Is the SIV Accounting Fraud Nearing an End?

Anyone who's read this blog for a while will know that one thing I absolutely can't wrap my head around is the Great SIV Loophole, and why it was never addressed until the "structured investment vehicles" exploded messily.

Here's how everything worked before the financial crisis/meltdown/blowup: A bank -- let's call it "Smitigroup" -- creates an off-balance sheet vehicle; let's give it a sexy name like "Xena." Here's what lil ol' Xena does: it borrows money by issuing short-term securities, then in effect lends money by buying longer-term securities. Now, if you're a Joe Blow reader who's wondering, "Exactly why does, ahem, Smitigroup, want to do this?," let's look at what's going on.

You make the short-term borrowing at say 2.8 percent, then you buy longer-term products at say 3.5 percent. Notice the "spread" between the two. Borrow a million for $28,000, buy a longer-dated asset (like, say, a security backed by mortgages from some fast-appreciating neighborhood of homes in south Florida) that pays $35,000 a year, and you're pocketing a cool $7,000 annually from the difference.

Rinse. Wash. Repeat. (As they say.)

Now, if you're financially savvy, you may be thinking, "Hey, that sounds kind of like what my bank does." And bingo -- you'd be right. Your bank basically "borrows" short term from its depositors (paying them a small amount for their funds) then lends long term (home and business loans). But there is a difference: the structured investment vehicle is more like an invisible bank; it doesn't show up on regulatory radar.

So let's return to that riveting question: Why does Smitigroup want to create an SIV? Well, one reason obviously: the potential profit. But why create the SIV off-balance sheet?

Yes, why oh why? Because Smitigroup operates under capital constraints and other regulations. It must have a certain amount of underlying capital to be able to expand its traditional banking operations. But here's the neat thing about its SIV: Smitigroup waves a magic wand and Xena takes shape and starts operating at arm's length from its creator, so that the business doesn't eat into Smitigroup's capital base. Meanwhile Xena funnels profits back to Smitigroup.

Hey! What's not to like?

But then, you have a liquidity seize up -- those long-term assets Xena is holding become worth less, no one wants to buy its short-term securities, and Xena begins to circle the bathtub drain at an increasingly frenetic clip. So, no problem for Smitigroup right? Smiti won't be on the hook.

Wrong. Suddenly you see the tractor beams appear. Smiti hoovers up Xena, moving the vehicle's operations onto its own books. So much for the fiction that Xena was "independent"!

This is the absurd crap -- sorry, crap is the most polite term I can think of -- that was allowed to persist in the financial industry. Now though, that may be changing, Floyd Norris of the NYT reports in a meandering column:

The issue is a couple of new accounting rules that are forcing banks to put back on their balance sheets some strange creations that bad accounting rules had allowed them to shunt aside in the past.

Strange creations indeed! But as weird as these SIVs were, there was an even odder beast out there:

Bank holding companies have been allowed to issue something called “trust preferred securities.” The beauty of those securities was that they were really debt that the holding companies could call capital. Having that “capital” meant the bank could take on more debt. A system that lets a bank borrow more money because it has already borrowed money — rather than because it has sold stock — is hardly a wise one.

Got that? As a bank, I'm supposed to hold capital against my assets (mostly loans). The more loans I make, the more the capital ratio shrinks toward my regulatory minimum. But a "trust preferred security" turns that equation upside down. As I make more loans, my capital ratio grows. Ain't bank accounting grand!

Broadly, such tricks fall under the rubric of "capital arbitrage." Ah, if only we could have seen these capital arbitrage tricks at the time. But zee banks, zey are so clever! It would take a man of almost unimaginable perspicacity and intelligence, a man possessing perhaps clairvoyance even, to divine the gravity of the game that was afoot ...

Or maybe not:

In Spain, some smaller banks are in trouble from real estate loans, but the big banks seem to have emerged in good shape. One reason is that Spanish regulators were not fooled by things like SIVs, and insisted that if any bank wanted to create one, it could, but would have to hold reserves anyway. Since there was no business reason — other than capital arbitrage — for a SIV, those banks shied away.

Oh well. But at least the Financial Accounting Standards Board is finally getting around to sewing this loophole shut:

The FASB, in an attempt to save face and a degree of integrity, has pushed back on Wall Street by passing FAS 166 and 167 which would require investments in off-balance sheet vehicles to be brought on-balance sheet. The implementation of FAS 166 and 167 is imminent and would require financial institutions to set aside increased capital against selected assets.

This seems like a no brainer, a slam dunk. But the banks are squealing, predictably, because they have been hiding operations off balance sheet for a while and are worried about the impact of bringing them back onto the books.

The fact that we're even having this discussion is ridiculous. We need a better regulatory regime. I think capital-based requirements in a rule-based system (See Basel, incarnations I and II, e.g.) may be an obsolete approach, especially in an information-rich age of high-speed computers and financial innovation galore. The banks, with an army of lawyers in tow, will always twist and limbo their way around the requirements.

There is a better way. I'll look at that later this week.