WE’RE FUCKED, BASICALLY.”1
Just over a year before the financial crisis peaked, on July 11, 2007, Andy Forster, an executive at AIG Financial Products, worried aloud over the phone to Tom Athan, his subordinate. Forster had worried for some time now about the subprime mortgage market—and now those fears were being realized. On July 3 Forster told Athan, “If we actually mark it, we would physically have to take that hit.” In other words: if they accepted what the market deemed the value of their investments to be, AIG Financial Products would, in Wall Street lingo, have to record those lower values on its books, to “mark” the value of its investments down. The losses were—well, there was only one conclusion: they were fucked.
Over the next several months, AIG’s losses mounted. The firm fought with trading partners. Executives grappled with how to value the collapsing positions and fretted about what to tell the top AIG executives and the public. Paul Pelletier and some of his colleagues at the Department of Justice came to believe that during this anxious period, AIG executives committed crimes.
On July 27 Goldman Sachs came calling: the investment bank wanted money from AIG Financial Products. Goldman had been on the other side of AIG FP as its main trading partner. AIG FP had been selling insurance, in the form of derivatives called credit default swaps, on the supposedly safe portions of collateralized debt obligations. Goldman paid AIG a little bit on a regular basis for the insurance, and AIG had an obligation to pay out large amounts if the CDOs collapsed in value. The contracts contained an unusual feature: if the price of the securities dropped, the likelihood that AIG would have to pay rose, and AIG had to give money to Goldman as collateral. If the price moved back up, Goldman would give money back. Now, with the markets collapsing, Goldman wanted AIG to honor the contract. The investment bank did not want a little bit of money; it asked for $1.8 billion—a big number but nothing that would have come close to putting the insurer at risk of collapse.
The implications were clear. Forster told Athan that he was going to tell the big boss, the head of AIG FP, Joseph Cassano, exactly how clear. On August 2 Athan reported to Forster about a “tough call” he had with Goldman. He told Forster that they needed Cassano to “understand the situation 100 percent and let him decide how he wants to proceed.” Athan complained, “This isn’t what I signed up for—where are the big trades, high fives, and celebratory closing dinners you promised?”
If Goldman’s estimate of the value of its positions was right, AIG, the parent company to AIG Financial Products, faced a loss of around $1 billion to $2 billion on its derivatives portfolio. Its top executives did not know about this loss. Indeed, they were telling the public quite the opposite: that all was fine. On August 9, AIG executives held a conference call with investors. The parent company’s chief risk officer played down the tumult in the subprime market, telling investors, “The risk actually undertaken is very modest and remote.” Joe Cassano, who was also on the call, went further, making a comment that would become infamous as AIG imploded: “It is hard for us, without being flippant, to even see a scenario within any kind of realm of reason that would see us losing one dollar in any of those transactions.”
That evening, Forster, who served at Cassano’s right hand, and Athan had a conversation. The call had gone over well with Wall Street; analyst notes afterward had been positive. Athan marveled that most analysts seemed to believe that AIG had high-quality investments and that it could handle its subprime losses.
“Yeah, but it’s not right, right?” Forster replied.
“Well, I know that, but,” Athan responded, “it doesn’t seem like everybody . . . knew it.”
Over the next several weeks, the problems mounted. In August and September, at least four more parties demanded collateral from AIG. In August, Elias Habayeb, the CFO of AIG’s financial services division, to which AIG Financial Products notionally reported, began to raise questions about the potential losses. Cassano, a power unto himself within the behemoth AIG, cut Habayeb out of the discussions, ordering his own chief financial officer not to talk to him about valuations. “They are fish way out of their depth,” he wrote in an email. Cassano, an authoritarian boss, badgered his employees and sought control over information flow. “Why are you talking to Elias?” he yelled at the AIG vice president of accounting policy, Joseph St. Denis. “You don’t work for that fat fuck!” During the third quarter of the year, Cassano took over a project to value the positions, excluding his own CFO.
AIG may not have made public the demands from trading partners, but that didn’t stop those demands from continuing. When the French bank Societe Generale made a request for collateral, Athan emailed Forster on September 6: “Our only option is to delay and then dispute.” St. Denis, who felt bullied by Cassano, tried to resign, worried that the firm hadn’t been clear about the losses. Cassano dismissed it to colleagues as “[a]ll very dramatic.” He refused to accept St. Denis’s resignation and prevailed on him to stay. But a couple of weeks later, Cassano blew up at St. Denis, telling him, “I have deliberately excluded you from the valuation” process “because I was concerned that you would pollute the process.”
By September 11, 2007, AIG had received multiple collateral calls. That day, Forster and Athan discussed the potential for losses in the “catastrophic-type numbers.” They said they needed to “hold off the accountants.” The day after that, Habayeb met with Cassano and Forster. He advised them that they had to take the hit and mark down the portfolio to reflect the losses. Cassano and Forster argued against it. They contended that Wall Street was wrong and they were right about how to value the esoteric securities. Since the securities weren’t trading on an open market, everything had to be done with complex models. Cassano argued that AIG’s positions were special. They were more secure and more difficult to value because the other CDOs were made up of real bonds: cash instruments. But AIGs were derivatives. The differences between the two, he contended, required differing estimates.
