"But the emperor showed them the label sewn inside his worn-out suit. 'Goldman,' they said admiringly. 'They are the finest of weavers. We were wrong to doubt him.'"
-- "Emperor Corzine's Goldman clothes," Financial Times, Nov. 2, 2011
The failure of broker MF Global (OTC: MFGLQ) is a unique event in the annals of American corporate history: To my knowledge, it's the first time a CEO singlehandedly bankrupted his firm through actions that the board of directors was not only knowledgeable of, but had indeed expressly sanctioned. "That takes some talent!" quipped Roderick Hills, a former chairman of the SEC, when we put this to him -- and the deeper you dig, the more surreal things become. We looked at three groups that had an oversight role with regard to MF Global and how they performed in this farce. Only one came out of the review unscathed.
If the opening premise wasn't enough for you, let me add that I've never seen a board with more financial experience -- all of the directors had worked in finance. David Gelber, in particular, had been the chief operating officer of ICAP -- the world's largest inter-dealer broker -- as well as COO of HSBC Global Markets, on top of having held senior trading positions in foreign exchange and derivatives at Citibank and HSBC.
Overseer No. 1: The board
Let's imagine that you're in charge of hiring a senior trader who will be responsible for a new proprietary trading desk that is a key part of your firm's turnaround strategy. You're doing this in the context of extremely challenging markets in the aftermath of the worst financial crisis since the Great Depression.
A headhunter you sometimes work with thinks she's got a great candidate. "In fact," she says, "you'd be lucky to get him."
"There is one thing, though," she continues, "Nothing serious, mind you. It's just that he hasn't worked in finance for over a decade and the last time he held day-to-day trading responsibilities was almost a quarter of a century ago."
Relevant vs. non-relevant
Once you set aside the very impressive title of former senior partner of Goldman Sachs, and focus on specific, relevant information, the proposition above reveals itself in all its glorious absurdity, a choice that could only be the product of collective madness or extreme desperation.
And yet, that's exactly the choice that the MF Global Board of Directors made.
The board hired Jon Corzine for the job and almost immediately gave him the authority to make gazillion-dollar trades. I'm not referring here to the official hiring of Corzine as CEO, but to his unofficial hiring as senior trader when the board allowed him to become directly involved in the firm's proprietary trading in addition to his executive responsibilities. Doing so, they violated a cardinal principle in risk management: Segregation of duties.
Breaking the first rule
Nick Leeson bankrupted Barings Bank in 1995 because he could settle his own trades, but in this case, Jon Corzine didn't exploit the situation to conceal any activity -- on the contrary: The European sovereign debt trades were disclosed to the board, and to analysts, accountants, and shareholders. Instead, it enabled him to "strongly advocate the [European sovereign debt] trading strategy" -- his words -- at the board level.
As CEO, Corzine reported exclusively to the board, so he was outside the established risk management hierarchy. Normally, a trader would need to justify any limit request at multiple levels as it escalates through the organization. Although Corzine did have to go through the charade of having the chief risk officer formally request increased position limits for the European sovereign debt trade, he was effectively negotiating directly with his board. Once that was the case, no amount of risk committees or sophisticated analytics could help.
Deja vu all over again
Are you now ready to step deeper down into the rabbit hole with Jon Corzine? It turns out that he was in an almost identical situation earlier in his career at Goldman Sachs.
As we mentioned in the previous chapter, 1994 was a traumatic year for Goldman Sachs. As the Fed executed a series of interest rate hikes beginning in December 1993, Goldman's fixed income division -- of which Corzine was the co-head -- produced losses totaling hundreds of millions of dollars over a period that stretched from the beginning of the year through the summer. The losses had a devastating effect on morale and partners were gravely concerned as they watched Goldman's losing positions deteriorate month after month the wealth they had accumulated in their capital account. Forty partners left Goldman at the end of 1994 -- a much higher number than the normal attrition rate.
