It was, similarly, beyond the comprehension of Obama's Wall Street patrons, many of whom had gathered at a restaurant in Denver the night before Obama's acceptance speech at the Democratic convention on August 28. There were, at that point, rumors about the possibility of the government having to take over its huge, public-private mortgage guarantors, Fannie Mae and Freddie Mac. What's the likelihood? someone asked. They went around the large, twenty-seat table. The vote? Fewer than half thought it could happen. Nine days later, on Sunday, September 7, it did, with Paulson stressing this was a special event, due to the federal guarantee that was still the defining feature of the mortgage giants. Similarly, the Congressional Budget Office had estimated in August that a government bailout of the mortgage giants could amount to $25 billion. To start the fateful second week of September, a CBO spokesman said that would be "optimistic."

Greg Fleming, of course, read the New York Times headline, too. He'd been secretly planning for this eventuality-that if Lehman went, Merrill would be next-since the late spring.

He had been busy selling off parts of Merrill to strengthen the firm's balance sheet. First was Merrill's sizable stake in Bloomberg, which the financial information and media conglomerate bought back for $6.8 billion. Next were nearly $30 billion in troubled mortgage-related a.s.sets that were sold for just $7 billion, a hit that lightened Merrill's toxic load, but pressured other firms, including Lehman, to further write down their a.s.sets. With Merrill's balance sheet now as good as it could possibly be, he turned his attention to the only potential suitor.

He had to get to Bank of America to make an overture first, but he couldn't involve Merrill CEO John Thain, who had made it clear he wasn't interested in selling the company that he'd just been brought in to lead. Ed Herlihy, Wachtell Lipton's top mergers and acquisitions lawyer, had worked with Fleming on various deals and had also been JPMorgan's lead counsel on the government-a.s.sisted purchase of Bear Stearns. Fleming decided that Herlihy, a close friend, could act as a discreet facilitator. Herlihy, who was also Bank of America's lead counsel, called Greg Curl, the bank's number two, and set up a dinner in New York for Fleming and Curl in late July.

Now the question was what to do next. Fleming went back and forth. To meet with a senior player at Bank of America to discuss a sale of his firm, without notifying his own CEO, was a fireable offense. On the other hand, if he didn't move now, Merrill might find itself in the abyss. Finally, Fleming could take it no more. The day of the dinner, he phoned Curl, also an old friend, and told him it wasn't going to work, that they couldn't meet, that he'd be crossing too many ethical lines. But before they hung up, Fleming deftly suggested that Bank of America was the only fit for Merrill; Curl said, from his side, that Bank of America would be interested in looking at such a deal. Message pa.s.sed.

Now, sitting in his office as Lehman began to teeter, Fleming ran the endgame calculus: if Lehman went first, Paulson-no matter how often he said there'd be no government bailout-would be in a "too big to fail" nutcracker. He'd offer Lewis and Curl anything they demanded to buy Lehman, a "Jamie Deal" plus. Government cash would go to facilitate that deal, while Merrill would be left without a suitor or government support. Panic began to set in.

By the next morning, Thursday, September 11, the financial markets around the globe were like an overweight man worrying about the tingling in his left arm and the tightness in his chest.

Hank Paulson got on the phone with Ken Lewis at Bank of America. He and Lewis had already chatted several times that week, as Paulson, talking the talk of the mergers and acquisitions banker he once was, tried to convince Lewis that Lehman was a good buy and the right fit for Bank of America. Lewis, despite his long-standing desires to buy a Wall Street bank, was coy.

Paulson, of course, was no longer Goldman's M&A banker in chief. He was the top domestic appointee of the United States, in an ornate office with a desk used by Alexander Hamilton. His problems in properly defining his role were, more broadly, dilemmas shared by the wider U.S. government. Over the past three decades, it had acted as partner and booster of the profits of large corporations, especially on Wall Street. Were the profits themselves the goal, or was there a large public purpose to government's engagement?

That question was now being posed. Neither Paulson nor President Bush provided an answer. Paulson's limited response since the spring was simply that Bear Stearns was a special event that would not be repeated. Even as the fears of financial contagion began to grip global markets since Fannie and Freddie's fall on Sunday, Paulson stuck to his script. The Treasury would not again be opening the government's accounts to help a private inst.i.tution. A "market-based solution," he said, was the only path. They were on their own.

Ken Lewis, representing the only domestic hope for Lehman, didn't buy it.