Cassano also maintained they differed given that AIG FP wasn’t trading the securities but holding them like an insurance contract. An insurer holds a homeowner’s policy and makes money if the home never has a fire. Same arrangement here. AIG FP only needed to adjust for the chance that the values would collapse (that the fire would wipe out the building), and not add in any liquidity risk (that it would be forced to sell at a time not of its choosing). Anything that is illiquid and hard to sell requires a discount. But Cassano said AIG’s positions didn’t need that discount because they were never going to try to sell them.
Still, the executives kept returning to the incoming truth. On September 19 Athan discussed with another colleague the potential losses, worrying they would be about $5 billion. A few days later, however, on September 26, AIG FP executives and auditors from its accounting firm PricewaterhouseCoopers met to discuss how to value AIG FP’s positions. Cassano played down the Goldman collateral dispute and said that no other calls had been made, which was not true. His lawyer would later tell prosecutors that the executives and accounting discussed Goldman only briefly in the meeting, almost as an afterthought, and that Cassano and his employees emphasized that Financial Products had not finished estimating the values.
On October 3 Athan told a colleague that AIG FP’s losses were “a couple of billion,” saying that Forster and Cassano knew it. That day, Forster made his stress apparent, exclaiming, “Fuck, this is an ugly portfolio!” The model the company was using to value its positions was “hokey” and “too made up,” he said.
In October AIG Financial Products executives met to discuss the problem. They determined the division had a minor loss on the portfolio, on the order of $45 million. That was nothing for a behemoth the size of AIG. The insurer had exposure to $34 billion worth of pre-2006 CDOs at the time. (As of the end of 2007, AIG had credit default swaps of $527 billion, of which $78 billion were written on CDOs.)2
In a conference call with executives and outside accountants on November 1, the accountants learned for the first time that Goldman was now requesting $3 billion and that other firms had made collateral calls. Cassano now reported to the top executives that his group’s estimates of the losses had risen, to somewhere between $45 million and $350 million. The next day, PricewaterhouseCoopers told Cassano that the $45 million figure was too low. He reran the estimates and arrived at $352 million.
By now, banks from all over were demanding money from AIG FP. Forster and Athan worried about the insurer’s liquidity, wondering if it could actually come up with the cash to give its trading partners. On November 6 PricewaterhouseCoopers learned that AIG FP had received collateral calls from Merrill Lynch, JPMorgan Chase, French banks Societe General and Calyon, Swiss bank UBS, and Morgan Stanley. It was a pile-on. In early November AIG Financial Products received $6.5 billion worth of collateral calls, another indication of how many billions in losses AIG faced.
On November 14 Cassano and Forster spoke by phone and discussed how big the write-down was looking. “We’re gonna end up with about—my guess is about a $4 billion write-down, right?” Cassano said.
“Yeah, slightly more, I think, actually,” Forster responded. It depended, they agreed, on what the model spit out.
A few days later, Cassano found out from the top executives that he would have to make a presentation about the credit default swaps portfolio during a December 5 meeting AIG held for investors. On November 17 Forster and another executive discussed the rapidly escalating collateral calls, now estimating that they had reached as much as $16 billion.
Some executives at AIG understood that losses were serious. In an “early warnings and lessons learned” memo to top executives, Kevin McGinn, AIG’s chief credit officer, wrote, “Not all credit events are foreseeable; this housing correction was, in my judgement. And we saw it, perhaps before anyone else and warned the business units accordingly. Some paid heed; others did not. But no one can suggest that they were not told.” Yet the top AIG management still did not know how bad the situation was.
On November 23 AIG transferred $1.55 billion to Goldman as “good faith” collateral. That was it. An acknowledgment of losses, an admission of the truth. Cassano told Habayeb the opposite, that this deal should have no effect on the valuations because there was no “science” to the amount. The number was the product of haggling between the two disagreeing firms.
Meanwhile, Athan had an ongoing battle with Merrill Lynch over its collateral call. He began to lose that fight, too. The bank “called our bluff,” he wrote to Forster. He was going to see “if there is some way to wiggle some right to dispute on a technicality.” The same day, Cassano emailed a report on the collateral calls to top management, stating “all of the counterparties are understanding and working with us in a positive framework.”
On November 29 the top AIG executives held a conference call to plan for the investors’ meeting. All the senior corporate officers, including AIG’s CEO, Martin Sullivan, and its CFO, Steve Bensinger, along with auditors from PwC were on the line. During the call, Cassano tried to explain why his division’s valuations differed so starkly from the much lower valuations Goldman was using as a basis for its collateral call, which now was $3 billion. Cassano explained that if AIG used Goldman’s methodology, AIG’s loss would be $5 billion. The estimate stunned the group. After a moment, Sullivan said that loss would “eliminate” the quarter’s profits. Using AIG’s own methodology, Cassano attempted to reassure the gathered executives, would reduce the loss to $2.5 billion.
Immediately after the conference call, the two senior PwC partners took aside AIG’s CEO and CFO, telling them they were concerned about the company’s lack of understanding about how Cassano’s valuation model worked. The accounting firm warned top officials that the problems could be so serious that it would have to classify them as a “material weakness,” an accounting term meaning that a corporation’s numbers cannot be relied upon. The partners cited AIG’s “managing” of the valuation process—a damning phrase suggesting that the company was gaming the numbers to make the problem seem smaller than it was.
After the call, Cassano scrambled with his lieutenants to adjust how it valued the portfolio. He met a few days later with executives, who came up with a new way to estimate the value of its losses. One technique put the loss at $5 billion. Another pegged it at “only” $1.49 billion.
