Bethany McLean

The Republican candidates for president have some major differences in their policies and their personal lives. But they have one striking thing in common—they all say the federal government is responsible for the financial crisis. Even Newt Gingrich (pilloried for having been a Freddie Mac lobbyist) says: “The fix was put in by the federal government.”

The notion that the federal government, via the Community Reinvestment Act (CRA) and by pushing housing finance giants Fannie Mae and Freddie Mac to meet affordable housing goals, was responsible for the financial crisis has become Republican orthodoxy. This contention got a boost from a recent lawsuit the Securities and Exchange Commission (SEC) filed against six former executives at Fannie and Freddie, including two former CEOs. “Today’s announcement by the SEC proves what I have been saying all along—Fannie Mae and Freddie Mac played a leading role in the 2008 financial collapse that wreaked havoc on the U.S. economy,” said Congressman Scott Garrett, the New Jersey Republican who is chairman of the financial services subcommittee on capital markets and government-sponsored enterprises (GSEs).

But the SEC’s case doesn’t prove anything of the sort, and in fact, the theory that the GSEs are to blame for the crisis has been thoroughly discredited, again and again. The roots of this canard lie in an opposition—one that festered over decades—to the growing power of Fannie Mae, in particular, and its smaller sibling, Freddie Mac. This stance was both right and brave, and was mostly taken by a few Republicans and free-market economists—although even President Clinton’s Treasury Department took on Fannie and Freddie in the late 1990s. The funny thing, though, is that the complaint back then wasn’t that Fannie and Freddie were making housing too affordable. It was that their government-subsidized profits were accruing to private shareholders (correct), that they had far too much leverage (correct), that they posed a risk to taxpayers (correct), and what they did to make housing affordable didn’t justify the massive benefits they got from the government (also correct!). Indeed, in a 2004 book that recommended privatizing Fannie and Freddie, one of its authors, Peter Wallison, wrote, “Study after study has shown that Fannie Mae and Freddie Mac, despite full-throated claims about trillion-dollar commitments and the like, have failed to lead the private market in assisting the development and financing of affordable housing.”

When the bubble burst in the fall of 2008, Republicans immediately pinned the blame on Fannie and Freddie. John McCain, then running for president, called the companies “the match that started this forest fire.” This narrative picked up momentum when Wallison joined forces with Ed Pinto, Fannie’s chief credit officer until the late 1980s. According to Pinto’s research, at the time the market cratered, 27 million loans—half of all U.S. mortgages—were subprime. Of these, Pinto calculated that over 70 percent were touched by Fannie and Freddie—which took on that risk in order to satisfy their government-imposed affordable housing goals—or by some other government agency, or had been made by a large bank that was subject to the CRA. “Thus it is clear where the demand for these deficient mortgages came from,” Wallison wrote in a recent op-ed in The Wall Street Journal, which has enthusiastically pushed this point of view in its editorial section since the crisis erupted.

But Pinto’s numbers don’t hold up. The Financial Crisis Inquiry Commission (FCIC)—Wallison was one of its 10 commissioners— met with Pinto and analyzed his numbers, and concluded that while Fannie and Freddie played a role in the crisis and were deeply problematic institutions, they “were not a primary cause.” (Wallison issued a dissent.) The FCIC argued that Pinto overstated the number of risky loans, and as David Min, the associate director for financial markets policy at the Center for American Progress, has noted, Pinto’s number is far bigger than that of others—the nonpartisan Government Accountability Office estimated that from 2000 to 2007, there were only 14.5 million total nonprime loans originated; by the end of 2009, there were just 4.59 million such loans outstanding.

The disparity stems from the fact that Pinto defines risky loans far more broadly than most experts do. Min points out that the delinquency rates on what Pinto calls subprime are actually closer to prime loans than to real subprime loans. For instance, Pinto assumes that all loans made to people with credit scores below 660 were risky. But Fannie- and Freddie-backed loans in this category performed far better than the loans securitized by Wall Street. Data compiled by the FCIC for a subset of borrowers with scores below 660 shows that by the end of 2008, 6.2 percent of those GSE mortgages were seriously delinquent, versus 28.3 percent of non-GSE securitized mortgages.

