Tom Adams, an attorney and former monoline executive, provided considerable input into this post.
There is nothing more useful to people in authority than when a writer with an established brand name does their propagandizing for them.
Harvard Law graduate and Pulitzer Prize winning author James B. Stewart penned a remarkable little piece in the New York Times over the weekend. Titled “,” it contends that against a proposed $285 million SEC settlement with Citigroup over a $1 billion CDO (Class V Funding III) that delivered $700 million in losses to investors and $160 million in profits to Citi is misguided. Stewart argues, based on “some reporting,” that the SEC is unlikely to do better in the trial that Rakoff has forced on the agency by nixing the settlement.
We will look at the caliber of Stewart’s “reporting” in due course, since his article reads like dictation from the SEC’s head of enforcement Robert Khuzami (the SEC’s interests are aligned with Citi’s in wanting the settlement to go through). Stewart either did not read or chose to ignore critical information in the underlying complaints, which the Rakoff ruling cites, and he also overlooked relevant cases.
But let’s first examine the dead body in the room that Stewart and other commentators have conveniently managed to overlook.
Why hasn’t the SEC been tough on CDOs, even though they were at the heart of the crisis? As we discussed at length in ECONNED, it was the collapse in the value of AAA tranches of so-called asset backed CDOs (ones based heavily on subprime-related exposures) that blew holes in bank balance sheets around the world. They had not only served as collateral for short term funding (repo) but were also used for bonus gaming by traders (the “negative basis trade”). This is the big reason why so many banks had large CDO exposures when the credit markets started backing up. While they were stuck with some unsold inventory,the big reason is that traders loaded up on them to loot their own firms.
So why has the SEC not pursued this area more vigorously? Rakoff is puzzled by the SEC’s willingness to enter into the Citi agreement:
It is harder to discern from the limited information before the Court what the S.E.C. is getting from this settlement other than a quick headline.
The SEC went after Goldman only on one Abacus deal out of 25 in its program. Even though the $550 million settlement was limited to that transaction, it was widely understood that the SEC was not going to pursue Goldman on other CDOs. And it hasn’t. The SEC has gone through the motions in this arena: it poked around some Magnetar deals (not surprisingly, after the hedge fund got some real press about its destructive strategy) and negotiated a on a particularly noxious Magnetar trade, a CDO squared imaginatively called Squared.
Look no further for an answer than the SEC chief of enforcement, Robert Khuzami, Stewart’s primary and probably sole source for this article. He was General Counsel for the Americas for Deutsche Bank from 2004 to 2009. That means he had oversight responsibility for the arguable patient zero of the CDO business, one Greg Lippmann, a senior trader at Deutsche, who played a major role in the growth of the CDOs, and in particular, synthetic or hybrid CDOs, which required enlisting short sellers and packaging the credit default swaps they liked, typically on the BBB tranches of the very worst subprime bonds, into CDOs that were then sold to unsuspecting longs. (Readers of Michael Lewis’ The Big Short will remember Lippmann featured prominently. That is not an accident of Lewis’ device of selecting particular actors on which to hang his narrative, but reflects Lippmann’s considerable role in developing that product).
Any serious investigation of CDO bad practices would implicate Deutsche Bank, and presumably, Khuzmami. Why was a Goldman Abacus trade probed, and not deals from Deutsche Bank’s similar CDO program, Start? Khuzami simply can’t afford to dig too deeply in this toxic terrain; questions would correctly be raised as to why Deutsche was not being scrutinized similarly. And recusing himself would be insufficient. Do you really think staffers are sufficiently inattentive of the politics so as to pursue investigations aggressively that might damage the head of their unit?
And notice: this article quotes Khuzami directly, Khuzami, and not a single other source. There is no “reporting” of opposing views, say of attorneys pursuing CDO litigation or law professors who are willing to criticize financial services industry practices like John Coffee at Columbia or, perhaps William Bratton at Penn.
It’s also in keeping that Stewart missed In this case, Deutsche (!) sold an ABS CDO to M&T Bank which promptly tanked.
M&T Bank sued for various breach of contract and fraud claims. The court denied Deutsche’s motions to dismiss on a variety of grounds, including fraudulent misrepresentation. The also granted some specific fraud claims, leaving more than enough for M&T Bank to proceed on.
This is an obvious precedent that Rakoff is almost certainly aware of, even if Stewart is not. If a complaint can survive a motion to dismiss, then it is strong enough to force some real fact finding and compel the defendants to come to a settlement that reflects the real costs and conduct (and that per Rakoff includes meaningful punishments for culpable individuals).
In addition, the appellate division of the Supreme Court of New York by China Development Industrial Bank against Morgan Stanley.