The amount was still unacceptably high. Then the executives did something that government prosecutors would eventually regard as devastating actions. The AIG FP executives came up with a valuation trick. Changing valuation measures to prettify a loss when a market turns against a position can be, but is not always, criminal. The executives called it the “negative basis” adjustment. “Negative basis,” in the technical jargon, means that the bond bought with cash is less valuable than the bond created through derivatives; in this case, credit default swaps. Why would that be? To buy a bond with cash, the purchaser had to come up with that money. That often meant borrowing. However, a buyer could go “long” a credit default swap with no money down, or at least with little cash. The borrowing raised the cost of owning the cash bond. Therefore, in normal times, the derivative was slightly more valuable by a fraction—mere hundredths of a percent, which are called “basis points.” AIG FP’s bonds were all derivatives and therefore supposedly more valuable. With the new valuation technique, the company could reduce its loss by overlaying a bit of juice, a boost, because they were derivatives. The Financial Products executives took the losses and then reduced them by a percentage they selected themselves. If the losses were a billion, and AIG decided the negative basis was worth 10 percent, the loss would shrink to $900 million.
Just as AIG was fumbling for a way to reduce the estimates of its losses, the market was coming to the opposite conclusion. Many traders were skeptical about the housing market, and betting against it with credit default swaps. Given the demand for the derivatives, the market was flipping. Investors and traders began to treat the derivatives as less valuable than the cash bonds, not more so. Just as AIG was granting itself a “negative basis” adjustment, the credit market was signaling that an opposite “positive basis” adjustment was needed.
Some top executives got suspicious. Habayeb, skeptical for months, bore in. On Saturday, December 1, he emailed Cassano, pounding him with a series of questions. He asked directly if AIG Financial Products’s valuation methodology had changed. Cassano replied that his unit had used various models to estimate the losses and they “have come in within a range of each other.” In fact, they had a wide disparity. One had produced the $1.49 billion loss, while the other had come up with an estimate of $5 billion in losses. Cassano implied that the loss estimates were up to the minute, when they had not taken into account the latest market drops. He elided the specifics of the “negative basis” magic. Prosecutors and government investigators came to regard the exchange as one of the most damning interactions that took place in the days, weeks, and months before AIG’s crisis.
Cassano and Forster understood that the trading partners—Goldman, Merrill, Societe Generale, UBS, Morgan Stanley—posed the problem. As more banks claimed AIG owed them money, their insistence that AIG Financial Products wasn’t facing significant losses became ever more incredible. On December 3 Forster relayed instructions he said came from Cassano: Athan was to withdraw a collateral offer AIG had made to Merrill. Forster explained that the auditors might force them to use it as a basis to value the entire portfolio, creating an immense loss. Athan said he would continue to try to stall.
On December 5 AIG held the investors’ meeting. Martin Sullivan reassured the shareholders and the public that AIG was confident in its derivatives valuation processes. The CEO also said that management understood how much subprime risk the insurer had. On the call, Cassano told investors that the losses in November were in the $500 million to $600 million range and that they stood at about $1.5 billion for the year. Neither Sullivan nor Cassano specified just how many billions in collateral calls AIG had received, though Cassano dismissed the requests as “drive-bys.” An analyst asked a question about how the firm was estimating its position, because by his estimate, using the public indexes, the losses would be more than $5 billion—much closer to the number that AIG had come up with internally using the more accepted techniques, when not using the negative basis adjustment. Cassano replied that the question was “nonsensical.”
Over the next several weeks, Cassano and his group continued scrambling to value its assets—if you believed the account the executives would give later. Forster admitted to Athan in a phone conversation that AIG would be “lucky” to be able to continue to reduce its losses through its negative basis adjustment.
By January 30, 2008, Habayeb had figured out how much trouble they were in. He now understood how much Cassano and his team had used the negative basis technique to reduce its estimates of the losses. He brought it to the attention of a top partner at PricewaterhouseCoopers. The auditing firm stopped permitting AIG FP from using the technique, as Forster and Athan had feared. Habayeb then called CFO Steve Bensinger. AIG’s top executives could no longer pretend. The giant insurer had to reveal the devastation to the markets.
Bensinger told Habayeb to ask Cassano whether he had been aware of how bad the loss was before the December 5 investors’ call and to ask why he had not disclosed it to corporate officers. Habayeb called Cassano. “Who is asking?” Cassano said. When Habayeb told him “Steve,” Cassano said he had to go and hung up.
On February 4 AIG senior management and the outside auditors from PwC met with Cassano and Forster. The two AIG FP executives attempted to justify how they had valued the positions, but convinced no one. Martin Sullivan talked with Cassano on February 8 and suggested he “retire.” A week later, on February 11, AIG came clean to the public: its December 5 disclosures had been erroneous; it had understated its losses by $3.5 billion, and PwC had determined that AIG had “a material weakness” in its internal controls.3 AIG’s stock fell 12 percent. It would continue plummeting.4
• • •
A year and a half later, by the fall of 2009, at Main Justice, Paul Pelletier and Adam Safwat had assembled that narrative of the AIG Financial Products collapse. Government agents and prosecutors had gathered emails, pored over documents, listened to hours and hours of taped calls, and talked to witnesses. Pelletier supervised Safwat, other prosecutors from Main Justice, and the postal inspectors who worked on the investigation. They had been on this case since February 2008, right after AIG had made its shocking confession to the markets. If this was not a white-collar crime, what was? AIG FP executives knew they were facing huge losses, had motive to hide them, took steps to do so, made erroneous statements, and misled investors.