To recap: If private-sector loans performed far worse than loans touched by the government, how could the GSEs have led the race to the bottom?

Another problematic aspect to Pinto’s research is that he assumes the GSEs guaranteed risky loans solely to satisfy affordable housing goals. But many of the guaranteed loans didn’t qualify for affordable housing credits. The GSEs did all this business because they were losing market share to Wall Street—their share went from 57 percent in 2003 to 37 percent by 2006. As the housing bubble grew larger, they wanted to recapture their share and boost their profits.

Indeed, the SEC lawsuit specifically says Fannie and Freddie began to do more risky business not to meet their goals, but rather to recapture market share—and they began to do so aggressively in 2006, when the market was already peaking. So while the GSEs played a huge role in blowing the bubble bigger than it otherwise would have been—and the numbers in the SEC complaint are huge—they followed, rather than led, the private market.

It’s also very hard to look at what happened in the crisis and conclude that nothing went wrong in the private sector. Note that the other Republican members of the FCIC refused to sign on to Wallison’s dissent. Instead, they issued their own dissent. “Single-source explanations,” they said, were “too simplistic.”

Yet despite all that, the one-note Republican refrain hasn’t changed. The explanation is obvious: The “government sucks” rant polls well with conservatives. Mix in an urge to counter the equally simplistic story from the left—that the crisis was entirely the fault of greedy, unscrupulous bankers—and you get a strong resistance to the facts. Maybe there’s a deeper reason, too. For many, belief in the all-knowing market was (and is) almost a religion. This financial crisis challenged that faith by showing the market would indeed allow loans to be made that could never be paid back, and by showing that highly paid financial services executives aren’t gods, and that many of them are stupid and venal and all too human.

So maybe the Republican orthodoxy is understandable, but that doesn’t mean it isn’t scary. Of course, there’s the great line from Edmund Burke: “Those who do not know history are destined to repeat it.” Our housing market is a mess that threatens to drag down the entire economy, and whoever is president in 2013 needs to have a plan. Denying the facts is not a good start.

PHOTO: Republican presidential candidates (L-R) former U.S. Senator Rick Santorum (R-PA), former Massachusetts Governor Mitt Romney, former Speaker of the House Newt Gingrich and U.S Representative Ron Paul (R-TX) arrive on stage before the Republican presidential candidates debate in Tampa, Florida January 23, 2012.  REUTERS/Brian Snyder

Juries are sometimes told that in the eyes of the law, all Americans are created equal. But if that’s the case, then why does the Securities and Exchange Commission’s treatment of former top Fannie and Freddie executives seem to be so much harsher than its treatment of Citigroup and its senior people for what appear to be similar infractions?

Recall that on Dec. 16, the SEC charged six former executives at mortgage giants Fannie Mae and Freddie Mac with fraud for not properly disclosing the companies’ exposure to risky mortgages. In Fannie’s case, the SEC alleges that former CEO Dan Mudd and two other executives made a series of “materially false and misleading public disclosures.” The SEC says, for instance, that at the end of 2006, Fannie didn’t include $43.3 billion of so-called expanded approval mortgages in its subprime exposure and $201 billion of mortgages with reduced documentation in its Alt-A exposure. In Freddie’s case, the SEC alleges that while former CEO Dick Syron and two other executives told investors it had “basically no” subprime exposure, they weren’t including $141 billion in loans (as of the end of 2006) that they internally described as “subprime” or “subprime like.”

There are some gray areas in the SEC’s case. Start with the fact that there is no single definition of what constitutes a subprime or Alt-A loan, or as Mudd said in a speech in the fall of 2007, “the vague, prosaic titles that pass for market data — ‘subprime,’ ‘Alt A,’ ‘A minus’ — mean different things to the beholders.” In Fannie’s 2006 10(k), Mudd noted that apart from what Fannie was defining as subprime or Alt-A, the company also had “certain products and loan attributes [that] are often associated with a greater degree of credit risk,” like loans with low FICO scores or high loan-to-value ratios. And in a letter to shareholders in 2006, Mudd noted that “to provide an alternative to risky subprime products, we have purchased or guaranteed more than $53 billion in Fannie Mae loan products with low down payments, flexible amortization schedules, and other features.” These are the very holdings that the SEC says should have been disclosed as subprime exposure.

