Friday, September 16, 2011

Regarding UBS, Their Rogue Trader, John Mack, Morgan Stanley & Howie Hubler

Yesterday's two big financial stories, outside of the continuing soap opera "As The Euro Implodes," were John Mack's retirement from Morgan Stanley and the revelation that UBS rogue trader Kweku Adoboli lost $2B for the bank. Both stories dutifully appeared as prominent pieces in this morning's Wall Street Journal.

There is, however, a deliciously ironic link between the two tales. For me, it took finishing Michael Lewis' The Big Short to realize the connection. Here's the clue from the Journal piece,

"But in 2007, it hit a rough patch, losing $9 billion on a proprietary mortgage bet that cost the jobs of several of Mr. Mack's lieutenants, including Zoe Cruz, who was widely viewed as a top candidate to succeed Mr. Mack."

Now, it's tempting to think of the Morgan Stanley loss as not caused by a rogue trader. In fact, elsewhere in the Journal article on Mack, it contends,

"He's a battlefield commander who knows how to lead troops......Mr. Mack pushed traders to take more risk and sell more esoteric but profitable mortgage products. Briefly, that move paid off. In 2006, the firm enjoyed record results."

Funny, that's not at all how Michael Lewis described Howie Hubler's calculated $9B loss on mortgage-backed derivatives.

How do you suppose UBS suffered only a $2B loss and blamed it on a rogue trader, whereas John Mack presided over a $9B loss in 2007 and kept his job? In fact, he essentially lost the firm, as it required a federal bailout, plead to convert to a chartered commercial bank, and had to arrange other financing to remain solvent.

Yet Mack kept his job and never called Hubler's loss a rogue trade.

We don't yet know precisely how Adoboli accomplished his allegedly-undetected losing trade(s). One assumes, like Nick Leeson who brought down Barings Bank, he somehow fooled the various electronic systems purporting to monitor positions, desk P&Ls and risk.

But here's how Michael Lewis explains Morgan Stanley's star fixed income trader Howie Hubler's stunning $9B loss. For those wishing to follow along, I'm synopsizing Lewis' Chapter 9 in The Big Short (Norton, 2010), entitled A Death of Interest, pages 200-225. I'm not going to retype the chapter, nor major parts of it, as that would, I think, be a waste of my time, and probably violate Michael Lewis' copyright. Instead, I'm going to abstract the highlights of what went on at John Mack's (and Zoe Cruz's) Morgan Stanley during 2006 & 2007. But I heartily recommend, if you find any of this of interest, that you run over to Amazon and buy a copy (new or, as I prefer, used but 'like new' for only about $6) of Lewis's book.

Howie Hubler was a star trader of mortgage-backed bonds. Sometime in 2004, Hubler begins to realize that a lot of the bonds he's selling, for which Morgan Stanley, like Goldman Sachs, has built its own origination system to capture the entire profit stream, are of highly suspect quality. So he, in conjunction with a colleague, Mike Edman, hatch a proprietary swap which they convince some of their clients to sell to Morgan Stanley. This allows Hubler's desk to own protection, for a small annual fee, on billions of dollars of dodgy subprime-backed mortgage bonds.

Lewis writes,

"It's now April 2006, and the subprime mortgage bond machine is roaring. Howie Hubler is Morgan Stanley's star bond trader, and his group of eight traders is generating, by their estimate, around 20 percent of Morgan Stanley's profits. Their profits have risen from roughly $400 million in 2004 to $700 million in 2005, on their way to $1 billion in 2006. Hubler will be paid $25 million at the end of the year, but he's no longer happy working as an ordinary bond trader......"

Along the lines of my long-expressed belief, Lewis notes that "the best and the brightest Wall Street traders are quitting their big firms to work at hedge funds, where they can make not tens but hundreds of millions."

So Hubler wangles a deal with Morgan Stanley to set up his own proprietary trading group, take his existing desk's swaps positions with him, and get a sweet deal for his group to own a stake in the group, to be subsequently spun out from Morgan Stanley. In short, Hubler extorts Mack and Cruz to give his group a semi-private business and stake in its future value, in return for not bolting from Morgan Stanley.

Here is where the story becomes interesting as it relates to today's WSJ articles about a rogue UBS trader and John Mack being paid homage as a wise senior Wall Street CEO.

Hubler's new group is given a profit bogey of $2B, but is paying about 10% of that, or $200MM, in fees to maintain its swaps, or shorts on bad mortgage bonds. Hubler wants to eliminate this drain, so he sells credit default swaps on a much larger amount of allegedly higher-quality mortgage bonds. About $16B of bonds.