Nevertheless, everyone on the probe always knew it would be a tough criminal case to bring. They had no doubt that Cassano had tried to fool the top executives and the auditors. But he had been artful. He dropped subtle mentions of what his group was doing. He’d been cute in answering questions from top management, which had no idea what he meant. He’d been smart not to put anything in email.
But, in Cassano, the prosecutors felt they had one of the great villains of the crisis to present to a jury. Vanity Fair had labeled him “the man who crashed the world.” Along with Lehman Brothers CEO Richard Fuld and Countrywide CEO Angelo Mozilo, Cassano ranked in the first tier of most-despised figures of the financial crisis. He had ruled the unit through terror and intimidation. He repeatedly pronounced Habayeb’s name incorrectly, a reflection of his haughty disdain, which the prosecutors expected to exploit.
Was it enough? Could they prove Cassano lied to anyone? What about Forster and Athan? Would they flip if indicted? Pelletier believed in the case. So did Safwat and James Tendick, the postal inspector doing the investigation. The government should bring the indictments based on their evidence, present the facts to a jury, and let it decide. According to the US Attorneys’ Manual, if the prosecutors believe in good faith that the evidence shows wrongdoing, they are justified in bringing the case.
But the criminal division’s leadership did not agree. In the spring of 2010, Lanny Breuer and his lieutenants resolved not to bring any case against AIG executives. The consequences of this decision wouldn’t become clear for a while. Each time top government officials had a difficult choice, they did not take the risk and did not bring charges. By shying away, they created a moral hazard, making bankers more likely to take risks the next time. It was also a moral failure.
In the fall of 2008, the government effectively nationalized AIG, investing $182 billion (originally putting in $85 billion and more than doubling it later) and taking a stake in the company of 79.9 percent.5 More than $75 billion of that passed through to AIG’s trading partners—mainly Goldman Sachs and Societe Generale, the banks on the other side of Cassano’s fateful contracts.6 The government conducted a backdoor bailout of the banks that had taken advantage of AIG. Public fury about the bank bailouts peaked in March 2009 when it came out that AIG had paid out $165 million in bonuses to employees.7 After bailouts in the trillions, this relatively small number set off the public.
The US House of Representatives held angry hearings and readied a bill to tax the bonuses at 90 percent.8 New York Attorney General Andrew Cuomo announced an attempt to seize the bonuses. Senator Charles Grassley of Iowa suggested that AIG employees should, like the Japanese, either resign or commit seppuku. President Obama denounced their greed.9
Safwat, Pelletier, and others at Main Justice worked the investigation through 2008, in the waning years of the Bush administration, without interference from a checked-out front office. They interviewed the auditors at PricewaterhouseCoopers as well as third parties such as the Goldman bankers who had tangled with AIG over collateral.
By the fall of 2008, they had a working theory of what the accounting fraud had been. They homed in on the three executives at the Financial Products division—Cassano, Forster, and Athan—believing that they had misled auditors, management, and the public about the true value of the company’s assets. The investigative theory had three prongs:
One, Cassano shared information with only a close circle of executives, while terrorizing his employees to make them stop asking questions. People who needed to know information didn’t get it and were afraid to ask. Two, the executives took steps to avoid making the collateral calls. They did not tell the complete truth to the auditors or top management about the size and seriousness of those demands. Three, when they could no longer do so, they changed their valuation techniques, using the negative basis adjustment without fully disclosing to anyone its impact.
Pelletier’s and Safwat’s problems started when the Obama administration came in. Early in his tenure, in March 2009, Attorney General Eric Holder wanted an update on the investigation into AIG. Holder stayed calm to the point of evincing boredom. Corporate prosecutions didn’t fire his emotions. He kept his feelings to himself. Noting that the insurer had reached both a deferred prosecution agreement and a nonprosecution agreement in recent years with the Justice Department, Holder said, “We ought to be considering a charge” to the corporation if the case materializes.
The head of the fraud section, Steven Tyrrell, spoke up. The taxpayers now owned the majority of the company. An indictment of AIG might shut down the insurer, handing the new owners—American taxpayers—a massive hit. They couldn’t indict AIG. They couldn’t reach another deferred prosecution agreement that saddled taxpayers with the penalties. “Oh,” said Holder, appearing to remember.
Main Justice would, by necessity, have to pursue individual executives, which Pelletier preferred. As he and Safwat pressed forward, the administration resisted. The investigation became a flashpoint within Main Justice; a clash between the old, aggressive cowboy culture personified by Pelletier, and the new, cautious one with Breuer as its figurehead. Breuer disliked Pelletier When Breuer arrived, Pelletier seemed like the mayor of the building. Many staff prosecutors had been inspired by Pelletier. In May 2009, as a young line attorney, Sam Sheldon attended one of the white-collar training sessions that Pelletier ran with a colleague down at the DOJ’s National Advocacy Center, where prosecutors go to learn how to manage investigations and trials, in South Carolina. Pelletier mesmerized Sheldon, who was coming from Laredo, Texas, where assistants handled a “rocket docket”: a furious load of three hundred to four hundred cases a year dealing with crimes related to guns, drugs, and illegal immigration. Pelletier contended that white-collar cases, while obviously more complicated, should be treated with similar urgency and efficiency, harkening back to Larry Thompson’s and Jim Comey’s credo of “real-time” prosecutions. He urged prosecutors not to neglect circumstantial evidence, which has the same standing in court as direct evidence. If prosecutors were 100 percent convinced of their target’s guilt, it would be okay to take the risk of a difficult trial.