Fool me once
One of the reasons 1994 was so painful was that Corzine did not order the positions reduced quickly enough, and one of the reasons for that was that he had two roles: co-chief financial officer and co-head of fixed income. I can do no better than this former Goldman partner who described the problem (the quote is from William Cohan's superb book, Money and Power: How Goldman Sachs Came to Rule the World):
He [Jon Corzine] wasn't independent. You just really need independent control functions and it is just critical that you have that to run any kind of trading business. You need to have people on the control side and the compliance side to independently mark the books, to go head to head with the traders, to have a totally independent career track. And you need to look at everything in terms of size.
That lesson applies no less to MF Global in 2011 than it did to Goldman Sachs in 1994.
The evidence is all over his bond sheets
Does all this sound too bizarre to be true? We know for certain that Corzine was MF Global's CEO and there is plentiful evidence of his trading role. The Wall Street Journal reported that he used to walk out of management meetings to check the markets and that he himself placed orders on occasion, based on a list of prices left with an assistant. One trader we spoke to said that, in contrast to the former CEO, Corzine "loved to prowl the trading floor every day he was here." On Oct. 25, Corzine stated openly in his last earnings call, "Our [European sovereign debt] positions and the judgment about risk mitigation steps are my personal responsibility and a prime focus of my attention."
Another design flaw in the reporting lines contributed to marginalizing the chief risk officer, and with him, the risk function. According to the most recent 10-K report, "the Chief Risk Officer ('CRO'), who reports to our President and Chief Operating Officer, is delegated certain authorities from the Board." Given the structure, logic would dictate that the CRO at minimum report to the board of directors or a board committee rather than the president and COO. In this case, the president and COO happened to be Bradley Abelow, a former Goldman Sachs partner who was the treasurer and then the chief of staff in the New Jersey statehouse while Corzine was governor.
It's certainly possible to identify other weaknesses within MF Global, but we prefer to repeat the two most important points: First, other than Jon Corzine himself, the board of directors bears the heaviest responsibility in MF Global's failure. Second, the board's gravest mistake -- other than hiring Corzine in the first place -- was to ignore a cardinal rule of risk management: the segregation of duties. Once they allowed Corzine to wear two hats as an executive and a trader, the odds of the firm failing increased exponentially.
Overseer No. 2: Regulators
For now, we'll ignore the role of the regulators in regard to the issue of segregation of funds -- we'll be addressing this later in the series. Instead, we'll focus on regulators' performance in monitoring the firm and its risks.
MF Global is a situation in which I'm impressed by regulators' actions -- and that's not often the case. Back in mid-March, the Securities and Exchange Commission sent a letter to the broker asking them to defend their accounting treatment of repo-to-maturity transactions and explain the impact on their financial ratios (including leverage) for the fiscal year ending on March 31, 2010. John Corzine had come onboard only a week prior to that date and the only RTM transactions the broker had on were backed by U.S. Treasuries. This suggests MF Global may already have been using aggressive accounting in "de-recognizing" (i.e., removing) assets from their balance sheet through that mechanism.
De-recognizing assets for fun and profit
This favorable accounting treatment may have been part of the attraction of these particular transactions for Jon Corzine as an easy way to make some large bets without increasing the stated leverage of the balance sheet. Here's how it works: It's possible -- under certain conditions -- to treat the repo-to-maturity transaction as if you were selling the bonds to someone even though all you're doing is putting up bonds as collateral on a loan. If you retain the market risk, the default risk, and the credit risk (and a few others), you own the bonds -- full stop. Your house is collateral on your mortgage; do you say you've sold the house to the bank and remove it from your personal balance sheet?
To address the very distortion this treatment creates, the Financial Industry Regulatory Authority (FINRA, a self-regulatory body) asked MF Global in June to boost the amount of capital at its U.S. broker-dealer in recognition of its exposure to its European sovereign debt trades (which didn't figure on the broker's balance sheet). Around mid-August, FINRA informed the broker it would have to comply and make a public filing to disclose it.
MF Global disclosed the capital increase in an 8-K filing on Sept. 1. Then The Wall Street Journal picked up the story of the capital increase on Oct. 17, and the death march began.