He said that Bank of America would not help Lehman unless they received government a.s.sistance similar to the JPMorganBear Stearns deal. Paulson appealed to what he hoped Lewis would see as a wider, even enlightened, self-interest: if Lehman fell, it would surely be bad for the financial sector at large, of which Bank of America, the nation's largest bank, was a flagship.

Not his problem. In fact, in a shakeout, Bank of America might pick up some deals. Lewis reiterated his desire for a "Jamie Deal," stressing that buying Lehman meant a.s.suming tens of billions in troubled a.s.sets. If Paulson wanted Lewis to act in the "greater good," as a bulwark against "systemic risk," then the government would have pay for it, period. After all, Dimon didn't buy out Bear Stearns because of some wider interest in supporting his industry. He did it because of the Fed's role in guaranteeing $30 billion in toxic a.s.sets. Paulson said that doing that sort of a thing again would put the government on a slippery slope, that the markets had to remain sacrosanct as a basic principle, and that it wasn't going to happen. He and Lewis agreed to talk again, but as they clicked off, there was a chill on the phone lines.

What Paulson didn't realize: he'd been beaten to the punch by Fleming.

In a panic, Paulson and his Treasury team shifted their focus to the London-based Barclays, which was eager to widen its Wall Street footprint and had desires for Lehman. Paulson, Geithner, and Bernanke all huddled on a conference call to discuss a known hurdle: British banking regulators would have to approve such a deal, and fast. They were not known for either speed or decisiveness. Paulson groused, "The thing about these Brits is that they always talk and they never close" a deal. Worse, Paulson said he considered the person at the helm of Barclays, John Varley, a "weak man."

On that last score, Paulson had no worries. Barclays was being driven forward these days by its hard-charging number-two executive, Bob Diamond, an American who'd been a top executive at Credit Suisse First Boston and, before that, Morgan Stanley. Diamond had pulled together a tentative bid for Lehman. Paulson summarily got on the phone with Alistair Darling, who carried the staid and dusty t.i.tle chancellor of the exchequer of Britain. Darling, along with Barclays, had been running their due diligence on Lehman, looking at balance sheets and projections, and were concerned by what they saw. Darling feared that an acquisition of Lehman would expose Barclays and, by extension, the UK economy to much more risk than they were prepared to accept. Paulson attempted to a.s.suage Darling's concerns.

Still bluffing like a deal maker, Paulson cited Bank of America as a "backup buyer," even though talks with the Charlotte-based behemoth had already frosted. But Paulson's princ.i.p.al argument to Ken Lewis, that of systemic risk, carried even less purchase with a British-bank regulator. He needed to convince Darling that a Lehman failure could have ramifications across the pond. Yes, Darling felt, consequences that would be devastating if Barclays went down trying to effect a rescue.

Nonetheless, Paulson felt he'd made the case, a strong one, about the shared interests between the United States and Britain in a global financial system that all but mocked national borders and long-standing definitions of sovereignty. It had, in fact, been nearly two hundred years since Britain set the precedent for government's role in stopping a financial panic, firmly establishing the concept that certain inst.i.tutions in a society were simply "too big to fail." It was, specifically, the Panic of 1825, when a financial bubble, grown large with speculation on textiles and shipping, burst, leaving many of the banks insolvent. Those in Parliament cried that the banks had been warned about overspeculation, which many bankers had profited from, and that they should be left to their own demise. As the panic spread, depositors crowded into bank lobbies and were turned away by bankers trying to h.o.a.rd capital to stay afloat. The result: credit froze solid, commerce halted. After a weekend of heated meetings with the prime minister, his chancellor of the exchequer ordered the Bank of England's bailout of London's banks. The panic soon pa.s.sed and all was well, until the next panic.

Now, as the U.S. government tried to avoid having to bail out financial inst.i.tutions-which in this era acted like banks and then co-opted traditional banking functions with wild speculation-Hank Paulson and Tim Geithner felt that, in a crunch, the British would do what was needed.

Obama was on a run through New Hampshire on Friday afternoon, giving a few speeches before flying back to Chicago for a precious weekend of downtime with Mich.e.l.le and the girls. But already he sensed opportunity.