The SEC’s case on the Alt-A mortgages is equally tricky. As blogger David Fiderer points out, “reduced documentation” can refer to a loan for which the borrower proves his income with a W-2 for last year in addition to a 1040 for the previous year. Or “reduced documentation” can refer to a “stated income/stated asset” loan, otherwise known as a liar loan.” As of September 2008, the default rate on the loans that the SEC says Fannie should have labeled Alt-A was less than a third of the default rate on those Fannie did call Alt-A. In other words, you could argue that if Fannie had mixed those two groups of loans together, thereby lowering the reported default rate, the company could have been criticized — or possibly even sued — for making the default rate on its Alt-A loans look artificially low.

Be that as it may, the SEC, which has been under fire for not being aggressive enough, brought its charges. “Fannie Mae and Freddie Mac executives told the world that their subprime exposure was substantially smaller than it really was,” said Robert Khuzami, director of the SEC’s Enforcement Division, in the press release. “These material misstatements occurred during a time of acute investor interest in financial institutions’ exposure to subprime loans … all individuals, regardless of their rank or position, will be held accountable for perpetuating half-truths or misrepresentations about matters materially important to the interest of our country’s investors.”

The companies themselves entered into non-prosecution agreements and agreed to cooperate with the SEC’s litigation against the former executives. (The SEC said it “considered the unique circumstances presented by the companies’ current status,” meaning that because the companies were wards of the state following the government takeover in the fall of 2008, it wasn’t in taxpayers’ best interests to go after them.) The SEC didn’t charge any of the more junior employees who were presumably involved in the actual drafting of the disclosures. But the SEC is throwing the book at the former executives, charging them with securities fraud — its most serious charge — and seeking financial penalties plus a bar on any of the six serving as an officer or director of a publicly traded company. It’s one of the harshest treatments of big corporate executives in the aftermath of the financial crisis.

Just weeks earlier, a judge in New York had thrown a wrench in another SEC proceeding. Judge Jed Rakoff refused to approve a settlement in which Citigroup (or rather, its shareholders) would have paid $285 million to settle SEC charges that the bank misled investors when it sold a $1 billion subprime CDO, made itself a profit of about $160 million and cost investors about $700 million. The charge involved only negligence, not fraud. In his rejection of the settlement, Rakoff wrote: “If the allegations of the Complaint are true, then Citigroup is getting a very good deal; even if they are untrue, it is a mild and modest cost of doing business.” Separately, the SEC filed a complaint against a former Citigroup employee named Brian Stoker, who the SEC describes as “primarily responsible” for the deal’s marketing documents. Stoker was charged with fraud — but Stoker was a relatively junior guy.

That attempted settlement, though, isn’t the only deal the SEC has cut with Citigroup. In 2010, Citigroup’s shareholders paid $75 million to settle charges that Citigroup — get this — misled investors over its exposure to risky mortgages. According to the SEC, Citi repeatedly told investors that its exposure was only $13 billion and declining, when in reality it was more than $50 billion and not declining. Citi neither admitted nor denied the charges, as was the standard convenience afforded to companies until Judge Rakoff began to derail the settlement machine.

There is probably gray in the Citi complaint as well, but at least on the surface, it’s harder to find. The allegedly undisclosed exposure consisted of “super senior” tranches of subprime mortgage-backed securities (the stuff that was supposed to be really safe, but wasn’t) and “liquidity puts” (which required Citi itself to buy outstanding commercial paper that was backed by subprime mortgage-backed securities if the paper couldn’t be sold to investors). According to the SEC’s complaints, Citi executives didn’t feel they had to disclose these exposures because they felt the risk of default was low. Yet by the fall of 2007, Citi had had to buy “substantially all” of the $25 billion in commercial paper and had started taking losses on the super-seniors. Even then, Citi still didn’t disclose its exposure, but rather bundled these losses in with others. According to the SEC, “senior management was aware” that the super-senior tranches were a source of the losses, but the company “nevertheless continued to exclude” the additional exposure. Only on Nov. 4, 2007, did Citi issue a press release in which it acknowledged the existence of this junk — and the after-tax losses of $5 billion to $7 billion it would have to take as a result of all of its subprime exposure.