Sheldon wanted to get to Main Justice and join Pelletier. But to the Breu Crew, Pelletier was a puzzle they didn’t bother to solve. They saw him as an out-of-control maniac who could blow up a case, taking his and their careers with them. It wasn’t just Pelletier; after Andersen, KPMG, Ted Stevens, and Bear Stearns, they sought to avoid overly zealous prosecutions at all. A victory in court was golden, but a humiliating loss damaged prosecutors more than a win boosted them. It could crush careers, as the KPMG case showed.
Pelletier, however, did not worry about losing if he thought the case was just. It requires an unusual person to lack regard for one’s future viability. Most people husband their credibility both within the bureaucracy, which prizes collegiality and hierarchy, and with the outside professional world, which values success. The same forces push people to reject others who aren’t as politic or ambitious. Pelletier might have been dangerous. He was a rebuke, forcing others to confront their own cowardice. In the new, cautious order, those who are not conscious of their reputations at all times are so unusual that they are dismissed as cranks, their exhortations regarded as rants.
• • •
Lanny Breuer was not the only one at the Department of Justice fearful of prosecuting top executives or corporations. The entire institution, laboring under the legacy of Arthur Andersen and the fight over the Thompson memo, became even charier following the financial crisis. Almost immediately after taking power, the new administration faced a decision about what to do about a major bank. In early 2009, just months after the financial crisis reached its apex, the Obama White House deliberated over serious misconduct by UBS, the Swiss bank. The IRS had sued the bank for names of wealthy American clients suspected of offshore tax evasion. Now the Justice Department was negotiating over a settlement and to get the names. Delivering the names might violate Swiss law, however. The case fell to the new Obama Justice Department officials. Breuer had yet to arrive. Holder was recused, since Covington had represented the bank.
The Swiss government had bailed out UBS the previous October. The bank still teetered. The Department of Justice did not want to make the decision alone. The Treasury Department consulted. “There were policy-level deliberations about whether or not [to indict] and whether whatever penalties UBS faced would impact the fragility of the bank itself, but more importantly, systemic risk,” says a former top Obama official, “as there should be.” It was a “very precarious situation,” this person says. Prosecutors believed they could not indict any of the big global banks. On February 18, 2009, UBS entered into a deferred prosecution agreement, on charges of conspiring to defraud the United States by impeding an IRS investigation.10 While the media covered the UBS negotiations extensively, the Justice Department’s settlement instead of an indictment or a guilty plea generated little criticism.
Three years later, public opinion about the Department of Justice’s soft treatment of the big banks had shifted. The Holder administration had not recognized the change. The department was trying to resolve a six-year investigation into HSBC, a British bank that had flouted money-laundering rules. HSBC had become the preferred bank for Mexican and Colombian drug cartels, had financed banks with links to terrorism, and had conducted transactions with countries under American sanctions, including Iran and Libya. The amounts reached into the trillions.
HSBC executives repeatedly ignored internal warnings of its shortfalls. At one point, Mexican law enforcement officers told the head of HSBC’s Mexico unit that they had a wiretap of a drug cartel leader saying that his bank was the preferred money-laundering destination. The bank did nothing. The cartels even had especially wide boxes to fit into the deposit windows at local Mexican branches to facilitate their cash deposits.11 The HSBC investigation was not about arcane mortgage securities or arguments about how to value assets on its balance sheet. Here was a simple, straightforward case that would be understandable and outrageous to jurors. The staff pushed for criminal charges.12
Holder and Breuer deliberated. By 2012, markets had stabilized, so the worry about disruption from a tough sanction should have receded. As the department worked to resolve the investigation, Breuer sought advice to understand whether the Justice Department could indict a bank of HSBC’s size without collateral consequences to the financial markets and global economy. He surveyed Washington regulators. He flew to London to consult regulators there. He had sleepless nights. It troubled him that no regulator discounted the possibility that markets might be disrupted. Meanwhile, George Osborne, the United Kingdom’s Chancellor of the Exchequer, and the UK banking regulator, the Financial Services Authority, reached out to their counterparts in the States. Osborne wrote to Ben Bernanke, the chairman of the Federal Reserve, and Treasury Secretary Timothy Geithner to warn that charging HSBC criminally could lead to a “global financial disaster.”13
Breuer also canvassed a lawyer who had a direct interest in the outcome: H. Rodgin Cohen, one of the preeminent banking attorneys in the country. Cohen, a partner at Sullivan & Cromwell, was representing HSBC. Breuer pulled Cohen aside during a meeting and asked, “Is any bank too big to indict?”
Cohen, soft-spoken and courtly, replied, “That cannot be the rule.”
The Sullivan & Cromwell partner then outlined the questions he would ask if he were assessing whether to indict: Had the company made serious efforts to fix the problem and prevent it from reoccurring? Had it removed the responsible senior managers or punished them? If so, no indictment should be necessary. As it happened, HSBC had taken those steps. This was from the standard defense counsel playbook. Davis Polk’s Bob Fiske had used it in the Andersen case, Skadden, Arps’s Bob Bennett had done likewise in the KPMG matter, as had countless defense lawyers representing corporations. Emphasize the guilty company’s cooperation, remorse, and remediation, the rule goes.