"The good news is that in fifty-three days, the name George Bush will not be on the ballot. But make no mistake: his policies will," he said to a large crowd at a gymnasium in Dover. "A few weeks ago, John McCain said that the economy is 'fundamentally strong,' and a few days later George Bush said the same thing. In fact, Senator McCain has said that we made 'great progress economically' over the last eight years. And here's the thing. I think they truly believe it."

Obama, from his many economic briefings, knew how wrong they were, and how, if a financial industry meltdown now further bruised the economy, such statements would seem nonsensical.

Working his phone nonstop between hits in New Hampshire, Obama got word that afternoon from his economic team that Paulson, Geithner, and representatives of all the major banks would be meeting briefly that evening and then all day Sat.u.r.day and Sunday at the New York Fed headquarters. Wolf sent a note to Obama saying he'd be representing UBS. Obama was delighted; he'd have a source in the room. He told Wolf to make sure to give him regular reports over the weekend.

On Friday evening inside the New York Fed's large conference room, Paulson and Geithner sat across from the heads of Wall Street. It was a replay, with much higher costs and stakes, of what the banks had done in 1998, when Long-Term Capital Management imploded: they gathered to divide up the damage to keep the financial system from locking up, each taking a share of the hit, except for Bear Stearns. For a decade, Bear Stearns' intransigence was a bitter pill, making it a moment of prairie justice that the fifth-largest investment bank had its comeuppance in the spring. Now Paulson, who until so recently would have been sitting in the seat now occupied by Lloyd Blankfein, told the group that there would be "no government money" helping this time, and anyone who didn't cooperate-in a thinly veiled reference to Bear Stearns-"would be remembered."

Paulson counted on their dreading a domino effect, that if the contagion of fear spread, they'd go down, one after another.

But fear was a hard sell to this crowd: in fact, over the course of nearly a year, it had failed to conquer deeply ingrained hubris and self-regard. Like European monarchs in the centuries before democracy, the financial industry CEOs-like so many American CEOs whose behavior Wall Street had shaped and then rewarded with 1980s-forward "innovations" in compensation practices-acted as sovereigns, untouchable. It was clear since the late summer of 2007 that many of their inst.i.tutions had been busily swapping and subbing debt-rolling it over, or keeping it invisibly tucked away far from their balance sheets-so they would not have to recognize their underlying insolvency. Vikram Pandit, presiding over the disaster that was Citibank, was sitting in a chair until recently occupied by Charles Prince, who had left in December with a compensation package worth $52 million. Bob Rubin, the bank's chairman, was in line for $126 million in compensation. Thain's predecessor, Stan O'Neal, had slipped out of a crippled Merrill the previous fall with a $72 million package. Hank Paulson himself, who oversaw Goldman's powerful, viral machinations in mortgage securities until his departure for Treasury in 2006, had left with a pay package worth $700 million.

John Mack of Morgan Stanley, looking up and down the table, asked where d.i.c.k Fuld was. Paulson said he was "in no condition to be helpful" right now. Some CEOs nodded.

Everyone in the room, men who knew one another as members of a club, understood the darkest scenario of dominos falling: that if Lehman, the fourth-largest investment bank, went down, then Merrill, the third-largest, would be next. After that, Morgan Stanley, the second-largest, and finally Goldman, at the top. Of course, the last three were all close to the same size, each of them three times the size of Lehman. But, in many ways, the more incisive disaster scenario might have put Merrill in its own special category of destructive capability, in terms of shattered confidence. Lehman, like the other investment banks, had only a small arm, $20 billion or so in a.s.sets, that came directly and regularly from the public-it was mostly an operator in funds, inst.i.tutional trading, bank-to-bank operations, and a wide array of investment activities.

The biggest part of Merrill, conversely, was still its ninety-four-year-old consumer brokerage business, now called "wealth management." That business had grown with breakneck speed over the past thirty years as millions of savers moved from traditional banks to investment funds. Driving the migration was a combination of the government's 1970s creation of tax-exempt 401(k)s and IRAs, to encourage saving, and the 1980s heady rise in stocks. Merrill, always the biggest brokerage, was soon enough handling the life savings of a significant slice of the country, as traditional banks once did, and there was no going back. By 2008 it boasted fifteen thousand financial advisers handling four million customer accounts worth $1.5 trillion. That would be the nest eggs of more than ten million people.

But, of course, it isn't a bank. So, those accounts are not federally insured, as bank accounts have been since the Great Depression. That means Merrill was actually like a huge national bank . . . but in 1929.

