Saturday, September 13, 2008

Early Saturday morning, Jim, Christal, and I, accompanied by my Secret Service detail, left the Waldorf-Astoria Hotel in Midtown Manhattan, climbed into a car, and sped down a deserted Park Avenue, arriving at the New York Fed just after 7:00 a.m. It was quiet in the gray light and early enough that the television crews had not yet set up. Though everything had been hush-hush the night before, the news of our meeting was all over the morning’s papers. By the time Dan Jester arrived a few minutes later, reporters had begun to swarm outside the building.

We rode the elevator up to the 13th floor, where Tim Geithner had arranged for me to work in an office borrowed from his Information Technology department, just down the hall from his own suite. I went straight to work and called Ken Lewis, who reported that after closer inspection his people now believed that Lehman’s assets were in even worse shape than they had thought the previous evening—when they had said they wanted to leave $40 billion behind. I wasn’t surprised to hear Lewis put forward a new obstacle: it was increasingly obvious that he didn’t really want to buy Lehman. Nonetheless, we arranged for his team to come over to brief me later that morning.

I joined Tim in his office for a conference call with Barclays at about 8:00 a.m. Bank chairman Marcus Agius and CEO John Varley were on the line from London, and Bob Diamond was at Barclays’s Midtown Manhattan offices. Varley said they were working hard on a possible deal but needed to hear that Tim and I were serious. Barclays did not want to be used as a decoy. Varley said he had serious concerns about some of Lehman’s assets and indicated that Barclays would need to leave $52 billion of them behind. In addition to the problematic commercial mortgages, the list of dubious holdings included undeveloped land and Chrysler bonds that hadn’t been marked down.

I told Varley to focus on the biggest problems first—the assets he thought were going to be the most troubled—and tell us what he needed to take care of them. If Barclays gave us its best offer that day, we believed we could deliver a private-sector consortium that would fund whatever shortfall there was. Even as we spoke, I explained, the leaders of virtually the entire banking industry were assembling downstairs at the Fed. The Barclays bankers said they would keep working, and I ended the call encouraged that Lehman might have found its buyer.

We were scheduled to meet the Wall Street CEOs in the first-floor conference room at 9:00 a.m., but just before then Dick Fuld called. I briefed him on my unpromising conversation with Lewis and told him that it was more important than ever that he work with Barclays. He expressed great disappointment, bordering on disbelief, at BofA’s findings. He wanted to know more, but I had to cut him off to get to the meeting.

Addressing the CEOs for the second time in 12 hours, I tried to be totally open. I knew I had to give them crucial information as soon as I received it so that we could all quickly make informed judgments. I told them that Barclays appeared to be the most likely buyer for Lehman. I added that we had a meeting scheduled with BofA for later that morning, but I didn’t dwell on the prospects of a deal with the U.S. bank, and it must have been clear to the group that those talks weren’t going anywhere. I emphasized that we couldn’t do anything without their help.

“We’re working hard on a transaction, and we need to know where you guys stand,” I said. “If there’s a capital hole, the government can’t fill it. So how do we get this done?”

I can only imagine what was going through their minds. These were smart, tough businessmen, and they were in a difficult spot. We were asking them to rescue one competitor by helping to finance its sale to yet another competitor. But they had no idea of the true state of Lehman’s books or how much they would have to cough up to support such a deal. Without this information, they were flying blind: they couldn’t possibly predict the consequences of any course of action they chose. They knew how important it was to maintain a smoothly functioning market and how much we needed them to keep lending to one another if Lehman did go down. But their own institutions were all under grave pressure, and they had no idea what tests they might face in the days ahead—or whether they would be strong enough to survive this crisis.

As a group, the CEOs were nonetheless working hard to agree on a plan, but there was, understandably, some pushback. John Mack wanted to know why the government couldn’t arrange another assisted transaction, like the Bear Stearns rescue.

Tim quickly dismissed the possibility. “It’s not a feasible option,” he said. “We need to put another plan in place.” He made clear that the Fed could not lend against Lehman’s dubious assets but asserted that it wasn’t the government’s place to dictate the terms of any deal.

The three groups that Tim had organized to examine Lehman scenarios had worked through the night and reported on their progress. Citi, Merrill Lynch, and Morgan Stanley had been looking into an LTCM-type rescue, but that approach faded quickly as an option because it was impractical to liquidate Lehman without incurring huge losses, given the poor quality of its assets.

The team looking into how the industry might assist an independent buyer had spawned a series of subgroups to, among other things, scour Lehman’s books, identifying and valuing its toxic assets, and devise a deal structure that would allow an industry consortium to finance the purchase of, and absorb the losses on, those assets. Credit Suisse and Goldman Sachs led the way on valuing Lehman’s dubious real estate (Goldman had taken a look at the portfolio on its own earlier in the week). Credit Suisse’s Brady Dougan reported that private-equity assets carried by Lehman at $11 billion were worth around $10 billion, while real estate assets carried at $41 billion were more accurately valued at between $17 billion and $20 billion.