In its own press release announcing the settlement, the SEC summed it up this way: “Even as late as fall 2007, as the mortgage market was rapidly deteriorating, Citigroup boasted of superior risk management skills in reducing its subprime exposure to approximately $13 billion. In fact, billions more in CDO and other subprime exposure sat on its books undisclosed to investors,” said Khuzami. “The rules of financial disclosure are simple — if you choose to speak, speak in full and not in half-truths.” Added Scott Friestad, the associate director of the SEC’s Enforcement Division: “Citigroup’s improper disclosures came at a critical time when investors were clamoring for details about Wall Street firms’ exposure to subprime securities.”

In addition to the fine paid by Citi’s shareholders, the SEC gave two executives, Citi’s former CFO, Gary Crittenden, and its former head of investor relations, Arthur Tildesley, light slaps on the wrist for their roles in causing Citigroup to make the misleading statements. Crittenden and Tildesley agreed to “cease and desist” from anymore violations of the securities laws, and they paid $100,000 and $80,000, respectively. That’s a far cry from being charged with fraud and being barred from serving as an officer or director of a publicly traded company. Neither man admitted or denied the charges. Tildesley continued to work at Citi.

The SEC doesn’t comment on enforcement cases. But it might be worth noting that the Citi charges came at the end of a settlement process, whereas the charges against the Fannie and Freddie executives might be just the beginning of a process that could end in something far less extreme than fraud charges. Nor is the SEC a monolithic agency, meaning that there may not be any coordination between the teams in charge of each case. And there could, of course, be differences between the cases that can’t be gleaned from the surface of the complaints.

Still, given the leniency seemingly accorded Citigroup’s people, it’s hardly surprising that some would charge that Citi received special treatment. In fact, the SEC’s Office of the Inspector General investigated an anonymous complaint that there were “serious problems with special access and preferential treatment” in the Citi affair. While the OIG did not find that that was the case, its redacted report still offers insight into the vagaries of the SEC’s process.

Most notably, Tildesley had agreed to settle to a lesser charge of fraud than that faced by the former Fannie and Freddie executives — but it was still fraud. Crittenden, however, wouldn’t settle. After SEC officials had meetings not just with Crittenden and Citi’s lawyers but even one in which Dick Parsons, Citigroup’s then-chairman, “made a personal pitch,” the SEC agreed to the slap on the wrist for Crittenden. So Tildesley’s settlement was changed to match Crittenden’s. The SEC enforcement staff had notified other individuals that it planned to recommend charges against them, the report said, but their identities are blacked out, and the non-redacted text doesn’t explain why those individuals weren’t charged. But an unidentified SEC official testified that Citigroup was “trying to use whatever leverage they had … to get us to … lay off the individuals.”

These days, Fannie and Freddie, far from having any “leverage,” are political punching bags. And they deserve to be punched: The companies were disasters. In the wake of a crisis the magnitude of the one we experienced, it’s also probably better that the SEC is too aggressive, rather than not aggressive enough. But above all, the SEC has a duty to be fair. Which means making sure that there can’t even be the perception that a company’s “leverage” influences the treatment accorded its employees.

By Bethany McLean
The opinions expressed are her own.

There were a lot of things that were supposed to save Europe from potential financial Armageddon. Chief among them is the EFSF, or European Financial Stability Facility.

In the spring of 2010, European finance ministers announced the facility’s formation with great fanfare. In its inaugural report, Standard & Poor’s described the EFSF as the “cornerstone of the EU’s strategy to restore financial stability to the euro zone  sovereign debt market.”  The facility itself said in an October 2011 date presentation that its mission is to “safeguard financial stability in Europe.”

That of course hasn’t happened. And the evidence suggests that the EFSF may have only exacerbated the problems.