Breuer believed his department was getting tough on the bank, telling Cohen, “The era of banks no longer getting indicted is over.” Concerned, Sullivan & Cromwell drafted a press release outlining dire consequences for the bank and told prosecutors it would be made public if HSBC was charged criminally. The statement said other financial institutions would no longer do business with HSBC, forcing the bank to leave certain markets. The implications were clear: indict HSBC, and global finance markets will be disrupted—and HSBC would fan the disruption by putting out the press release.
But the era of avoiding indictments was not over. The staff debated it but never recommended bringing charges or asking for a guilty plea, instead opting for a DPA. Holder and top officials went along. In December 2012 the Justice Department entered into a deferred prosecution agreement with HSBC. The bank paid a sum that seemed spectacular to DOJ officials: $1.3 billion in forfeited profits and $665 million in penalties. The bank agreed to management changes, to make large investments in oversight and compliance, and to have a monitor oversee the agreement.14
Not only did the Justice Department not charge HSBC itself, but also it did not charge a single employee. The news was met with public outcry. Department officials explained later that it was too difficult to single out responsible individuals, since the behavior had gone on for so long and had been overseen by so many. They defended the reforms that HSBC had undertaken, but even those were flimsy. By April 2015, the special monitor assigned to oversee them assailed the bank’s progress as “too slow.”15
In testimony in front of the Senate in March 2013, Attorney General Holder acknowledged the problem. “I am concerned that the size of some of these institutions becomes so large that it does become difficult for us to prosecute them when we are hit with indications that if you do prosecute—if you do bring a criminal charge—it will have a negative impact on the national economy, perhaps even the world economy. I think that is a function of the fact that some of these institutions have become too large,” he said, adding, “It has an inhibiting impact on our ability to bring resolutions that I think would be more appropriate.” He had acknowledged the truth. But then, in response to widespread opprobrium, Holder walked back the remarks.
As the AIG Financial Products probe continued throughout early 2009, Paul Pelletier worked on another investigation, into the R. Allen Stanford Ponzi scheme case. Stanford had run a financial services company called Stanford Financial Group that had in actuality been a $7 billion fraud—not quite as big as Bernie Madoff’s but still massive. Pelletier had tried to get the FBI to intervene in August 2008, but he couldn’t persuade any agents. Stanford collapsed on its own early the next year, and the authorities—lamentably, serving as archeologists rather than detectives—swooped in.
As Pelletier worked on the investigation, Obama won the presidency, and his appointees came into the building. Almost right away, the front office worried both about the case and Pelletier. Once, after a briefing on Stanford, Breuer, now the head of the criminal division, demanded, “Now everybody tell me about the admissible evidence.” He regarded Pelletier as a bullshit artist. Gary Grindler, then Breuer’s number two, rode Pelletier. In May 2009 Grindler forwarded a news report of a governmental filing, asking if Pelletier and his boss, Steve Tyrrell, the head of the fraud section, had reviewed it. “We need to be apprised of the positions that are being taken in this matter,” he wrote. Pelletier responded that he’d reviewed it and was “truly confused as to what you need to be apprised of,” adding that the government had not taken any new position. Grindler explained that since it was such a high-profile case, the front office required constant updating. Pelletier pushed back. “Sorry, sorry, sorry,” Grindler replied. “Does this place such a large burden on you?”
Pelletier felt so. He’d never seen a front office dun him with petty requests and nettlesome tasks. They chastised him on small details and raised alarms at any possible hint of negative press coverage. The micromanagement was annoying enough. But the Breu Crew’s behavior raised a worse specter. The Main Justice staff believed that the front office—the political appointees—shouldn’t get involved in cases to that degree of detail; it increased the potential for political meddling.
In June 2009, after weeks of delay, Pelletier urged Breuer and the front office to give a thumbs-up or thumbs-down on the Stanford indictment. Breuer called Tyrrell, Pelletier, and the other prosecutors involved to his office, where his inner circle surrounded him. The new assistant attorney general of the criminal division appeared unhappy to be put in this position so early in his tenure.
“These are my first months on the job, and you guys have me in a tough position,” he said. “You are advocating bringing this case very strongly, and my back is really up against the wall. I am going to remember this. I am not going to forget this.”
Pelletier was barely four feet outside Breuer’s office when he said loudly, “Holy shit!” Breuer cared, he thought, about his image. If something went wrong, Breuer would exact revenge. Pelletier was on notice.
Pelletier would say, “An orangutan could prosecute Stanford.” His crimes were obvious and easy to prove. When Pelletier and a colleague trained prosecutors at the National Advocacy Center, they explained a white-collar conviction is only half the job. Then comes restitution. To help make Stanford’s victims whole, Pelletier wanted to go further: to look at the lawyers, banks, and all the Stanford enablers. He wanted to go after Societe Generale, which had done banking for Stanford. Greg Andres, who replaced Grindler as the number two in the fraud section, told him, “I don’t give a shit about Stanford II,” meaning the follow-on cases. The Main Justice front office passed on investigating the French bank further. The Obama administration Department of Justice was demonstrating its preference for cases against the scammers who made the headlines rather than pursuing the larger, systemic actors who allowed crimes like Stanford’s to flourish. The Enron Task Force, by contrast, had gone after Merrill Lynch and NatWest bankers. Up in the Southern District of New York, Madoff prosecutors took years to go after his enablers at the banks and never charged individual executives from firms that helped him.