No one was quite sure what would happen if Lehman went down-trading desks at the big houses were trying to figure that out on Friday, running what-if scenarios. But, being focused on Lehman-with Paulson saying first let's save Lehman, then we'll think about what happens next-no one thought much about how a Merrill collapse would have sparked a public panic: people lining up at brokerage houses, banging on the windows and maybe worse, just like they lined up at the locked doors of banks seventy-nine years before, on Black Tuesday.

Such on-the-ground insights-about the behavior of panicked customers or how fear spreads across landscapes far from Wall Street-was not in the line of sight of those at the table. They were mostly looking at one another, measuring themselves against the only men, maybe anywhere, they considered peers and compet.i.tors. John Thain, the former number two at Goldman, who after Paulson left was beaten out for the top job by Blankfein, was eager to continue as Merrill's chief, something he'd long wanted to be, and didn't see that his house was already on fire, with a weight of toxic a.s.sets every bit as bad as Lehman's.

So many of the dynamics of the crisis, in fact, were exacerbated by the ego-addled dance of the CEOs-over the years and in this very room-marking an era when the imperial chief executive often existed in a cloud city immune from accountability, even to quarterly earnings. Critics of astronomical CEO pay, and captive boards, in which CEOs supported fellow CEOs, often said the interests of chief executives were no longer woven with those of either shareholder or employee. The idea was get to be a CEO, by any means necessary, and you'd live in your own separate universe that defied traditional laws of business physics.

Even d.i.c.k Fuld, atop his Midtown office tower, couldn't imagine a world without Lehman. A week before, the Koreans were ready essentially to merge with Lehman. Once his subordinates had it all set, Fuld, who'd been kept at a distance, burst into the deal room p.r.o.nouncing that they'd undervalued Lehman's real estate-that "plenty of those a.s.sets were good as gold"-ultimately scuttling the deal. Tonight, he was holed up atop his castle, where he'd presided for two decades, wrestling with wounded pride, and outrage-"Thain's worse off than we are!" he was yelling at subordinates, and "Hank will never let us go down." Paulson, and a few other CEOs in the know, were, meanwhile, livid at Fuld for having shooed away the Koreans.

Blankfein, for one, was concerned about how Lehman would pay its obligations to Goldman. Goldman, in fact, was owed money by many of the men in the room, having been early, and most active, in hedging and selling swaps on the great piles of toxic mortgage debt. For any banks wanting to restructure those debts, Goldman was poised like the sword of Damocles. Any attempts to restructure CDOs, to ease pressure on overburdened balance sheets, would trigger a contractual violation of the CDS contracts Goldman held, underwritten by AIG and other banks.

Jamie Dimon, who became CEO of JPMorgan in December 2005, was fortunate that his predecessor, Bill Harrison, expressed distaste in the late 1990s for the mortgage-backed securities and never wavered, even as JPMorgan's earnings sagged in the coming years compared with those of CDO-trading compet.i.tors. When Dimon took over in early 2006, with Harrison still chairman, he never loaded up on toxic mortgage securities, even though many of what would prove disastrous "innovations" in how to trade and account for CDOs occurred under Dimon's 1990s tenure at Citigroup. Now with the strongest balance sheet at the table, Dimon was looking for more "Jamie Deals," drawing suspicions from chairs on all sides.

Meanwhile, John Thain, sitting across from his old boss Paulson, and three seats down from his onetime rival Blankfein, wasn't even considering a world without Merrill. He'd just made it to the table. He wasn't going anywhere.

Greg Fleming, of course, was not a member of the CEOs club. Having spent two years enduring an array of cleansing ego-adjustments, he went home that evening to his home in Bedford, New York, ate a late dinner, and fell into fitful sleep. At 4:00 a.m. he awoke, and began wandering the halls of the silent house.

There were too many variables, too many to game. Padding barefoot in his kitchen, he found the world quiet and settled and coming into focus. Merrill needed to be sold this weekend or it would either die or be sold on its knees for a few bucks a share. A sense of panic began to rise from his gut. While Merrill had profited in recent years from exotic trading in CDOs, Fleming, after all, was overseeing the firm's old core business, "wealth management." He'd traveled the country, edge to edge, many times, talking to brokers at the big offices. Merrill was vastly, systemically woven into the global financial fabric, just like Lehman, but was significantly larger. Its accounts held the money of real people in the real country. Panic at trading desks and in corner offices was horrific, but nothing compared to angry mobs.

