John Dizard, who writes a pretty-much-weekly column for the Financial Times, typically presenting an exotic investment idea, has long given me the impression he spends much of his day gossiping with people on trading desks. Which means he is very much plugged in, and some of the remarks he makes in passing can be more interesting than his main argument.
This week, in “Useful strategy to take advantage of the Bear mess,” he provides quite a few of those very helpful observations-in-passing. For example, he mentions casually that not all of the “housing heavy fund managers” have reported their June results, and some are facing 30% to 50% losses of capital and preparing to fold.
We’ll give you the full text of Dizard’s piece at the end, but in brief, he discusses the thinking of Barak Laks of Alpha Beta Capital Partners, a fund of funds, who sees the current mess in subprimes as a buying opportunity.
Note that people like Laks start discuss the merits of an investment idea either 1) after they have established their position and are trying to get other buyers in to start an upward price movement or because 2) they want out and are talking up the other side of the trade. Given that his FOF has “heavy exposure to ABS strategies,” I wouldn’t discount the possibility that he is talking up his own book.
Nevertheless, there is an argument to be made that CDOs are oversold:
Right now, we are in a liquidation phase for subprime housing debt, and some other risky housing securities. Not the liquidation phase, because there will be several. Presently, the liquidation – a nice word for forced sale – of these assets is being compelled by the securities dealers, who have just gone from being the best friends of the hedge fund managers to their worst enemies. The dealers’ proprietary trading desks are also being harshly right-sized, as GE’s Jack Welch used to put it.
Dealers are aware they’re probably shredding positions that will eventually be worth more. They have highly leveraged balance sheets, which can only work if they are prepared to remain liquid.
But here we get to the juicy part: we’ve long known that these securities are complicated, each deal is sui generis, and the stuff doesn’t trade much (and I have argued that the paper is so heterogeneous that you’d need to have not merely a few issues, but quite a lot of issues and tranches trading to provide enough benchmarks).
It turns out the complexity is much worse than I imagined:
Accurately determining the value of many of the complex ABS securities is a labour-intensive job, given the securities’ highly differentiated capital structures, and the dealers don’t have the time. Instead, they do “matrix pricing”, in which three or four variables – such as average credit ratings for individual borrowers, average loan to (estimated) value, and where the security is in the “waterfall” of payments – are used to come up with a good-enough-for-now estimate of value. Then they wind up selling them for less than that, because there aren’t enough bids.
“This [process] can be a source of mispricing,” says Mr Laks. “They’re leaving out at least another three or four variables. To price one of these things correctly takes a week with a smart person, and they have to mark thousands of them. The manpower isn’t available to the dealers.”
A man-week to price a single security? Too labor intensive for the dealers to price it accurately? And this stuff was sold in volume to consenting buyers?
Now I have to tell you, in all my days in and on the periphery of the capital markets, I have never heard a anything issued in the volume of CDOs in which the pricing was beyond the skill not only of most investors, but most dealers as well (even if it’s a labor rather than skill shortage, the net result is the same). I wasn’t in the MBS business at the beginning, so I imagine the early deals might have been hard to price by virtue of the methodology being new, but the market evolved quickly beyond that stage. Similarly, in the early days of OTC derivatives, clients would infrequently ask for options that were very hard to price. Only one or two dealers were sufficiently skilled, and even then it would take them several days (they needed to figure out how they’d hedge the risk….). But in that case it was the dealers who were at risk, not the end buyers.
Dizard later tells us there are perhaps 50 CDO managers who have the resources and skills to price this stuff (and would Ralph Cioffi have been presumed to be on this list before the Bear hedge fund blowup?) Only 50 buyers who know what they are doing and a universe of chumps. And most of these buyers have fiduciary duties!
This could end up even worse than I thought if the credit markets take a nosedive. But Dizard’s buddy Laks thinks we are near the bottom of the current phase of liquidation, even as Dizard cautions us that there are likely future chapters in this tale. Both agree that rating agency downgrades would force “mechanical” selling by insurance companies and pension funds. In this context, “mechanical” means “with no concern as to price or outcome.”