Fed up with Pelletier, Breuer pushed him off the case in August 2010. In March 2012 the Department of Justice won at trial against R. Allen Stanford. A jury convicted him of fraud and obstruction charges. A judge sentenced him to 110 years in prison. The government never pursued any enablers of his Ponzi scheme.
The Obama administration’s cautious attitude worked to the advantage of the AIG Financial Products executives. AIG executive Andrew Forster’s lawyers managed to protect many of his materials, saying they were subject to attorney-client privilege. After the withdrawal of the Thompson memo, prosecutors had little ability to fight the assertions. Forster’s attorneys kept at least eight of the executive’s phone conversations, which they said contained “legal advice” about the derivatives contracts and collateral calls. The defense pressed in other ways. The government would have a conversation with the company’s counsel. Then it would hear that the lawyers for the executives knew what had been said. In the old days, prosecutors would have scolded the company’s lawyers and might have prevented those conversations. Now they weren’t allowed to object at all.
In early 2009 Pelletier and Safwat began telling defense attorneys that some executives were “targets,” including Cassano, Forster, and Athan. They particularly thought that Athan might flip. Like many prosecutors, Pelletier and Safwat believed that witness testimony makes white-collar cases. Documents are often technical and dry. To some less aggressive government types, calling Athan a “target” this early in the investigation violated unspoken white-collar mores. In Justice Department parlance, people who testify fall into one of three categories: “witnesses,” who may have observed something prosecutors want to discuss; “subjects,” who may have been involved more directly; and “targets,” who prosecutors believe could be indicted based on the current evidence. Investigations are supposed to proceed slowly. An early threat is viewed as just not cricket. The Breu Crew felt Pelletier and Safwat had been too aggressive.
Tom Athan would prove no pushover. He had a powerhouse legal team protecting him: Debevoise & Plimpton’s Mary Jo White and Andrew Ceresney, the same pair that had represented Bank of America chief executive Ken Lewis. Tyrrell heard that White was complaining about Pelletier, concerned that Main Justice had already come to its conclusions about the case. She succeeded in coloring how the case was viewed within the Justice Department.
Concerned, Rita Glavin, a front office official, sat in an interview (known as a proffer) with Athan. White and Ceresney accompanied him. Pelletier, Safwat, and Tyrrell were outraged. The front office’s micromanagement during Stanford was one thing, but this meddling took it a step beyond. Such active front office participation in investigations was highly unusual.
White and Ceresney laid out their arguments for why Athan was not culpable. Glavin interrupted Safwat and asked questions that seemed to those closest to the investigation as sympathetic to White’s and Ceresney’s positions. In one melodramatic moment, Glavin pulled Tyrrell, Pelletier, and Safwat out of the meeting to chastise them. She found White convincing and exited the Athan interview even more discomfited than she’d entered it. She viewed Pelletier’s and Safwat’s conduct as unprofessional. They were overreaching. In her view, Pelletier and Safwat kept representing their case as stronger than it was. She warned Breuer that the case was weak.
By early 2010, Breuer and Andres and the front office had lost faith in the AIG FP case. The micromanagement hurt the case. The front office received too many updates, seeing the investigative ups and downs. Pelletier understood that every complex investigation took a natural course. The government gathered information, developed theories, and then sometimes discovered new information that made the conclusion less clear. At each new turn, the front office grew more worried. Officials were upset that no one at AIG had yet pointed the finger directly at Cassano. The government had no damning emails from Cassano himself. Perhaps he was too clever. Top AIG officials had been incurious about the state of Financial Products’s business and had let Cassano skate with vague reassurances too often. Sometimes Justice Department officials would mock Safwat and Pelletier for the case’s lack of direct evidence, as if that were the only way to make a case.
By late 2009, the AIG investigation case shifted. Early on, auditors from PricewaterhouseCoopers insisted that Cassano and AIG Financial Products never told them about the negative basis adjustment. But now the government found out that wasn’t exactly true. Investigators found a handwritten note in the margin of a document from the PwC auditor suggesting that Cassano had briefed the accountants on how AIG Financial Products was booking its credit derivative positions. The scribbled note suggested at least one auditor understood something about the negative basis technique. This raised the possibility that Cassano hadn’t been entirely misleading, as auditors had told prosecutors initially. AIG FP’s defense lawyers viewed the note as reasonable doubt in and of itself.
But this piece of evidence was only one elliptical handwritten piece of marginalia. The partner couldn’t say why he wrote it down, who said it, or what the person had meant. Was this truly an investigation killer? People conducting the probe didn’t think so. The note did not change that AIG Financial Products executives had resisted taking losses or reaching collateral deals with counterparty banks in order to avoid losses and then had taken steps to hide their actions and mislead people. James Tendick, the postal inspector who worked the case, found the decision confounding. “We still saw red flags of accounting fraud,” he says. If the office shrunk from complex investigations in the face of one problematic fact, nothing or nobody would be indicted, he believed.