Brady’s report wasn’t a complete surprise, given the Street’s doubts about Lehman’s health, but it was shocking nevertheless. There was a more than $20 billion difference between what Lehman said its assets were worth and their true value. The CEOs were left wondering how their firms could fill a hole that size and what other bad assets—and losses—they would be asked to take.

With their background as major custodian banks, JPMorgan and Bank of New York Mellon had led the way on the “lights out” scenario. Noting the frailty of the market, and especially of the banks’ funding sources, Bob Kelly of Bank of New York Mellon remarked: “We have to figure out how to organize ourselves and how to do something, because we’re toast if we let this thing go,” he said.

I reiterated the severity of the situation. “I’m just going to say bluntly that you need to help finance a competitor or deal with the reality of a Lehman failure,” I told them.

“We must be responsible for our own balance sheet and now we’re responsible for others’?” Blankfein asked. “If the market thinks we’re responsible for other firms’ assets, that ups the ante.” The market, he believed, would now see all the investment banks as more vulnerable.

His observations had to trouble every free-marketer in the room. At what point were the interests of individual firms overridden by the needs of the many? It was the classic question of collective action. If the firms were forced to jointly support one failing institution, would they have to pony up aid for the next player to run into trouble? Where would it end? And what would the impact be on anyone’s ability to discern the industry’s true health? Potential investors assessing any bank’s balance sheet would have to consider not only its assets and liabilities, but whether it had properly accounted for the risk that it might have to bail out any one of its competitors. Under the circumstances, how could the market accurately gauge the condition of any financial institution?

When we stepped out into the main lobby, I noticed that the Fed building was filling up quickly. Before long, it seemed as if everybody I knew from Wall Street was there—CFOs, chief risk officers, heads of investment banking, senior staff from financial institutions groups, and specialists on lending, real estate, and private equity. Dozens of bankers were working on foldout tables spread throughout the lobby, in rooms off the lobby, and in offices all over the building, trying to come up with a rescue plan. Barclays had set up shop four floors above; Lehman was on the sixth floor; Bank of America was working at its New York offices. Each bank had a team of lawyers, and an unmistakable war-room atmosphere was evolving.

Tim and I decided we should meet individually with Jamie Dimon, Lloyd Blankfein, and John Thain. Jamie and Lloyd were the CEOs of the two strongest institutions and had been reducing their exposure to Lehman. We believed others would likely follow if they stepped up as leaders of a collaborative effort to save the stricken bank. John was a different matter entirely. Tim and I were concerned that if Lehman went down, his firm, which had the next-weakest balance sheet among investment banks, would be the next to go. We planned to ask him to find a buyer for Merrill Lynch.

Shortly before 11:00 a.m., Tim, Dan Jester, and I met in a 13th-floor conference room with Bank of America’s deal team: CFO Joe Price, head of strategy Greg Curl, financial adviser Chris Flowers, and legal adviser Ed Herlihy. Price and Curl explained that after poring over Lehman’s books, Bank of America now believed that to get a deal done it would need to unload between $65 billion and $70 billion worth of bad Lehman assets. BofA had identified, in addition to $33 billion of soured commercial mortgages and real estate, another $17 billion of residential mortgage-backed securities on Lehman’s books that it considered to be problematic. In addition, its due-diligence team had also raised questions about other Lehman assets, including high-yield loans and asset-backed securities for loans on cars and mobile homes, as well as some private-equity holdings. The likely losses on all of those bad assets, they estimated, would wipe out Lehman’s equity of $28.4 billion.

We asked if they would be willing to finance any of the assets they wanted to leave behind or take more losses. They said no.

To say the least, it was a disappointing session. Price and Curl weren’t even working off paper—they simply sat back in their chairs, reeling off ranges of huge numbers that would require an enormous private-sector bailout. At another time it might have been a humorous charade, but we were desperate to find a solution. Still, I wasn’t prepared to give up just yet, so I asked them if they would be available for a meeting or a call later to discuss in more detail what assets they wanted to leave behind. At a minimum I wanted to keep BofA warm as a bidder, because the presence of another buyer would help us negotiate more effectively with Barclays.

As everyone got up to leave, Chris Flowers motioned me aside and said, “Hank, can I tell you what a mess it is over at AIG?” He produced a piece of paper that he said showed AIG’s day-by-day liquidity. Scribbling arrows and circles on the sheet to outline the problem, Flowers told me that according to AIG’s own projections the company would run out of cash in about ten days.

“Is there a deal to be done?” I asked.

“They are totally incompetent,” Flowers said. “I would only put money in if management was replaced.”

I knew AIG was having problems—its shares had been pummeled all week—but I didn’t expect this. In addition to its vast insurance operations, the company had written credit default swaps to insure obligations backed by mortgages. The housing market crash hurt AIG badly, and it had posted losses for the last three quarters. Bob Willumstad, who had shifted from chairman to CEO in June, was expected to announce a new strategy in late September.