As Lewis puts it,

"In effect, Howie Hubler was betting that some of the triple-B-rated subprime bonds would go bad, but not all of them. He was smart enough to be cynical about his market, but not smart enough to realize how cynical he needed to be."

For some perspective, the original credit default swaps which Hubler and his colleagues sold to their customers only required a 4% default rate among the subprime mortgages backing the bonds, which Lewis writes was expected in good times, to allow the swaps to pay off.

Next, Lewis writes about how Zoe Cruz' risk management people ask for stress tests of Hubler's aggregate positions under scenarios involving a default rate of 10%. A rate Hubler's people, who, remember, are just traders- not PhDs in economics or seasoned mortgage industry researchers (like, for example, Lew Ranieri's original Salomon group contained)- protested would never occur.

I need to make a brief side point here on which I'll elaborate in a subsequent post. Value-at-risk is the main component of most trading desk risk management systems. I've worked on and been around these systems since my days at what is now Accenture consulting back in 1995. One of the problems with VAR systems is that, since they require and assume variance in valuation to impute the capital required for a position, and, thus the losses possible at some probability level, they don't work well with bespoke instruments. Like, say, credit default swaps or so-called 'off the run' fixed income instruments.

Since the entire credit default swap market, as it evolved among AIG, Goldman Sachs, Deutsche Bank, Morgan Stanley,, was a telephone bid/ask market in which valuation was an exercise in judgement, variation in values was meaningless. As my old, sometime-business partner Bob Mankin is fond of saying,

"A model can tell you what something was worth yesterday or may be worth tomorrow. But the only way to know what it's worth today is to sell a piece of it in the market to someone else."

Thus Morgan Stanley's risk group's slow realization that its VAR reports on Hubler's groups risks were also meaningless.

The 10% stress test showed that Hubler's group wasn't short subprime mortgage-backed bonds. It was long, and the 10% default scenario would create a $2.7B loss. Lewis notes that the actual eventual default rate on the bonds on which Hubler sold swaps, i.e., went long, became 40%.

Again, to synopsize, by mid-2007, Deutsche Bank, which had bought the swaps Hubler sold to get a $200MM income stream to offset his negative carry on his base position of being long credit default swaps on allegedly-worse subprime mortgage-backed bonds, called Hubler to demand payment on the shifting, now higher value of the swaps Hubler sold. Again, because they aren't exchange-traded or continuously-quoted instruments, their value was the subject of what, in effect, became a verbal pissing contest between Deutsche and Morgan Stanley. By later in 2007, Morgan Stanley had paid Deutsche Bank at least $3.7B, eventually losing the net $9.2B. Hubler had, as Lewis writes,

"...been allowed to resign in October 2007, with many millions of dollars the firm had promised him at the end of 2006. The total losses he left behind him were reported to the Morgan Stanley board as a bit more than $9 billion: the single largest trading loss in the history of Wall Street......Hubler and his traders thought they were smart guys put on earth to exploit the market's stupid inefficiencies. Instead, they simply contributed more inefficiency."

To further add to today's trading loss/John Mack retirement irony, Lewis continued, on page 216,

"The other, bigger, buyer was UBS- which took $2 billion in Howie Hubler's triple-A CDO's, along with a couple of hundred million dollars' worth of his short position in triple-B-rated bonds.....A few months later, seeking to explain to its shareholders the $37.4 billion it had lost in the U.S. subprime markets, UBS would publish a semi-frank report, in which it revealed that a small group of U.S. bond traders employed by UBS had lobbied hard right up until the end for the bank to buy even more of other Wall Street firms' subprime mortgage bonds.....said one UBS bond trader close to the action, "It was a very controversial trade in UBS. It was kept very, very secret.....He further explained that the traders at UBS who executed the trade were motivated mainly by their own models- which, at the moment of their trade, suggested they had turned a profit of $30 million." "

Small world, indeed, eh?

Now with all this information, stop and reflect with me for a moment.

You are John Mack. It is 2006. You meet with your chief risk officer, Zoe Cruz, to ask for a complete examination of Howie Hubler's group's positions, strategies, assumptions, forecasts, etc., because that single eight-person desk is generating, according to Lewis, roughly 20% of Morgan Stanley's profits. You tell Zoe to set aside an entire day- maybe two. You surely want to understand in detail how these eight traders are producing 20% of your firm's profits, and what concomitant risks they are taking to do so.