He put on a pot of coffee and grabbed a pad. He needed to move-now. Even though he'd received a come-hither nod of interest in early August from Bank of America, nothing could happen unless Thain were fully on board. He wrote down lines for what he would say to his CEO as soon as the sun came up.

The first call went in at 6:30 a.m. Thain was already in the back of a Town Car, on his way to New York City for the Sat.u.r.day meeting at the Fed. He was surprised to hear from Fleming. The two men did not get along.

"What is it?"

"I think you need to talk to Ken Lewis."

"About what?"

"About a deal."

"What are you talking about? Greg, this really isn't the time."

"If you don't talk with Bank of America, I think this company is going to fail."

There was a pause on the other end.

"Go on," Thain said.

Fleming laid out the variables, all the scenarios of what might happen over the weekend, and how Merrill needed to move, today, not in spite of those uncertainties, but because of them. "It's Sat.u.r.day. We have until Monday in early morning, before Asia opens, to get this done. If those markets open, and we don't have a buyer, we'll have the whole world breathing down our necks." He raced through a calculus of how Merrill's stock, currently $17, "could lose $15 in a day-and we'd be at $2. The next thirty-six hours is like eight years in deal land. Now's our moment."

"Greg, you're panicking," Thain said, dismissively.

Fleming was sitting on the front steps of his house, script in hand, and it was all slipping away. He made a desperate bid. "We need to do the right thing for our sixty-five thousand employees and the shareholders," he said evenly. "It's not about me and you and you being p.i.s.sed about what I'm saying and how I'm saying it."

Silence. After years living under Stan O'Neal's explosive bravado, Fleming had become a survivalist, expert in managing the blend of insecurity and willed confidence common to the modern, wildly compensated, and ever more imperial CEO.

"Listen," he said to Thain, his voice softening. "You're going to be a hero if you save this company."

Thain cleared his throat. "Well, umm, that's not the focus." And, finally: "All right, I'll think about it."

A green light? Not really. But Fleming decided to see it as one.

He immediately called up Greg Curl, Bank of America's number two, and Ed Herlihy, the bank's lead lawyer at Wachtell Lipton. "We're on."

Curl and Fleming each began to pound through their contact lists, alerting and a.s.sembling their respective SWAT teams of lawyers, accountants, and key executives within each inst.i.tution. The two number twos, both with a long history of buying and selling financial inst.i.tutions, had to get their two number ones together to get things launched officially. They decided on a 2:30 meeting in the Bank of America apartment in the Time Warner Center, on Columbus Circle. Curl called Lewis in Charlotte and said he needed to be on the corporate jet to New York, and fast. Lewis, following CEO protocol, said he wasn't getting on any jet until he'd heard directly from Thain.

While the CEOs settled into the New York Fed to try to make Lehman saleable to Barclays-which, Paulson had informed them, was the sole potential buyer-an actual Wall Street deal, the last deal of the golden era, was taking shape.

Or almost.

Fleming was back on the phone to Thain.

"You've got to call Lewis."

"No way," said Thain. "I don't want to do it. I'll be at a tactical disadvantage."

Fleming was speechless. A tactical disadvantage? It's these guys or n.o.body, he wanted to shout at Thain. Either they want Merrill or they don't. He composed himself. "Look, you just have to tell him it's a beautiful day in New York and you're looking forward to seeing him. Talk about the weather."

Thain wouldn't budge-nope-and then he hung up.

Fleming was back on the phone with Curl, staking everything-"my reputation, my whole career"-on a guarantee that Thain would show up at that apartment by 2:30. But Curl had his own imperial CEO to deal with. Without a call, Lewis wasn't coming-and unless the call came quickly, Lewis would be hard-pressed to get to New York today.

But now Thain wasn't picking up. He was in the conference room at the Fed, looking over Lehman's books. The clock was ticking. A half hour pa.s.sed; it was already past 11:00 a.m. Fleming had talked to his boss six times already that morning, and was now dialing into the ether every few minutes. Finally he called Ed Herlihy, Wachtell's lawyer, telling him, "this whole thing is about to collapse," because he couldn't get Thain on the phone. Herlihy paused. "He's sitting right next to me here," in the New York Fed conference room. "I think he just doesn't want to talk to you."

A few minutes later, Thain finally picked up. "I'm not happy with this Greg. I'm not happy with the way you're handling this!"