I relayed Flowers’s information to Tim, and we agreed to invite Willumstad over. He surprised me by saying Flowers shouldn’t attend. “Flowers is the problem, not the solution,” Willumstad said. I suspected that Chris was trying to buy pieces of AIG on the cheap, and I promised he would not be part of the meeting.

Tim and I met privately with Jamie Dimon. A number of CEOs had expressed concern to us that he was using the crisis to maneuver his bank into a stronger position. Indeed, some were convinced that he wanted to put them out of business entirely. We led off by raising these complaints. Jamie assured us that JPMorgan was behaving responsibly but pointed out that he ran a for-profit institution and had an obligation to his shareholders. I emphasized that we needed him to play a leadership role in averting a Lehman Brothers failure.

Then, because I respected his judgment, I pressed Jamie for his assessment of the situation. Did he think we had a chance of putting together an industry agreement to save Lehman? He said it would be difficult but possible. The European banks would have a tougher time getting a quick decision from their boards and regulators, but he felt they would probably come through, too. In the end, I felt reassured that I could count on Jamie’s leadership.

Tim and I spoke to Lloyd in the afternoon. He was still questioning the idea of a private consortium, given the weakness of the industry.

“Do you think this makes sense?” he asked us. “What will you ask for next week when Merrill or Morgan Stanley goes?”

“Lloyd, we’ve got to try to stop this thing now,” I said.

“Goldman will act responsibly,” he replied. “We’ll do our part, but this is asking a lot, and I’m not sure it makes sense.”

Tim and I believed that both Lloyd and Jamie would ultimately support a private-sector consortium, and I was optimistic that the CEOs would come up with a plan. Now we had to make sure that Barclays was on board.

Tim and I returned to the first floor about 3:30 p.m., shortly after Lloyd left, and reconvened a group meeting with the CEOs. I assured them that Barclays seemed interested and aggressive. I didn’t bother talking about BofA. It was obvious from the morning meeting that the Charlotte bank had lost interest. I asked the group to intensify its efforts and find a way to finance any assets Barclays might want to leave behind.

The CEOs were testy, but in what I felt was a productive way. They were being asked to risk billions of dollars. They had been getting due-diligence reports on the quality of Lehman’s assets from their people, and they knew that to make the math work, they would have to make a loan secured by assets worth much less than their stated value. In other words, they would have to take a mark-to-market loss the moment a deal was completed. The question was: how much would they eventually get back?

Vikram Pandit asked why banks like Citi, which had retail-based funding sources, should have to put up as much as those that relied on wholesale funding. After all, it was the investment banks, which lacked consumer deposit bases and depended on the institutional money markets, that were in trouble.

“You’ve got as much wholesale funding as anybody here,” Lloyd Blankfein shot back at Vikram. “And because you’ve got the Fed behind you, you’re like a big utility.”

As ever, Jamie Dimon zeroed in on specifics. “Barclays is going to buy all the assets they want and assume all the liabilities they want, but what liabilities are they going to leave behind?” he asked. “Are they going to take tax liabilities and shareholder litigation from prior years, or is that being left for the Street?”

Tim and I met one last time, for just a few minutes, with Curl and Price from Bank of America. But we made no progress. By the time we had our third call with Barclays that day, at 4:30 p.m., BofA was out of the picture. Everything now depended on the British bank.

Each time we had spoken on Saturday, our discussions had become more granular as Barclays focused on the quality of Lehman’s assets and the due diligence they needed to perform. Earlier, Barclays had also mentioned that its regulator, the Financial Services Authority, wanted to be sure the British bank had an adequate capital plan in place to back the deal, an understandable requirement that we expected could be met.

Now Bob Diamond raised a new, troubling issue. Given the size of the transaction being contemplated, he said, it appeared that Barclays might be required, in accordance with its London listing requirements, to hold a shareholder vote to approve the merger. He said he hoped a vote wouldn’t be needed, but if it was, would the Federal Reserve guarantee Lehman’s massive trading book until the deal was approved? The vote could take 30 to 60 days.

Tim carefully replied that the Fed was unable to provide any such blanket guarantee. But if a vote proved to be necessary, Barclays should quickly come up with their best ideas on how to deal with it, and the Fed would examine its options.

Even as I strived to maintain industry backing for a Lehman deal, Merrill Lynch had been weighing on my mind. The weekend had bought the firm a little time, but I hated to think what would happen come Monday—especially if we couldn’t save Lehman.

Around 5:00 p.m. John Thain, responding to my invitation, walked through the door of my 13th-floor office. He had never been good at hiding his emotions; now he looked somber and uneasy. Tim had to take a phone call, so I began the meeting alone.

By this point, I had begun to suspect that BofA had set its sights firmly on Merrill and the legions of retail stockbrokers that I knew Ken Lewis had long craved. But I wasn’t positive that this was the case, and I felt the need to make sure John understood the seriousness of the situation: Merrill was in imminent danger and he needed to act.