In theory, the facility is supposed to provide a way for a country that the market perceives as weak to still borrow money on good terms. The initial idea was that instead of the financially troubled country itself trying to sell its debt to live another day, the EFSF would be the one to raise the money and lend it to the country in question. The logic was simple: country X might be shaky, but the EFSF deserved a triple-A rating.

For all of its would-be financial firepower, the EFSF isn’t much to see—it’s just an office in Luxembourg with a German-born economist CEO named Klaus Regling, who oversees a staff of about 20. Its power—and that rating—is derived from the assumption that any debt it issues is guaranteed by the members of the euro zone. Initially, each member pledged unconditionally to repay up to 120% of its share of any debt the EFSF issued. (A country’s share is determined by the amount of capital it has in the European Central Bank.)

On paper, it all sounded great. The reality is that the EFSF wasn’t meant to be an active institution; it was supposed to be a fire extinguisher behind glass: never to be used. “The EFSF has been designed to bolster investor confidence and thus contain financing costs for euro zone member states,” wrote Standard & Poors in its initial report granting the triple A rating. “ If its establishment achieves this aim, we would not expect EFSF to issue a bond itself.”  Moody’s, for its part, wrote that the EFSF “reflects the political commitment of the euro zone member states to the preservation of the euro and the European Monetary Union.” That show of commitment alone was supposed to be enough to reassure the market.

In granting the EFSF the all-important triple-A rating, the rating agencies were somewhat cautious. They weren’t willing to assume, as they did with subprime mortgages, that any subset of debt guarantees—no matter how small—made by countries that weren’t themselves triple-A rated would be worthy of the gold-plated standard. Instead, they insisted that the EFSF’s loans had to be covered by guarantees from triple-A rated countries and cash reserves that the EFSF would deduct from any money it raised before it passed those proceeds on to the borrowing country. Based on that, euro zone members initially pledged a total of 440 billion euros ($650 billion) in guarantees. However, S&P said in its initial report that the EFSF would be able to raise less than $350 billion of triple-A rated proceeds.

Of course the euro zone did break the glass: the fire extinguisher was used, first in support of Ireland, and then Portugal, and then Greece. This summer, the European Powers That Be agreed to bolster the EFSF’s lending capacity by increasing the maximum guarantee commitments of the member states to 165%, instead of 120%. That was supposed to enable the EFSF to borrow up to 452 billion euros “without putting downward pressure on its ratings,” according to S&P.

As we now know, that’s not nearly enough money to end the crisis, especially given that the EFSF’s commitments to Ireland, Greece and Portugal leave the facility with a lending capacity of just 266 billion euros, according to a recent report from Moody’s. (I found it difficult to add up where the money went.)

Hence the politicians’ latest idea: leverage the EFSF’s remaining ability to borrow. Either have the EFSF offer first-loss insurance when a country issues debt, or turn its remaining capacity into the first-loss tranche of a collateralized debt obligation, which would raise money by selling bonds to other countries like China.

Besides being undersized for the job, there’s a core structural problem with the EFSF: It is only as strong as its triple-A members—not just Germany, which according to that October 2011 EFSF presentation contributes 29% of the total value of the EFSF’s guarantees, but also France, which contributes 22%, and the Netherlands, which contributes 6%.

Some financial analysts have questioned why any European country deserves a triple-A rating, seeing as EU members can’t print their own currency to pay off their debts.  As the financial turmoil has unfolded, it’s become clear that the only EU country the market views as a bona-fide triple-A is Germany. So as much of Europe crumbles, the EFSF’s triple-A, in the eyes of the market, is supported by a lone country. As one bond market participant says, “Eventually, only the German guarantee will matter, and it isn’t big enough to cover this.”

Indeed, Germany represents less than a quarter of the EU’s GDP, and obviously the nation’s economic health is to no small degree dependent on other members of the EU buying German goods. (Such interconnectedness was the whole point of the European Union—and that helps explain why the cost to insure against a German default has more than doubled since the summer, to $93,655 a year to insure $10 million of 5-year German debt.)