Fleming had nothing more to say. He pleaded, he begged.

"All right," Thain finally groused. "I'm going to call him just so I don't have to talk to you again!" And he hung up.

A few feet away, in the hallway outside the Fed conference room, Wolf looked for a quiet nook and punched in a number.

"Wolf, what have you got for me?"

"It's a f.u.c.king mess, Barack. Just getting our arms around the problem will be a feat."

Inside, he explained to Obama that they were, first, trying to figure out the depth of "the hole inside of Lehman"-meaning the value of its toxic a.s.sets and how far underwater that left the firm. Once that could be established-and the midday numbers looked to be about $70 billion-every bank would decide how much capital it was willing to put up to keep Lehman whole while it found a suitor. He said that the likely buyer was Barclays. But it would want only the profitable parts of Lehman. The toxic a.s.sets, in some of kind of bad bank, would have to be a.s.sumed by those in the room. Obama asked if they could accurately gauge the "depth of the hole," because "aren't lots of these securities difficult to value?"

Wolf said, yes, that was a problem, "especially considering how much stuff is not on the balance sheet" in terms of counterparty risks on instruments such as CDSs. "So what happens, Robert, on Monday morning if this doesn't work?"

"A s.h.i.t storm, Barack. We'd have to take the company apart, piece by piece, but it would be a nightmare. They're in the middle of trading relationships all over the planet. The value of billions of financial instruments would go zero, because they can no longer be funded."

Wolf had been closely following the solvency-versus-liquidity game since the previous summer, and he had his eye on Wall Street's insurance broker, AIG. That was not just an intermediary's business, like much of financial services. Insurance was different-a miracle product invented in the 1600s after the then-newfangled "theory of probability" was matched with statistical breakthroughs to create the actuarial tables still used today. Insurance was, in fact, the only proven model to manage and price risk, a leap of progress as great as any in human history. Life, fire, flood, liability, maritime, property, and casualty-the familiar list of lines allowed for the modern economy to develop across three centuries, and AIG was the world's largest insurer. And now the most precarious, having strayed from core principles to sell the faux insurance of CDSs. For that there was no actuarial table, no informed oceans of data, but rather computer models a.s.sessing the probabilities of events yet to occur.

"I don't understand why no one is talking about AIG," Wolf told Obama. "There's no way they can survive, and the part of AIG that's gone bad, with all the CDSs, will pull down the side that still insures everything under the sun."

Wolf was being beckoned. He had to go. He'd call later, and Obama went back to his first quiet Sat.u.r.day afternoon at home in three months.

As Thain prepared to meet Lewis at 2:30 uptown, he told Fleming, "I don't want to sell the company; they can buy 9.9 percent." At the New York Fed, Thain had been talking to Blankfein about Goldman Sachs taking a 9.9 percent stake, infusing some new capital into Merrill. He'd now decided that Bank of America would be convenient as someone for Goldman to bid against. Fleming was succinct: Bank of America wanted to buy the whole company or nothing, which is what Lewis soon told Thain at their 2:30 meeting when the Merrill CEO proffered his minority stake idea. The meeting was brief, perfunctory. Thain left Columbus Circle to head back downtown to the Fed, and told Fleming to nonetheless have the negotiations move on two tracks: one to sell the whole company, the other to sell a minority stake. Fleming nodded and ignored the directive.

Now, with the CEOs out of the way, he and Curl could actually start to cut the deal, something Wall Street still knew how to do, starting with a discussion of Merrill's earnings potential, with those fifteen thousand advisers and million of customers, who weren't going anywhere. Merrill also owned 50 percent of the giant a.s.set manager and mutual fund company BlackRock, locked up in 1994 when Fleming, then a young Turk at Merrill, helped Larry Fink break his firm away from Blackstone, the huge private equity firm. How did those franchises fit with Bank of America's structure and product lines-were there synergies or overlap-and would their value be enough to counteract the heavy load of hard-to-value toxic a.s.sets built up by Merrill's other half, its trading operations? It was, in essence, old Wall Street versus new: traditional investment, mutual fund, and brokerage activities versus the new innovations of bets and bonuses based on high-stakes math compet.i.tions. Merrill remained valuable because in the trading frenzy of the past decade, no one had bothered to jettison its legacy operations.