Or maybe you don't.

The same thing would seem to apply to Zoe Cruz, would it not? Wouldn't she want to protect herself by conducting such a thorough examination of Hubler's group's business, in order to brief Mack prior to presiding over a blow-up of that desk's business?

Evidently not.

Instead, Lewis quotes verbatim from Mack's December 19, 2007 investor phone call. I won't republish the detailed exchanges between Mack and his questioners, including Goldman Sachs analyst William Tanona. Instead, here's what Lewis wrote about Mack's statements,

"The meaningless flow of words might have left the audience with the sense that it was incapable of parsing the deep complexity of Morgan Stanley's bond trading business. What the words actually revealed was that the CEO himself didn't really understand the situation. John Mack was widely regarded among his CEO peers as relatively well informed about his bond firm's trading risks.....Yet not only had he failed to grasp what his traders were up to, back when they were still up to it; he couldn't even fully explain what they had done after they had lost $9 billion."

In a footnote to the verbatim exchange on page 218, Lewis also wrote,

"What John Mack's trying to say, without coming right out and saying that no one else at Morgan Stanley had a clue what risks Howie Hubler was running, is that no one else at Morgan Stanley had a clue what risks Howie Hubler was running- and neither did Howie Hubler."

In an earlier footnote on page 210, Lewis provides a brief discussion of the differing explanations offered by those close to Hubler and Cruz regarding who was ultimately responsible for Hubler's positions' losses- Hubler or Cruz? Lewis sides with those who believe Hubler hoodwinked Cruz into believing the positions' net risks were minimal.

So, what is my point at the end of this very long post involving today's WSJ articles about yesterday's two big breaking stories at UBS and Morgan Stanley, and their mortgage-backed credit default swaps losses four years ago?

It's that, to me, John Mack shouldn't be retiring now as chairman of Morgan Stanley. He should have been fired by his board in 2007 for allowing Hubler and Cruz to lose $9B on one desk. Then Cruz should have been fired, and Hubler, Cruz and Mack all sued by the Morgan Stanley for fraud and breach of fiduciary duty to the firm's shareholders.

That if Howie Hubler was allowed to exit with tens of millions of dollars, and no criminal charges, after putting a $9B hole in Morgan Stanley's 2007 balance sheet, maybe UBS' Adibolo isn't guilty of anything, either.

Maybe both Hubler and Adibolo are rogue traders, or neither one is.

Now, I know that the foregoing story paints Adibolo as a trader trying to hide his known losses, whereas Hubler is portrayed as just too inept to realize what he thought was a net short position in derivatives was actually a net long position.

Personally, I have some trouble really belieiving Hubler was that stupid. Or, if he was, are we actually to believe that people that stupid can make $25MM a year? Plus, Cruz's people did the stress test which alerted them, and Hubler, to just how risky his positions actually were. So, from that point on, it seems unarguable that all concerned at Morgan Stanley could or should have known the truth about Hubler's positions.

But Lewis' book provides details of how Morgan Stanley dithered over exiting the worsening parts of the group's positions. Was that rogue behavior? By Hubler? Cruz?

Something doesn't add up. That's why I don't see a real difference between Adibolo and Hubler. They both were given license to risk their firms' shareholders' capital, and both lost it in positions which should never apparently ever been allowed to exist.

And, yes, Zoe Cruz did lose her job. But not quite the way I suggested above. And neither she, nor Mack, nor, of course, Hubler were sued for violating their fiduciary duty to their firm's shareholders through either gross incompetence, given their compensation and senior positions, or calculated deceit.

I think Lewis' account puts this week's comparatively paltry $2B UBS loss in perspective. And suggests that UBS has an evidently continuing cultural blind spot that makes it vulnerable to rogue trading, however you choose to define the term.

Lewis' book also begs the question, in my view, that Hubler was also a rogue trader. And perhaps Mack was a rogue CEO all the while.

I suspect, if you asked him, that Michael Lewis would say the whole lot- Mack, Cruz, Hubler, Adibolo, UBS's senior management, the boards of Morgan Stanley and UBS- are rogues.

And that anybody who buys shares in those firms, or any of their Wall Street ilk, are foolish and deserve what happens to them. Because for less than twenty bucks, any of those shareholders could buy both of Lewis' appropriately well-regarded books, Liar's Poker and The Big Short, and thus be warned of the risks of owning shares of formerly-private investment banks or brokerage firms.

1 comment:

Reno said...

You are so very correct. Rogue, my butt. This is the biz model. Thanks for your clarity.