read, just in case you might go read it, though of course there's no accounting for tastes and I may enjoy many things that you don't.* And that's okay. In any case I doubt anyone will find the SEC's fraud complaints against Fannie Mae and Freddie Mac filed today all that fun to read. "Very, very boring" would be more like it. The only bits that I enjoyed were the names of some of the loan programs, including Freddie's "Touch More Loans" and the Fannie/Countrywide joint effort "Fast and Easy" which, boy, different times.

But there are some fascinating things about the case. A small one: I was kidding when I said "complaints against Fannie Mae and Freddie Mac." They're complaints against former Fannie CEO Daniel Mudd, former Freddie CEO Richard Syron, and a handful of their executives. The SEC signed weird neither-admit nonprosecution agreements with Fannie and Freddie themselves, in which the GSEs agree to help the SEC make its case against their former bosses.

This all seems like very good PR. You are learning, SEC. The neither-admit-nor-deny thing might be awkies, but slapping a big fine on the taxpayer-funded GSEs wouldn't make a whole lot of sense. And the people who are upset that the SEC are not going after big names connected to the financial crisis have to be happy about the fact that the SEC here is going after the CEOs of big entities that in most people's minds are intimately connected to the cause of the financial crisis. Suing them is not quite as good as throwing them in jail, but the SEC can't do that, and this is a start anyway.

The bad news is that the SEC's case sounds just absolutely terrible. Here it is:
(1) Fannie and Freddie guaranteed some mortgages that were risky.
(2) Internally, Fannie and Freddie noticed that those mortgages were not dissimilar to subprime mortgages in underwriting standards, FICO scores, LTVs, and default performance. Sometimes they even used the "S" word in emails to each other.
(3) But they told the market "we don't have much subprime in our single family portfolio."

Which, yeah, that sounds pretty fraudy. Except that "subprime" is not a word with a specific meaning. I am pretty unsophisticated re: mortgages and stuff, but I actually thought that "subprime" kind of meant "having credit quality or underwriting that was too bad to qualify to be guaranteed by Fannie or Freddie," which if it were true would mean that by definition none of the GSE's single-family portfolio could be subprime. (It seems not to be true, since F&F did disclose some minor amount of exposure to things that even they called "subprime.")

But, still, sounds bad. Until you read the complaints. Let's pick an example at random from the Frannie complaint (paragraph 131):

On November 9, 2007, for the quarter ended September 30, 2007, Fannie Mae also filed a "credit supplement" on Form 8-K with the Commission. The document contained a summary description of certain credit risk characteristics of its Single Family book of business in chart form. Included in this chart were separate columns identifying Fannie Mae's subprime holdings and designating that 0.3% of its Single Family holdings were subprime loans. This supplemental disclosure did not inform investors of the additional subprime exposure from EA and MCM loans, or loans originated by the subprime divisions of large lenders.

Terrible! Now here's the chart they're complaining about:

Is that chart the clearest and most user-friendly thing in the world? Probably not. Does it give you one clear picture of "we have this many loans that are just peachy, and this many loans that are probably going to default?" No. Is it accurate, outside of that disputed right-hand column? I have no idea.

But notice what it does. It tells you how many loans had FICOs below 620. How many had original LTVs above 90%. How many had both. How many loans in the Fannie single-family book were negative amortizing, and how many of those had sub-620 credit scores.

You could probably give a name to a mortgage given to someone with a bad credit score, a sub-10% down payment, and adjustable monthly payments that don't reduce the principal owed on the loan. You could call it something like "risky," or "crappy," or "not the sort of loan I would normally make if I were running a bank or something." Or you could call it "subprime."

I don't think I'm taking an unfair example here. Here's a Freddie disclosure that the SEC complains about for the line "Our acquisition of subprime loans has increased in recent years, but remains a very small percentage of our single-family mortgage credit book." But read Question 4, where Freddie asks itself "hey, besides 'subprime,' what else do you got that might go bad?" and then breaks down its high-LTV, low-FICO, and funky-structure loans. Is it perfectly transparent? No. Does it give you a good sense of how many loans look crummy in one way or another? Sure.

I've looked at a couple of other examples, and would invite you to do the same but, again, quite boring, and they all go like this. There's a half-hearted claim of "we don't have much of a thing that we will call subprime," and then clear quantitative disclosure of loans that have the risk factors that you and I might think of as making them subprime.

It. is. just. incredibly hard to believe that humans and analysts and investors, in thinking about Fannie and Freddie's credit exposure on their mortgage portfolios, stopped reading when they got to the part where Frannie said "we don't have much subprime in our single-family guarantee business." More likely, they acted like, y'know, analysts and investors and actually looked at the numbers - which the SEC, for all of its newfound quant skills, seems not to have done.

That doesn't mean that Fannie and Freddie did nothing regrettable - calling things "subprime" in internal emails and "not subprime" in SEC filings is the kind of thing that, just, come on people, really? Really? But the case that this behavior was important to investors - that this was "securities fraud" - or the broader moral-PR case that Frannie's subprime disclosure was a key element of the financial crisis seems really hard to make.

You could if you like speculate on why the SEC is bringing a case this weak. (To hurt Newt Gingrich?) One answer could be "they were sick of all the criticism they got for not going after big dogs, so they slapped together a weak suit against some of the least popular big dogs around." I prefer a more contrarian theory: they were sick of being told "you should take cases to trial instead of settling," so they found some individual defendants (who are less likely to settle) and brought a case against them that they'll lose, badly, at or before trial. And then everyone will go back to saying "yeah, SEC, you'd probably be better off settling the next one."

SEC charges former Fannie Mae and Freddie Mac executives with securities fraud [SEC]

In Hunt for Securities Fraud, a Timid S.E.C. Misses the Big Game [DealBook]

* Totally unrelated aside: I can't really tell if this paper is a fun read. I can't really make heads or tails of it and find it creepy and counterintuitive. But it seems sort of neat.

Related