As we talked about the lack of options for his firm, I could see that the full impact of the crisis had settled on John. Just as with Lehman, I stressed, the government had no powers to save Merrill. Under the circumstances, he should try to sell the firm. He said he was exploring his options and talking with Bank of America, Goldman Sachs, and Morgan Stanley. He asked what I thought about a merger between Merrill and Morgan Stanley. I told him it didn’t make sense: there would be too much overlap, and the market wouldn’t like it.

“I agree,” John said.

We also discussed Bank of America. I told him that I believed that BofA was the only interested buyer with the capacity to purchase Merrill. Still, John’s manner was somewhat evasive. I couldn’t tell if he really wanted, or intended, to sell the firm. He himself may not have known at that point.

AIG’s Bob Willumstad arrived at the New York Fed late in the day, accompanied by his financial and legal advisers. We sat down in a conference room on the 13th floor. Willumstad, a soft-spoken man who had once run Citi’s global consumer group, was very candid, admitting that AIG had a multibillion-dollar liquidity problem stemming from losses in its derivatives business and an imminent credit rating downgrade. He now told us that without a big infusion of money, AIG estimated it would run out of cash as soon as the following week. He described efforts to raise $40 billion in liquidity by selling certain healthy insurance subsidiaries to private-equity investors and by using some unencumbered securities from its insurance subsidiaries as collateral. Doing so would require the approval of Eric Dinallo, the superintendent of insurance for New York State. Bob said that the New York regulators supported the plan, and he was optimistic that the problem would be resolved by the end of the weekend.

I knew that Willumstad had gone to Tim earlier to see if AIG could have access to the Fed’s discount window in an emergency, and that Tim had said he couldn’t loan to a nonbank like AIG. It gave me a chill to think of the potential impact of AIG’s problems. The firm had tens of millions of life insurance customers and tens of billions of dollars of contracts guaranteeing 401(k)s and other retirement holdings of individuals. If any company defined systemic risk, it was AIG, with its $1 trillion balance sheet and massive derivatives business connecting it to hundreds of financial institutions, governments, and companies around the world. Were the giant insurance company to go under, the process of unwinding its contracts alone would take years—and along the way, millions of people would be devastated financially.

The company’s immediate difficulties stemmed from the fact that it had written huge amounts of credit default swap insurance on obligations backed by mortgages. Those contracts included triggers: if AIG got downgraded, it had to post additional collateral. AIG’s collateral requirements also depended on the fair market value of the securities it insured, which was eroding with the declining housing market. In this Saturday meeting, Doug Braunstein, AIG’s financial adviser from JPMorgan, described AIG’s books as aggressively marked.

“What do you mean by aggressively?” I asked.

“The opposite of conservatively,” the veteran banker shot back quickly.

Not long afterward, I shared my concerns about Lehman with Josh Bolten at the White House. “This is one of the most difficult situations I could have imagined,” I said. “There’s a big difference between what Lehman assets were marked at and what the buyers are willing to pay.”

Josh got an earful from me as I explained the other two balls we were juggling in New York. We had gone into the weekend to save Lehman Brothers, and now AIG was facing a liquidity crisis that had put it on the verge of bankruptcy, and we had become concerned enough about Merrill Lynch to urge John Thain to sell that firm.

Meantime, the CEOs and their teams were all working hard. It was an amazing scene, all these financial industry executives reviewing spreadsheets, crunching numbers, trying to devise a solution. Rivals from different firms were working together. Senior traders sat at one set of tables, figuring out how to net out firms’ exposures if Lehman went down. In another area, people studied Lehman’s private-equity portfolio, trying to get a handle on the losses their firms would have to absorb if they lent money against it. It was inspiring to see all these fierce competitors trying to save a rival.

By evening the CEOs had agreed to support in principle a proposal under which Barclays would leave behind a pile of bad real estate and private-equity investments and wipe out Lehman’s preferred and common shareholders. To make the deal work, Barclays wanted the consortium of Wall Street firms to agree to loan up to $37 billion to a special purpose vehicle that would hold the assets. These had been carried by Lehman at $52 billion, but after their analyses the firms estimated their value at closer to $27 billion to $30 billion. The firms stood to lose collectively up to $10 billion. Barclays was also going to contribute some of its own shares, which would reduce the loss to the firms. It would still cost them dearly, but Lehman would be saved.

I left the New York Fed before 9:00 p.m. optimistic about the prospects for a deal. The industry was doing its part to come up with funding, and I had reason to believe we would find a solution to Barclays’s need for a shareholder vote.

Anticipating another sleep-deprived night, I arrived back at the hotel exhausted. I went into the bathroom of my room and pulled out a bottle of sleeping pills I’d been given in Washington. As a Christian Scientist, I don’t take medication, but that night I desperately needed rest.

I stood under the harsh bathroom lights, staring at the small pill in the palm of my hand. Then I flushed it—and the contents of the entire bottle—down the toilet. I longed for a good night’s rest. For that, I decided, I would rely on prayer, placing my trust in a Higher Power.