Now let’s turn to Stewart’s brazen and remarkably fact free assertions in defense of the SEC and Citi against Rakoff. Now it is narrowly true that the SEC may well do less well at trial than it would in the settlement. If you force someone into a fight he does not want, he may underperform just to show he was right. And the SEC may want more to make its political point about Rakoff’s ruling than win the case. But any honest reading of the complaint that Rakoff discusses, the one against mid-level Citi employee Brian Stoker, as well as the SEC cease and desist order against Credit Suisse Alternative Capital for its action on this very same Citigroup deal shows that, contra Stewart, the SEC has a strong case. It’s no wonder Rakoff was ripshit.
As an aside, it’s also telling that Stewart didn’t even bother to understand the deal. He says Citi “put together a package of $1 billion of mortgages” when the deal was a CDO squared. That means it was uber dreck, a way for Citi to get rid of lower rated CDO tranches that it could not otherwise dispose of. But getting rid of toxic assets wasn’t good enough for Citi. The deal was part cash bonds, largely unsold lower rated CDO tranches in Citi’s portfolio, and part synthetic (credit default swaps, in this case on A rated subprime mortgage bonds) chosen by Citi so it could go short (more on that below).
Let’s start with the key section of Stewart’s whitewash:
But bad deals, even really bad deals like Citigroup’s, aren’t illegal. They’re not criminal. They’re not inherently fraudulent. If Citigroup’s clients, all of them sophisticated institutional investors, were foolish or careless enough to buy what Citigroup sold them, then arguably they deserved their losses. Their remedy, presumably, would be never to do business with Citigroup again.
This poses a major obstacle for prosecutors and the agency’s enforcers. They have to prove that bank executives misrepresented the terms of the deal and misled investors. Fraud has an even higher standard of proof: the statement must be intentionally false about a material fact….
The offering circular describing the deal states: “the composition of the Eligible Collateral Debt Securities will be determined by the selections of the Manager” which was Credit Suisse. Citigroup suggested mortgages for the collateralized debt obligation, most of which were included (along with others identified by Credit Suisse), but maintains that Credit Suisse “determined” the assets in that it had no obligation to accept Citi’s recommendations and that Citi had no veto power over what Credit Suisse decided to include.
Remember how these deals worked. The key party in the transaction was the asset manager, also known as the collateral manager, who chose the exposures that would go into the CDO. Investors relied on the belief that the asset managers were independent and were looking out for their interest. Thus an asset manager that was presented as independent but was in fact working in cahoots with the seller of assets (and worse, cooperating them in setting up the deal so their short trade would work out) is a HUGE omission. As we described in Chapter 9 and Appendix II of ECONNED, collateral managers of synthetic and hybrid (partially synthetic) CDOs knew full well whose interest they were really serving, namely, the short side, not the nominal investors.
Now let’s look at some sections of the SEC cease and desist order against the asset manager on this deal, Credit Suisse Alternative Capital, for its actions on this very transaction:
From the cease and desist order:
[Samir] Bhatt was the portfolio manager at CSAC primarily responsible for the Class V III transaction. Bhatt was responsible for selecting the assets in accordance with CSAC’s stated processes, as well as for negotiating and executing the purchase of those assets on behalf of Class V III. Bhatt and CSAC understood that Citigroup was seeking to short assets into Class V either for itself or for its customers (though did not necessarily know which), and thus that Citigroup was representing economic incentives potentially adverse to those of Class V III and its investors.
Rather than follow CSAC’s stated asset selection process, Bhatt provided Citigroup with a list of potential assets with which he had some familiarity, and allowed Citigroup to select from the list the names on which it wanted to purchase protection. The CDO securities on which Citigroup bought protection had a notional value of approximately $500 million, representing half of the Class V III investment portfolio. Citigroup’s selections were weighted towards assets that were regarded by the market as particularly risky
And the SEC does have the smoking gun. Here is part of the detail:
At approximately 9:58 AM on January 8, 2007, the Citigroup salesperson responsible for the CSAC account forwarded to Bhatt an email from a Citigroup CDO trader. The Citigroup CDO trader had written, “Here are the names where we would like to buy protection from CSAC,” and had selected 25 names from the Bhatt December 21 List (the “Citigroup January 8 List”). All 25 of the names on the Citigroup January 8 List were mezzanine CDOs, and 24 of the 25 were from the 2006 vintage. Sixteen of the 25 names on the Citigroup January 8 List were also on the Citigroup November 1 List. Five of the nine names from the Citigroup November 1 List that were not on the Citigroup January 8 List were actually on the CSAC December 21 List, but Citigroup did not seek to short those names on January 8.
By approximately 10:57 AM, less than one hour later, CSAC had agreed to include all 25 of the names from the Citigroup January 8 List in the Class V III investment portfolio.
So you gotta love it. The person ing Stewart information was narrowly truthful. CSAC did select the short exposures…a long list from which Citi picked the real dreck it wanted.