Update, 10:30 AM ED. Tanta at Calculated Risk was good enough to express doubts about Dizard’s comments via e-mail. He talks about valuation difficulties with “complex ABS” without specifying what he means. She pointed out what he says most certainly does not apply to seasoned mortgage backs, but I am pretty certain this isn’t what he is talking about. I have e-mailed him and will hopefully get a response.
From the Financial Times:
Last week was a slow one for most Wall Streeters, with the exception of people who manage portfolios of housing-backed securities. They had a lot of damage to assess and a lot of assumptions to bin. Given the growth of the asset-backed securities (ABS) market in recent years, and the imbalance between the complexity of the securities and the inexperience of the new entrants, the scale of the losses that followed the train wreck of Bear Stearns’ two hedge funds was predictable. While most have not reported their results for June, there are housing credit-heavy fund managers facing losses of 30 per cent or even 50 per cent of capital. Some are already preparing to shut down.
Other managers, though, were prepared for an event such as the post-Bear unwind. One of them, Barak Laks of Alpha Beta Capital Partners, a fund of funds with heavy exposure to ABS strategies, is already working to exploit the opportunities arising from forced liquidations.
About a third of Alpha Beta is committed to ABS tied to housing or other forms of consumer credit. Overall, the fund was about flat for the month of June. The ABS strategies cost it something over 2 per cent for the month. Now Mr Laks is committing more capital to the ABS strategies.
“This market today is like corporate credit was in 2001 and 2002, especially like 2002, when you had a tremendous risk/return opportunity. The real money in bonds is made when they are trading on price, not spreads.” Bonds tend to be quoted in price terms when they are trading at or below 75 per cent of face value, and that now describes many riskier housing-backed securities.
If we are at the beginning of a depression, what is cheap will get even cheaper. However, if a depression, defined as a prolonged liquidation of capital, is even a couple of years off, the historic pattern of distressed debt markets suggests a medium-term opportunity. “These bonds can go to 25 before they go back to 60. If they get to 45, you have to ask if you have the staying power to remain in the trade,” Mr Laks says.
Right now, we are in a liquidation phase for subprime housing debt, and some other risky housing securities. Not the liquidation phase, because there will be several. Presently, the liquidation – a nice word for forced sale – of these assets is being compelled by the securities dealers, who have just gone from being the best friends of the hedge fund managers to their worst enemies. The dealers’ proprietary trading desks are also being harshly right-sized, as GE’s Jack Welch used to put it.
Dealers are aware they’re probably shredding positions that will eventually be worth more. They have highly leveraged balance sheets, which can only work if they are prepared to remain liquid.
Accurately determining the value of many of the complex ABS securities is a labour-intensive job, given the securities’ highly differentiated capital structures, and the dealers don’t have the time. Instead, they do “matrix pricing”, in which three or four variables – such as average credit ratings for individual borrowers, average loan to (estimated) value, and where the security is in the “waterfall” of payments – are used to come up with a good-enough-for-now estimate of value. Then they wind up selling them for less than that, because there aren’t enough bids.
“This [process] can be a source of mispricing,” says Mr Laks. “They’re leaving out at least another three or four variables. To price one of these things correctly takes a week with a smart person, and they have to mark thousands of them. The manpower isn’t available to the dealers.”
He estimates there are perhaps 50 skilled managers of ABS securities out there. Each can find a lot of value in the growing piles of junk.
Soon after comes the next stage of liquidation. While dealers will make mistakes, they aren’t as mechanical in decision making as the more rigidly regulated institutions, such as insurance companies and pension funds. When the ratings agency downgrades hit, they have to get this stuff out of the door. So another big step down for a lot of this paper.
It would be convenient if there was a precise formula for predicting the length of the liquidation process, but there isn’t. “In my opinion,” says Mr Laks, “you’re looking at the end of the first quarter of next year,” based on his best guess of the dealers’ and institutions’ decision-making cycle.
He could easily be off by months in either direction. So the trick is to buy positions that will allow you to wait. There are two forms of leverage for ABS paper: structural and balance sheet. Structural leverage arises from taking the riskiest collateral, balance sheet from borrowing against better collateral. “My rule,” Mr Laks says, “is to go through the structural leverage every time. Every disaster is balance sheet leveraged.”
And all these bottom-picking strategies are dependent on this bump not being The Big One. If it isn’t, there’s a lot of money to be made from the Bear mess.