The prosecution team held heated debates about it. Some worried about the evidence, but others favored indictments, with Pelletier as the strongest advocate. But if that wasn’t forthcoming, then he, Tendick, and the agents wanted to continue their investigation—at the least. Sometimes the Department of Justice needed to bring a case even if victory wasn’t assured. The AIG cases were righteous prosecutions. Good things happen when you indict, they argued.
The defense lawyers sensed weakness with the DOJ’s potential case. In a series of meetings, defense lawyers argued that Athan and Forster had certainly struggled with valuing the difficult-to-understand securities but never misled anyone. As Kathy Ruemmler had known when she worked the Enron case for the Justice Department’s task force, more often than not, white-collar suspects truly believe they have done nothing wrong. Their defense lawyers reinforce those beliefs. Cracking them takes time and effort and pressure. If defense attorneys can see that the government is split or sense any vulnerability, flipping becomes almost impossible.
That spring, Breuer and Andres discussed the case with Pelletier, Safwat, and the team several times. The higher-ups probed the prosecutors on the evidence. They asked about their chances of winning. Chances of winning? Pelletier had never been asked that! No other boss had been concerned with trying to estimate the likelihood of victory. “You can always undermine a case in white-collar when you are required to guarantee a win,” he reflects. “If all they do is sit down to hear Athan’s attorneys give the closing argument and conclude the case is weak, the front office is creating an overwhelming burden. What they wanted was a perfect, airtight case. With these kinds of cases, you’re never going to get that.”
Pelletier did not want to provide an answer to Breuer. Juries and judges were unpredictable. Any white-collar case can be lost at trial. This case was even more complicated than usual, and with excellent defense attorneys.
Pelletier started to realize what had dawned on so many prosecutors before him: “These motherfuckers will never authorize the case,” he thought, “and if we somehow get them over that hurdle and we lose, I will be fighting for my job.” Pelletier and Safwat knew what had happened in the Ted Stevens case. He felt the front office had blamed the staffers for their mistakes. “I knew that the end was coming soon for me,” he says. “I knew I couldn’t stay in the department in the kind of high-stakes litigation we were doing; that with the first fuck-up, my career was over anytime. You are trying to make a difference, pushing the envelope; you are just doing your job, bringing hard but righteous cases. Things can go bad in a thousand ways, many of which you could never imagine and most of which you have little control over.”
Pelletier thought he had no choice but to undersell the case. If he didn’t, and they lost, he and Safwat would be dead men at the Justice Department. Pressed, Pelletier answered that they should bring the case. It was a righteous prosecution, but he’d put the chances of winning at 40 percent. That was all Breuer and Andres needed to hear. The Justice Department would abandon the investigation.
It’s possible that Breuer was right. Perhaps no provable crimes had been committed. Alternatively, perhaps Pelletier and Safwat hadn’t set their sights high enough. The Financial Crisis Inquiry Commission, the body Congress set up to examine the causes of the crisis, had the opposite interpretation: that the Justice Department should have gone after AIG’s top executives. The FCIC believed AIG’s top executives had misled the public and referred that case to the Justice Department.16 If Cassano had disclosed the methodologies and the losses, higher management should have been investigated more thoroughly. Pelletier and Safwat didn’t believe they had uncovered strong enough evidence that upper management knew of the problems and covered them up. Andres and Breuer certainly never pushed that line of inquiry. The FCIC made multiple criminal referrals covering a variety of financial firms involved in the 2008 financial crisis. All withered away without any significant attention paid to them at all.
Pelletier took one last desperate gambit. Andres, acting on Breuer’s order, told Pelletier to have Safwat write a memo explaining why they were going to decline to prosecute. Pelletier had Safwat fill the memo with all the details and evidence to compose a prosecution memo in the guise of a declination memo. Pelletier wanted to demonstrate just how much evidence they had in the case and to put it in the record. The front office would have to reckon with all of it before making a final decision to wind up the probe. Safwat wrote what both thought was a knockout twenty-six-page document. Andres took a look at this long document and flipped out. He insisted they cut it way down for Breuer’s benefit. They rewrote it, reducing it to four pages. The Justice Department officially passed on a prosecution.
Pelletier believes today that Cassano should have been indicted. “He conspired to mislead the American public about AIG’s losses. Period. Not a scintilla of doubt.” In May 2010 the Department of Justice dropped the AIG Financial Products investigation.
Passing on one investigation is understandable; passing on every single one starts to speak to something else: a cultural rot. Adam Safwat saw it more clearly than most Justice Department staffers. In addition to investigating AIG Financial Products, his job was to keep track of the financial crisis investigations across the country for Main Justice. Safwat made the rounds to prosecutors across the country: to New York to get an update on the Lehman investigation; to Los Angeles to get up to speed on Countrywide; to Seattle to hear about Washington Mutual. The Feds dropped plenty of subpoenas on banking officials, but that was more or less it.
Clearly, there wasn’t a commitment to digging in. One by one, the financial crisis criminal investigations fell to the decision not to take the risk: Countrywide, Washington Mutual, CDO wrongdoing, mortgage-backed securities transgressions, Lehman Brothers, Citigroup, AIG, valuation games, Bank of America, Merrill Lynch, Morgan Stanley. There were no indictments.
The aftermath of 2008 called for aggressiveness. The people demanded it. The politics were favorable. The Department of Justice had chances to bring cases against companies and executives. Public trials would have presented the evidence. Juries would have decided if crimes had been committed. That this fear of failure took hold when it did is tragic.