Saturday, 4 February 2012

NYT's DealBook Gets It All Wrong on CLOs

This was a painful piece to read at the New York Times DealBook: "A Debt Market's Slow Recovery Is Burdened by New Regulation."

Steven Davidoff whips out his violin, tucks it under his chin, and plays a mournful lament for the market for collateralized loan obligations, which is under threat by none other than the act that the bankers have all flicked out their switchblades for: Dodd-Frank. Specifically, Dodd-Frank meanies want CLO originators to retain 5 percent of their funds, to prevent them from stuffing the investments with, er, crap.

The Times should have top-notch journalists who know more about CLOs and junk debt than Davidoff appears to here. Because, honestly, in this piece I can't tell sometimes if he is playing the fiddle or he is being played by someone else like a fiddle.

Here are five problems I have:

1. "CLOs are largely made up of loans that are at much lower risk of default than the risky high-yield, or "junk," bonds that also finance private equity buyouts."

He's trying to make these loans seem safer. But let's take the buyout for the old TXU, which as Davidoff notes, was made possible by leveraged loans and bonds. Right now TXU (renamed Energy Future Holdings) is rated CCC, deep into junk territory, only four levels from D (which, for non-credit junkies, means "D" for default).

If Energy Future goes belly up, the loans and bonds should both default. The loans will probably enjoy a better recovery rate, because they're higher on the capital structure and secured, but they will default. I'm not sure where Davidoff gets this idea that the default risk is "much lower" for leveraged loans than bonds. When a company is rated Total Junk, the default risk will be high for both.

2. "CLOs had a default rate of less than 1 percent even as the loans underlying them had a default rate of about 6.5 percent."

So? Davidoff seems to imply that CLO structurers have somehow transmogrified straw into gold by using this numerical comparison. Look! The default rate is much lower! But it's disingenuous and incomplete.

The CLO is the loans, in aggregate. Structuring doesn't magically conjure up a new income stream out of thin air. Those 6.5 percent of loans that defaulted -- they defaulted whether they're in the CLO, out of the CLO, or wrapped in something called a super duper wonderbar GCLO (giant CLO).

In fact, thought experiment: let's say you create that GCLO that almost never defaults, even when a mess of underlying loans do. Let's say the default rate is 6.5 percent for the loans, 0.8 percent for the CLO, and 0.001 percent for the GCO.

But let's say when the GCLO defaults, because it's so huge, there's a good chance the entire financial system will implode. So if I tell you that 6.5 percent of the loans have defaulted, but the GCLO has never defaulted, is that a good thing (by Davidoff's logic, it would seem to be)?

Or is it bad that when that GCLO explodes it's probably taking down the financial system, so you'd be better off having the loans outside of that only ostensibly super-safe structure -- out where they'll do less harm? (Note: if you've been thinking a lot about the financial crisis, you'll realize this is not an accidental sort of thought experiment.)

(3) "So new CLOs are crucial to support the corporate loan market. Without them, banks will be hampered from originating credit since they will be unable to sell these loans off their balance sheet."

Well, for decades preceding the invention of the CLO, the financial system seemed to bumble along just fine. In fact, banks were regarded as being safer because they kept loans on their balance sheet, because that meant they were more circumspect about what lending they did. Here, Davidoff is mourning an end to the easy money that securitization threw open the spigots to. But he doesn't really seem to reflect on whether being awash in easy money was a bad or a good thing.

(4) "Most managers do not actually originate the loans underlying these financial instruments [CLOs]. Instead, the manager buys these loans from originating banks."

So he claims that leads to a "secondary-market check" that "may be lacking with other structured products."

Davidoff says this to argue that CLOs are unlike the mortgage-backed securities that crashed and burned, and thus deserve to be exempt from a rule about securitization originators retaining 5 percent of the deal. But wait a minute -- check out this post about "Why Structured Finance?" -- because the securitization model for mortgages doesn't require that a structurer originate the debt either.

In fact, before the crisis there were plenty of Wall Street banks bundling mortgages that they never originated. So Davidoff's "secondary-market check" looks like a rather weak safeguard.

(5) "And of course, private equity would also be hurt if the CLO market dried up."

Ah, private equity. You know, Mike at Rortybomb just showed us that much private equity activity occurs because of our screwball tax code -- not because it makes sense, business-wise, and James Surowiecki at the New Yorker had a good bash on the subject too, revealing that these raw-meat capitalists actually get a lot of hidden supports that make them look more like welfare queens.

Davidoff never pauses to consider that it might be a good thing if private equity buccaneering was curtailed.

Of course Davidoff misses the really big story about the CLO market: mathematically, these things just don't make much sense. With all the fees that they suck out, they should collapse under their own weight.

Before the financial crisis, they were priced close to real AAA, as investors were fooled by the complexity into thinking they really were AAA safe. Today, investors aren't that dumb: they're onto the game and demanding higher, non-AAA yields for debt that's graded AAA.

But why would investors play along with such shenanigans? Check out:

The Ratings Charade Continues: A CLO Investigation (Part I)
The Ratings Charade Continues: A CLO Investigation (Part II)