Not surprisingly, the EFSF’s last 3 billion euro bond sale, on Monday, November 7, met with what Moody’s called “significantly less demand” than a similar issue last spring. The issue was originally supposed to be 5 billion euros, and the spread, relative to German debt, that was required to lure investors was over three times the spread that was needed last spring. As Moody’s wrote in its report, “the demand for the EFSF bond issuance was dampened by a lack of confidence over the credit resilience of its guarantors.” Another way to think about this is that since the strength of the EFSF is dependent on the strength of its parts, the more individual countries have to pay to raise money, the more the EFSF itself has to pay.

In fact, it’s the definition of a vicious cycle. The EFSF’s funding costs rise along with those of its guarantors, and perversely, bond market participants say that the very existence of the facility also causes its guarantors’ cost to rise. That’s because there aren’t many investors who are interested in buying European sovereign debt these days. Those who are demand a very high premium if they’re going to buy, say, Italian debt instead of EFSF debt. This is why the EFSF was going to be hurtful, rather than helpful, if it had to be used: it competes with its very creators for investment, driving spreads higher and higher. One hedge fund manager calls the EFSF  a “self-inflicted killer” of Europe’s bond markets.”

The EFSF is due to expire, and is supposed to be replaced by the European Stability Mechanism, or ESM, in mid-2013.  But the ESM looks like it’s going to have same problem the EFSF does: Its finances depend on the very same countries that it is supposed to bail out.  In other parts of the world, this isn’t called stability; this is called a Ponzi scheme.

Photo: Men climbs steps next to a share price ticker at the London Stock Exchange in the City of London. REUTERS/Andrew Winning

By Bethany McLean
The opinions expressed are her own.

On Monday morning, MF Global, the global brokerage for commodities and derivatives, filed for bankruptcy.  The firm’s roots go back over two centuries,  but in less than two years under CEO Jon Corzine, whose stellar resume includes serving as the chairman of Goldman Sachs, as New Jersey’s U.S. Senator, and as New Jersey’s governor, MF Global collapsed, after buying an enormous amount of European sovereign debt. The instant wisdom is that he made a big bet as part of his plan to transform MF Global into a firm like Goldman Sachs, which executes trades on behalf of its clients, and also puts its own money at stake. Although the size of the wager has received a great deal of scrutiny, the accounting and the disclosure surrounding it have not–and may have played a role in the firm’s demise.

In the 24 hours since the filing, more ugly questions have piled up, with the New York Times reporting that hundreds of millions of dollars of customer money have gone missing, and the AP saying that a federal official says that MF Global has admitted to using clients’ money as its problems mounted. Whether this was intentional or sloppy remains to be seen; MF Global didn’t respond to a request for comment by press time.

At the root of MF Global’s current predicament was a simple problem:  the profits in its core business had declined rapidly.  That core business was straightforward, even pedestrian; what the firm calls in filings a “significant portion” of total revenue came from the interest it generated by investing the cash clients had in their accounts in higher yielding assets and capturing the spread between that return and what was paid out to clients. As interest rates declined sharply in recent years, so did MF Global’s net interest income, from $1.8 billion in its fiscal 2007 second quarter to just $113 million four years later. MF Global’s stock, which sold for over $30 a share in late 2007, couldn’t climb above $10 by 2009.

Enter Corzine in the spring of 2010, who had just lost his job as New Jersey’s governor to Chris Christie. He was brought in by his old pal and former Goldman partner Chris Flowers, whose firm had invested in MF Global. Fairly quickly, Corzine accumulated a massive net long sovereign debt position that eventually totaled $6.3 billion, or five times the company’s tangible common equity as of the end of its fiscal second quarter. I’m told Corzine’s move was highly controversial within the firm.  But no one overruled him, maybe because after all, he was Jon Corzine. In a mark of just how much Corzine mattered to the market, in early August, MF Global filed a preliminary prospectus for a bond deal, in which the firm promised to pay investors an extra 1% if Corzine was appointed to a “federal position by the President of the United States” and left MF Global.