Citi and CSAC also colluded on pricing to the detriment of investors (this is something Tom Adams has discussed at length in earlier posts):
The offering circular for Class V III represented that the assets in the portfolio were purchased at “fair market value.” This statement was inaccurate. Rather than seeking market bids, CSAC and Bhatt purchased most of the synthetic assets (i.e. those referenced by the sale of protection via CDS) in two separate portfolio trades with Citigroup. After determining that Citigroup had paid prices well below what was available in the market for individual assets (i.e. Citigroup had purchased protection for lower premium payments than it would have had to pay for the individual assets in a market transaction as of that day) for the first portfolio trade, CSAC and Bhatt nevertheless proceeded with a second portfolio trade with Citigroup. The prices CSAC and Bhatt obtained in that second trade were higher than for the first trade, but well below what was available in the market for individual assets. CSAC and Bhatt did not take meaningful action to verify that CSAC was obtaining market prices in the transactions with Citigroup.
The cease and desist order provides specific information to support its claim that Citigroup was getting falsely low CDS premiums (low premiums = cheap price). This ripped off investors two ways. Not only were they cheated out of a market price, but they also were deprived of market signals. If they realized how much spreads had widened (a sign of rising risk) that alone might well have put them off the deal.
Oh, and in case you doubt whether this was misleading:
The offering circular states in at least six separate locations that the portfolio was “selected” by CSAC, and emphasizes the importance of CSAC’s process for asset selection.
In addition, Citigroup also provided the warehouse line to CSAC, yet another way in which the relationship was less than arm’s length.
Stewart tries running the CDO version of the “rogue trader” defense: if anything was done wrong, it was Stoker acting alone. The SEC presented e-mails to the contrary in its complaint.
From page 10:
On or about November 3, 2006, the senior CDO structurer drafted an engagement letter for CSAC and circulated it internally with the subject line “CSAC CDO squared.”
Later that day, in response to receiving the draft engagement letter, the senior CDO structurer’s immediate supervisor inquired, “Are we doing this?” The structurer responded: “I hope so. This is [Trading Desk Head]’s prop trade (don’t tell CSAC). CSAC agreed to terms even though they don’t get to pick the assets.” The term “prop trade” is shorthand for “proprietary trade,” meaning a trade undertaken for a firm’s own account, rather than on behalf of the firm’s customer(s).
Notice the “prop trade” idea, which was understood from the very outset of the deal, also debunks another defense of the SEC settlement that comes straight from Mr. Something to Hide Khuzami, namely:
Citi will also argue that it disclosed that it was the initial short counterparty and said it might keep the position or sell it on.
What are you gonna believe if you are a judge, a contemporaneous e-mail trail, or Citi’s efforts to claim out of whole cloth that this wasn’t a prop deal?
And Stoker was not a lone actor. The head desk trader is usually at least a director, often a managing director. The separate Stoker complaint describes him as
…a Director in the CDO structuring group at Citigroup from March 2005 through August 2008. Stoker was the principal Citigroup employee responsible for overseeing the structuring of Class VIII and the drafting of the offering memorandum and pitchbook
In the earlier snippet, the “senior CDO structurer’s immediate supervisor” and the head trader are certain to be senior to Stoker. The idea that he cooked up the collusion in this deal all on his own and none of the higher ups had the foggiest idea is patent nonsense (although Citi loved making about how their deals were client-driven, as the FCIC testimony by its CDO business co-head attests). What may have happened is that Stoker was targeted, just as the Fabulous Fabrice Tourre of Goldman was, in the hope he’d roll more senior staffers, but that plan was trumped by the settlement deal (perhaps the plan was to quietly dispose of the Stoker case after Rakoff blessed the settlement, which now is not going to happen)
Let’s step back and review.
The facts in this case are strong. Citi’s behavior is much worse than Goldman’s on the Abacus trade in which the SEC reached a $550 million settlement. Citi was executing a clear and successful plan to short for its own profit at the expense of investors (while Goldman was working with a third party short, John Paulson, at least initially, and lost money on its trade). It hired a bogus manager to act as a cover, selected the majority of the bonds and short positions, and knowingly misrepresented the role of the manager in the offering documents and related communications (one section of the complaint even describes how Citi rescripted CSAC when its pitch was falling flat). The staff is to be commended for the caliber of investigation embodied in these documents.
The e-mails of “senior traders” should be enough to put Citi desk and business unit heads in jail (well, if you assume the Department of Justice believed in prosecuting anyone from A Company You Heard Of), but they go after a mid level structurer on civil charges instead.
Rakoff is right to be outraged that the SEC wants to allow Citi to get away without admitting or denying guilt. The SEC staff has Citi nailed. But the SEC senior management, meaning Khuzami, wants to let Citi get off easy. And with apologists like Stewart in his camp, he’s still trying to turn the public’s eyes away from the predatory and extraordinarily destructive conduct in this arena.