Sunday, September 14, 2008

I had gone to bed modestly optimistic about our chances of saving Lehman and hopeful that John Thain would find a partner for Merrill Lynch. I’d left Steve Shafran and Dan Jester behind, working at the New York Fed with Bob Diamond and the Barclays team to nail down their offer, and with the Wall Street consortium to structure the loan terms. When I spoke to Steve and Dan first thing Sunday morning, they’d barely had time to take a shower or shave, much less sleep. Reasonably confident that the Barclays bid was proceeding, they’d left the Fed at 4:00 a.m., when Diamond said he had to plug into a board meeting. They also reported that they had made good progress with the consortium on a preliminary term sheet for the loan that the Wall Street firms would need to provide for the Barclays deal.

Tim spoke with Diamond after the Barclays board meeting, at 7:15 a.m. New York time, and Bob warned him that Barclays was having problems with its regulators. Forty-five minutes later Chris and I joined Tim in his office to talk with Diamond and Varley, who told us that the FSA had declined to approve the deal. I could hear frustration, bordering on anger, in Diamond’s voice. He and Varley indicated that they were surprised and embarrassed by this turn of events.

We were beside ourselves. This was the first time we were hearing that the FSA might not support the deal. Barclays had assured us that they were keeping the regulators posted on the transaction. Now they were saying that they didn’t understand the FSA’s stance. We told them we would contact the U.K. officials right away and get to the bottom of this.

Subsequently, Tim and Chris spoke separately with Callum McCarthy, the FSA chairman. The British regulator, they learned, was not prepared to approve the merger, but at the same time, the FSA was careful to say it was not disapproving the merger, either. I recall both Tim and Chris saying that the FSA had raised concerns about the need for more due diligence, Barclays’s plans to raise capital to fund the acquisition, and guaranteeing Lehman’s trading book during the shareholder vote. All this added up to a delay, and delaying the deal was the same as killing it: we needed certainty today.

As I listened to Tim and Chris, I went over again in my head my Friday conversation with Alistair Darling, and it occurred to me that I had not caught his true meaning when he’d expressed concern about a British bank’s buying Lehman. What I had taken as understandable caution should have been taken as a clear warning.

Tim spoke with Callum McCarthy again around 10:00 a.m. in an attempt to get the British to waive the listing requirement for a shareholder vote so that Barclays could go ahead and buy Lehman. But the FSA chief put the onus on Darling, saying that only the chancellor of the Exchequer had the authority to do that.

With Bank of America gone and Barclays now in limbo, we were running out of options—and time. Treasury had no authorities to invest capital, and no U.S. regulator had the power to seize Lehman and wind it down outside of very messy bankruptcy proceedings. And unlike with Bear Stearns, the Fed’s hands were tied because we had no buyer.

Markets demand absolute certainty, and we had known all along that Lehman couldn’t open for business on Monday unless it had lined up a major institution, like Barclays, to guarantee its trades. That had been the crucial element of the Bear Stearns rescue. Even after JPMorgan, backed by the Fed, had announced that it would lend to Bear Stearns on Friday, March 14, the investment bank had continued to disintegrate. A collapse was only avoided on Sunday when JPMorgan agreed to buy Bear and guarantee its trading obligations until the deal closed. That halted the ongoing flight of counterparties and clients, averting Bear’s bankruptcy.

The Lehman situation differed from Bear’s in another important way. The Bear assets that JPMorgan left behind were clean enough to secure sufficiently a $29 billion Fed loan. But an evaluation of Lehman’s assets had revealed a gaping hole in its balance sheet. The Fed could not legally lend to fill a hole in Lehman’s capital. That was why we needed a buyer. And we hoped that the private sector would assist the buyer by providing $37 billion in financing that was exposed to $10 billion or so of expected losses from minute one.

The Fed had no authority to guarantee an investment bank’s trading book, or for that matter any of its liabilities. And without an acquirer with a big balance sheet to ensure solvency, a Fed liquidity loan would not have been sufficient to hold Lehman together during a shareholder vote. Instead, the Fed would have been lending into the same kind of run on Lehman that Bear suffered before JPMorgan came through. In the 30 to 60 days that could elapse before a shareholder vote, account balances would drain; huge amounts of collateral would be pulled as trades were unwound while hedge funds and other key customers fled; bank employees would quit. And then, most likely, Barclays shareholders would vote the deal down. The Fed would find itself in possession of an insolvent bank and out tens of billions of dollars.

I delivered the bad news to Josh Bolten, who had already spoken to the president about the possibility of a Lehman failure.

“You’ve got presidential approval to settle on a wind-down that doesn’t commit federal resources,” Josh told me. “Anything else, you should come back to the president and tell him what you’re planning.”

Tim, Chris, and I were running late for our scheduled 10:00 a.m. meeting with the CEOs downstairs. Believing we shouldn’t sugarcoat the situation, I told the bank chiefs we had run into some regulatory issues with Barclays but were committed to working through them.

The CEOs presented us with a term sheet for the deal. In the end, they had come much further than Tim and I thought they would. They had agreed to put up more than $30 billion to save their rival, and they had figured out how to spread the risk across the industry. If Barclays had committed to the deal, we would have had industry financing in place.