Inside the large conference room at the New York Fed, the heads of the major banks looked up and down the table. CNBC was now openly speculating about Lehman's impending bankruptcy, as the CEOs were forced to do with Lehman's books what many had avoided doing at their own firms: look hard at dizzyingly complex a.s.set-backed securities that a thirty-year debt fixation had bred and try to fix a value on credit default swap contracts linking all the banks in a daisy chain of disaster. Each of them was committed to kicking money into the Lehman hole, filling it, in essence, so Barclays would buy everything else. But had they really a.s.sessed all of Lehman's obligations? Everyone knew they hadn't. Who knew how many CDOs and CDSs Lehman had off its balance sheet, funded by overnight repos? Anybody's guess.

By the late morning on Sunday, after a sleepless night, Greg Fleming and Greg Curl were closing in on a final deal. Lewis had been peripherally involved. Thain, not at all. And that was the key. "This is it," Curl said. "It's $29 a share, a $50 billion acquisition. There is not a nickel more that we are willing to pay." It had been a night of insane brinksmanship inside Merrill, with factions forming and breaking up, price-per-share numbers flung in every direction-all the way down to $2 a share by some fearful Merrill advisers. Fleming immediately told Curl that $29 a share-a 70 percent premium to where the stock closed on Friday-would work. He now just had one call to make.

"John, it's done. We got them at $29!"

There was an unexpected silence on the other end.

"That's great," Thain said, "but I think we can get $30."

Fleming paused. A joke? "Come again?"

"I said we should go for $30. Call them back and get $30!"

Fleming felt the thread of sanity, the one that had helped him keep his cool over the most chaotic weekend of his life, tremble and break.

"You can call them back. Twenty-nine is it. I refuse to call them back."

How had a call that was supposed to be triumphant-euphoric, even-turned so acrimonious?

"You know what, Greg?" Thain shouted. "You are starting to p.i.s.s me off!"

The line went dead.

It took a minute for Fleming to see it. It was just human nature. Thain had been notably absent from the biggest deal in his company's history. In fact, he had been the biggest obstacle to getting it done. Now he wanted to put his thumbprint on the final doc.u.ment so he could go on to say, "I made this happen."

A half hour later, Thain called back. "Yes, it's time to move," he said quietly, as though breaking out of a trance. He threw in some caveats to create the illusion of having been involved in the historic deal, and in the loss of his prized seat at the CEO table.

Fleming said, sure, those were easy additions. And thanked his dazed CEO. Fleming had survived in his career by sticking with businesses handling money and risk that had been around, in various forms, for a century. And this core of Merrill, soon to be part of Bank of America, would survive with him-a twenty-fifth-hour maneuver that averted probable catastrophe.

A few hours later, as news of the Merrill deal spread through Wall Street's back channels, Paulson and Geithner a.s.sembled the CEOs in the conference room. They'd just finished a round of midmorning conference calls with Darling and the British regulators.

They wouldn't approve the deal. Barclays was out. Paulson was incredulous. Darling told Paulson he didn't want to take on "the cancer" of a flailing Wall Street giant. Paulson, while frustrated, knew that for Lehman Brothers, the prestigious investment bank, liquidation had become inevitable. Now all Paulson, Bernanke, and the country could do was brace for impact.

The CEOs were quiet; Paulson and Geithner, grim. They hadn't listened carefully to what the British were saying when they'd all talked on Thursday. They'd slammed into a brick wall of "no."

In some ways, this officially ended a year of colossal failure for Paulson, Geithner, and Bernanke. There was very little new information about the nature of the financial crisis that had emerged since the fall of 2007. All of what was knowable then was knowable now, and certainly discoverable with a round of modest, government-sponsored inquiry: the U.S. financial system was on the verge of collapse, ready to blow. What had been done in that year was emblematic of the modern dilemmas of projecting confidence, whether or not it is justified. Bernanke set up a "liquidity facility" in the fall of 2007. It had lent hundreds of billions in what was all but free money to banks, shadow banks, investment houses, and other companies to prop them up while the market could somehow correct itself. Insofar as the program was successful, as a stopgap, it was because it was secret-a fundamental violation of long-standing principles of corporate accounting, where the source of each dollar is supposed to be clear. As to any structural solutions, both the Fed and the Treasury ducked, as did the Bush administration, because acknowledging the need for dramatic action and then forcing architectural changes in the system would have undermined "confidence in the markets." The Treasury's main contribution was a grant of $30 billion to Dimon to take over Bear Stearns, and little else.