Buying European sovereign debt may not have been just a bet that the bonds of Italy, Spain, Belgium, Portugal and Ireland would prove attractive. An additional allure may have been the way MF Global paid for the purchases, and thereby, the way the accounting worked. MF Global financed these purchases, as its filings note, using something called “repo-to-maturity.” That means the bonds themselves were used as the collateral for a loan, and MF Global earned the spread between the rate on the bonds, and the rate it paid its repo counterparty, presumably another Wall Street firm. The bonds matured on the same day the financing did.

The key part is that for accounting purposes, MF Global’s filings say the transaction was treated as a sale. That means the assets and liabilities were moved off MF Global’s balance sheet, even though MF Global still bore the risk that the issuer would default; that means the exposure to sovereign debt was not included in MF Global’s calculation of value-at-risk, according to its filings. And that also means MF Global recognized a gain (or loss) on the transaction at the time of the sale. The filings do not say how much of the gain was recognized upfront.  But if it were a substantial portion, then these transactions would have frontloaded the firm’s earnings.  That, in turn, may have helped cover the fact that MF Global’s core business was struggling.

MF Global’s public filings also don’t say how much this contributed to earnings. But one indication of the size of the repo-to-maturity deals comes in this small excerpt from MF Global’s most recent 10K, under the heading of “Off balance sheet arrangements and risk”:  “At March 31, 2011, securities purchased under agreements to resell and securities sold under agreements to repurchase of $1,495.7 million and $14,520.3 million, respectively, at contract value, were de-recognized.” (“De-recognized” means moved off the balance sheet.) Of that $14.5 billion, 52.6% was collateralized with sovereign debt. One way to get a sense of the ramp-up of “repo to maturity” transactions is to compare the figures to those as of March 31, 2010:  The securities sold under agreements to repurchase increased by some $9 billion.

Once the regulators and rating agencies began to zero in on all of this, it didn’t matter that the trade itself may not have been that risky. (The debt all matured by the end of 2012, and MF Global, of course, had financing in place until it matured.) But it was European sovereign debt, after all, and the trade was huge—and it appears that part of the concern may have been the accounting, and certainly the lack of disclosure. On September 1, MF Global said in a filing that the Financial Industry Regulatory Authority (FINRA) was requiring it to “modify its capital treatment” of the European sovereign debt trades.  According to an affidavit filed by MF Global’s president on the day of the bankruptcy, FINRA was “dissatisfied” with the September filing and  “demanded” that MF Global announce that it “held a long position of $6.3 billion in a short-duration European sovereign portfolio financed to maturity.” Words like “dissatisfied” and “demanded” aren’t good in the context of a regulator!

By the time the market opened on Monday, October 24^th, MF Global’s stock had already fallen 62% from its high of almost $10 following the announcement that Corzine was joining the firm. Then, Moody’s downgraded the firm’s debt, citing MF Global’s “inability to generate $200 million to $300 million in annual pretax earnings and keep its leverage within acceptable range.” In other words, Moody’s was concerned about the real profitability of the business. The next day, MF Global reported its $6.3 billion position, per FINRA’s demand, and also reported that it had lost almost $200 million in the quarter ended in September—in large part because the firm had reduced its deferred tax assets by $119.4 million, a sign that the accountants were saying there wouldn’t be a return to big profits any time soon. By the end of the week, all three rating agencies had downgraded MF Global debt to junk. Moody’s wrote that its downgrade “reflects our view that MF Global’s weak core profitability contributed to it taking on substantial risk in the form of its exposure to European sovereign debt.” MF Global’s stock finished the week down 67%.

The actual details of the run aren’t clear yet, but according to the CFO’s affidavit, the ratings downgrades “sparked an increase in margin calls,” which drained cash. Plans to sell all or part of the business fell through, reportedly because of the discovery of the missing cash. Another part of the explanation might be that potential buyers found out just how weak the core business was.

Of course, if Corzine made the trades for an accounting play, there’s a deeper question of why he would feel the need to do this. And isn’t that always the question in situations like this?

PHOTO: Jon Corzine, MF Global Holdings Ltd. CEO, leaves the office complex where MF Global Holdings Ltd have an office on 52nd Street in midtown Manhattan, October 31, 2011. REUTERS/Brendan McDermid