Tim asked the group to keep plowing ahead, but I imagine everyone suspected that the deal was in jeopardy.

At 11:00 a.m. I went back upstairs, and within half an hour I was on the phone with Alistair Darling, who wanted a report on Lehman. I told him that we were stunned to learn that the FSA was refusing to approve the Barclays transaction. I pointed out that we had run out of options for Lehman, because U.S. officials had no statutory ability to intervene.

He made it clear, without a hint of apology in his voice, that there was no way Barclays would buy Lehman. He offered no specifics other than to say that we were asking the British government to take on too big a risk, and he was not willing to have us unload our problem on the British taxpayer. Alistair’s chief concern was the impact of a Lehman failure on the British financial system. He wanted to know what the U.S. would do once Lehman failed.

“We are very concerned over here,” he said. “Lehman has a significant business in the U.K., and we have real concerns as to whether it is adequately capitalized.”

The chancellor of the Exchequer was delivering a clear message: we would get no help from the British. Our last hope for Lehman was gone.

I hung up feeling deflated, and frustrated that we had wasted so much time with Barclays on a deal that could never have been done. I was frustrated, too, that unlike Barclays, the British were not simply asking directly that the Fed guarantee Lehman’s trading book, even if the Fed lacked that power. Frankly, I was beginning to believe that the British were afraid that if they did push, the Fed would somehow find a way to guarantee it, leaving them one less excuse for not approving the deal.

I could only surmise that if Darling wasn’t presenting any options or leaving any room to negotiate, it was because the British had their reasons for not wanting this deal done. In truth, I could understand their hesitation. The U.K.’s bank situation was more perilous than ours. Altogether, British banks’ assets amounted to more than four times the size of the national GDP; total U.S. banking assets were about the same size as our GDP. Moreover, individual U.K. banks, including Barclays, had capital issues of their own. It was understandable that the country’s officials might be reluctant to waive normal shareholder procedures for a deal that could have resulted in big losses to one of their largest institutions while carrying no risk for the U.S. government.

“Darling’s not going to help,” I told Tim. “It’s over.”

At that moment, I did not have time for regret, recriminations, or second-guessing. I could only think about the enormous challenge we faced.

I’d asked John Thain to come up to see me, and he arrived right after my conversation with Darling. I got to the point: “Have you done what I recommended and found a buyer?”

“Hank, I’m not thick,” he responded, slightly irritated. “I heard you. I’m doing what I need to do.”

John didn’t mention Bank of America, but I did. By this point I assumed he was in serious negotiations to sell Merrill to the bank, and I said he should focus on doing that deal.

John’s no actor, and I could tell he was deeply engaged in merger talks. I was relieved: with Lehman all but finished, I didn’t want to see Merrill dragged down next.

I phoned my Treasury team in Washington to brief them on the unhappy developments with Lehman and warn them that the markets were going to get very choppy. I asked Kevin Fromer to get ready to talk to the appropriate staffs on the Hill, and I made sure that Michele Davis was prepared to deal with the press, which was expecting a big announcement on Lehman before the Asian markets opened.

All weekend Dick Fuld had been holed up at Lehman headquarters, making phone calls. Now I called him back.

“Dick, I feel terrible,” I said. “We’ve come up with no options. The British government is not going to let Barclays go ahead. BofA isn’t interested.”

“This can’t be happening,” he said. “Hank, you have to figure something out!”

Fuld couldn’t understand that the BofA deal was gone. It was impossible not to sympathize with him. After all, I had run a financial institution; he had been one of my peers. I couldn’t help thinking what this would mean for the thousands of people who worked for Lehman Brothers, one of whom was my brother, Dick.

Fuld had also been calling Tim and Ben, but only I talked to him. Although I hadn’t been directly involved in the discussions between Barclays and Lehman, I knew that he had been shunted aside and that Lehman president Bart McDade had taken over the negotiations.

We’d scheduled another meeting with the CEOs for 12:30 p.m., but once again we were running late, because Tim was back on the phone with Callum McCarthy, fighting to the end for a Barclays-Lehman deal. I stood beside him, watching him jot notes on a pad—calm and methodical as always, although he must have been as frustrated as I was. He was pressing McCarthy about his reasoning and asking if there was anything that could be done to speed the FSA’s deliberations up or to get the deal done.

And then Tim hung up.

“I made no progress,” he said simply. The FSA continued to be unwilling to say what it would take to approve the deal.

With that, we walked to the elevators. To reach the conference room, we had to wade through all the Wall Street executives milling around the first floor. It was like pushing through a crowd at a stadium. Everyone, it seemed, wanted to speak to us. They were working hard and were eager for an update, and I felt as though they were all scanning my face or Tim’s to guess the verdict. I wish I could have been buoyed by their energy and effort, but I felt numb. The news I was about to deliver could only hurt them. Some of the crowd tried to follow us into the conference room, but we shut the door on them, limiting the meeting to the CEOs.