Ask and you shall receive
Investors and analysts could have asked themselves the same questions that FINRA must have asked itself: What is a repo-to-maturity? What is the size of the positions? What are the risks involved? The answers to all of these questions and more are available in considerable detail in the company's filings, particularly the 10-K report for the year ending on March 31, 2011. In terms of transparency, this was the furthest thing from a rogue trading scandal.
In this case, FINRA could have saved itself the time and effort if MF Global had no possibility of using repos to mask leverage. A change in the rule -- had it already been in place -- might have saved MF Global from itself by constraining its ability to put on proprietary positions in that size. Consider that once you add the $6.3 billion European sovereign debt exposure back to the firm's stated assets, the leverage ratio rises from an already-high 30-to-1 to 35-to-1. Despite AIG (NYSE: AIG ) , Bear Stearns, Fannie Mae, and Lehman Brothers, we continue to see some financial executives who are unwilling (or incapable) to exercise due caution in the use of leverage.
Overseer No. 3: Credit ratings agencies
Ratings agencies aren't regulators, but they certainly have an oversight function due to their official status in the financial landscape. How did the two most influential agencies, Moody's (NYSE: MCO ) and Standard & Poor's, perform? The answer: Not well.
To their credit, Standard & Poor's (a unit of McGraw-Hill (NYSE: MHP ) ) did point out in an April report on brokers that MF Global had a "material exposure" to European sovereign debt; unfortunately, the report said it was "concentrated in higher quality issuers." At least the part about the "material exposure" is right.
Standard & Poor's didn't, however, update its warning in the Industry Report Card of Aug. 31. By that time, MF Global had provided a sensitivity analysis specifically for the European sovereign debt positions which showed that an increase of 10 basis points in bond yields would produce a $10.7 million loss (a basis point is one-hundredth of a percentage point). This was almost identical to the equivalent figure for U.S. government and federal agency obligations ($10.6 million). But Spanish and Italian bond yields are a lot more volatile than U.S. ones. The higher the yield rises, the higher the losses you suffer on your bonds (bond yield and price are inversely related). Year-to-date, the yield on the Italian two-year bond has ranged between 2.3% and 7.9% -- a difference of 565 basis points. In comparison, the U.S. two-year bond yield has stayed within a range of 70 basis points.
Proprietary trading has a bright future behind it
Even the S&P update on Oct. 26 -- which placed MF Global on "CreditWatch with negative implications" -- still stated that the European sovereign debt trades were client-related. The ratings agency still hadn't understood that the trades were entirely proprietary -- a massive bet using the firm's own capital. The report even refers to MF Global's "plans for future proprietary trading activity" -- perhaps they meant funeral plans?
In an email, we asked the analyst who authored the report to explain on what basis he had written that. He referred me to the media relations manager for financial institution ratings who would only state that "we stand by our published statements." Personally, when I find out that I've written things that are demonstrably false, I prefer to retract them.
S&P waited until Oct. 31 -- the day MF Global filed for bankruptcy -- to downgrade MF Global.
As far as Moody's goes, we could find no evidence of any warning or even mention of MF Global's European sovereign debt positions at any time prior to the first downgrade on Oct. 24. By its own admission, Moody's understood the magnitude of the firm's European sovereign debt exposure only in the weeks leading up to the bankruptcy and only after discussions with company executives.
Al Bush, the Moody's analyst who covered MF Global, said in November that MF Global's public disclosure was "ambiguous and continues to be" and stated that "we were surprised to find out that they had a large off-balance-sheet proprietary position in sovereigns." I can certainly understand how an analyst that follows the company might be surprised -- if they had neglected to read the company's annual report. Those comments are a bald-faced attempt at whitewashing. As we described earlier, MF Global's disclosures were clear, comprehensive, and frequent. Bush did not reply to an email asking him how to reconcile his surprise with a set of specific disclosures we referred him to.
In any case, the "ambiguous disclosure" defense doesn't hold water. Even if MF Global's disclosures were ambiguous, the size of the positions alone should have compelled the analyst to address any uncertainty with the management.
In the next chapter, we get to the $1.2 billion in customer funds that seemingly vanished from MF Global. Click here to read all about it.