It was shortly before 1:00 p.m. when Tim, Chris, and I addressed the CEOs again. I was completely candid. Barclays had dropped out, and we had no buyer for Lehman. We were going to have to make the best of it.

“The British screwed us,” I blurted out, more in frustration than anger.

I’m sure the FSA had very good reasons of their own for their stance, and it would have been more proper and responsible for me to have said we had been surprised and disappointed to learn of the U.K. regulator’s decision, but I was caught up in the emotion of the moment.

“We’re going to have to all work together to manage this,” I went on. “We’ve got no buyer, and there’s nothing to do about it.”

Having been forewarned of this possibility at the morning’s meeting, nobody seemed shocked by the bad news. They may even have felt momentary relief not to have to commit billions to an iffy rescue. But as the reality sunk in they became somber. And then they quickly began to come together, focusing on a single question: How are we going to prepare for the markets’ opening on Monday?

Chris Cox talked a little about the process going forward. He said the SEC had been working for a long time on detailed plans for handling a Lehman bankruptcy.

As I made my way from the conference room, a number of executives rushed up to me for news. A contingent from Lehman crowded close to the doorway. Rodge Cohen, who was advising Lehman, approached me, accompanied by Bart McDade.

“Hank, what’s happening?” he asked.

I gave them the bad news. “We had the banks ready to do the deal, but the British wouldn’t approve it.”

Rodge grabbed hold of me and said, “Hank, this is terrible.”

I remember how he and McDade implored us to try something else. I could see the devastation in their faces as they took in the cold, stark reality: this was the end. They had scrambled all weekend, and I felt terrible for them, and particularly for McDade, a stand-up guy who had been thrust into an impossible job at the last possible minute.

Back in my temporary office on the 13th floor, a jolt of fear suddenly overcame me as I thought for a moment of what lay ahead of us. Lehman was as good as dead, and AIG’s problems were spiraling out of control. With the U.S. sinking deeper into recession, the failure of a large financial institution would reverberate throughout the country—and far beyond our shores. I could see credit tightening, strapped companies slashing jobs, foreclosures rising ever faster: millions of Americans would lose their livelihoods and their homes. It would take years for us to dig ourselves out from under such a disaster.

All weekend I’d been wearing my crisis armor, but now I felt my guard slipping as I gave in to anxiety. I knew I had to call my wife, but I didn’t want to do it from the landline in my office because other people were there. So I walked around the corner to a spot near some windows on the other side of the elevators and phoned Wendy, who had just returned from church. I told her about Lehman’s unavoidable bankruptcy and the looming problems with AIG.

“What if the system collapses?” I asked her. “Everybody is looking to me, and I don’t have the answer. I am really scared.”

“You needn’t be afraid,” Wendy said. “Your job is to reflect God, infinite Mind, and you can rely on Him.”

I asked her to pray for me, and for the country, and to help me cope with this sudden onslaught of fear. She immediately quoted from the Second Book of Timothy, verse 1:7—“For God hath not given us the spirit of fear, but of power, and of love, and of a sound mind.”

The verse was a favorite of both of ours. I found it comforting and felt my strength come back with this reassurance. With great gratitude, I was able to return to the business at hand. I called Josh Bolten and New York City mayor Michael Bloomberg to alert them that Lehman would file for bankruptcy that evening.

We had tried during the summer and more intensely in the last few days to be ready for this moment. Beginning right after I had informed the CEOs that Barclays was done, the Wall Street firms, under the guidance of Tim and the New York Fed, got down to work. Among other things, they divided the industry into teams to try to minimize the disruptions that were likely to occur the next day.

A group on the 13th floor worked through other issues. The Fed had decided it could and would lend directly to the Lehman broker-dealer arm to enable it to unwind its repo positions. (Over the next few days, it would lend as much as $60 billion for this purpose.) Separately, the International Swaps and Derivatives Association had agreed to sanction an extraordinary derivatives trading session. It began at 2:00 p.m., and though originally scheduled to run until 4:00 p.m., it would be extended another two hours. The aim was for the firms to unwind as much as they could, and to offset their exposure to Lehman, before the firm declared bankruptcy and threw the market into disarray.

With a company like Lehman that had operations across the globe, bankruptcy raised enormously complex issues. Which entities would file for bankruptcy, and which would not? Would the European and U.K. entities file before the New York holding company? The Federal Reserve and the SEC had to work these details out with Lehman in order to orchestrate the proper sequence of filings. Lehman’s broker-dealer had to be open for business on Monday for the Fed to be able to backstop the unwinding of Lehman’s giant repo book.

One of the biggest issues was that the firm did not appear to have taken seriously the possibility of having to file for bankruptcy until the last minute. A Lehman team, accompanied by their counsel Harvey Miller of Weil, Gotshal & Manges, would not arrive at the New York Fed to discuss bankruptcy options until early Sunday evening, and even then Lehman appeared to have no immediate intention of filing.

In the midst of all this, President Bush called me at about 3:30 p.m.

“Will we be able to explain why Lehman is different from Bear Stearns?” he asked.

“Yes, sir,” I replied. “There was just no way to save Lehman. We couldn’t find a buyer even with the other private firms’ help. We will just have to try to manage this.”

I had to add that Merrill, now in talks with BofA, was the next-weakest investment bank, and that AIG had a severe liquidity problem. I also told the president that in my opinion we might need to go to Congress to get expanded powers to deal with the crisis. The problems we had to contend with were coming at us fast and all at once. The case-by-case approach we had been using since Bear Stearns was no longer enough. President Bush—reassuring, as always—told me we would figure out how to work through the crisis. We agreed to meet the next day after I returned to Washington.

Even as we struggled with Lehman, AIG rushed to center stage. That afternoon, Chris Flowers called Dan Jester to say he’d made a proposal to AIG to acquire some of the company’s most valuable subsidiaries. It sounded to me like Flowers was trying to take the company for next to nothing. At the same time, other private-equity firms were doing due diligence on various parts of AIG’s operations. But Bob Willumstad had his own proposal for us.

A little before 5:00 p.m., Willumstad returned to the New York Fed with his advisers, and we again met in the conference room on the 13th floor. Willumstad delivered terrible news: The only proposal he had been able to generate from private-equity investors came from Flowers, and his board had rejected it as inadequate. Further, AIG had discovered another major problem: huge losses in its securities lending program. AIG had been lending out its high-grade bonds and receiving cash in return. It reinvested the cash in mortgage-backed securities, hoping to earn some extra income. As counterparties sought to unwind the deals to avoid exposure to AIG, the insurer faced the prospect of having to sell the illiquid mortgage-backed securities at big losses. It was clear that AIG’s cash crunch would likely occur sometime within the week—sooner than we had been told Saturday morning.

But Willumstad had a new plan, in which the Fed would provide a $40 billion bridge loan, in addition to the $10 billion AIG would generate from unencumbered securities. The company would sell some of its insurance company subsidiaries and use the proceeds to pay back the loan.

It was unnerving. Tim and I knew that an AIG bankruptcy would be devastating, leading to the failure of many other institutions. In one day the company’s shortfall had mushroomed to $50 billion. Tim said that the Fed was not prepared to lend to AIG and that the company should get a consortium of private lenders to make a bridge loan.

I joined Tim and Fed governor Kevin Warsh on a call with Ben, Fed vice chairman Don Kohn, and the rest of Ben’s team in Washington. We reviewed the day’s dreadful events. We were doing all we could, in Tim’s phrase, to spread foam on the runway to cushion the coming crash of Lehman.

Among these measures, the Fed had expanded the range of collateral that brokers could pledge to receive loans via the Primary Dealer Credit Facility (PDCF) to include anything accepted in the triparty repo system—such as stocks and non-investment-grade bonds. The big worry was that in the wake of a Lehman failure repo lenders would shy away from investment banks and other financial firms heavily dependent on that kind of financing. By expanding the PDCF’s eligible collateral, the Fed aimed to reassure repo lenders that if any investment bank counterparty ran into problems, it could get cash from the Fed for any collateral and use that to repay the triparty repo lender.

Separately, with encouragement from Tim and me, ten of the Wall Street firms had come together to create a $70 billion facility of their own that would provide emergency liquidity support for any of the participating banks that needed it.

After all these measures, though, we had run out of gas. None of us had any confidence that they would be sufficient. Some in the group asked if we should revisit the idea of putting public money into Lehman, but Tim said there was no authority to do that.

We were all frustrated to have worked so hard and come up empty. We knew that the consequences of the Lehman failure would be awful, but even so, we did not know what would face us in the morning—or in the days to come. I had a sense that the situation had gone beyond our ability to handle it on our own. I told Ben and Tim and the others on the call that the time had probably come to go to Congress for fiscal authorities to deal with the unfolding crisis. We had all wanted this for some time.

After the Fed call, I heard the only good news of the weekend: Bank of America was going to buy Merrill Lynch for $50 billion. Thain had managed to arrange a sale at $29 per share, a 70 percent premium over Merrill’s market price. I was relieved: without this, I knew, Merrill would not have lasted the week.

We had planned to announce Lehman’s bankruptcy at 4:00 p.m., four hours before Japan’s markets opened, to allow as much time as possible for market participants to prepare themselves. The SEC was supposed to take the lead on this, but all afternoon I got reports from the Fed that the commission was moving slowly. Chris Cox had been in his office for hours working on a press release to assure Lehman’s broker-dealer customers that they would be protected under SEC regulations. He was also supposed to discuss Lehman’s planned course of action with the company’s board of directors, but he had yet to do so.

Pressed by Tim and others, I finally walked into Chris’s office around 7:15 p.m. and urged him to move quickly to execute the SEC’s plan. “The Asia markets are opening!” I said. “You need to get your announcement out soon, and you can’t do that unless you are coordinating with Lehman. It is essential that you call the company now.”

Chris was waiting for Lehman to file for bankruptcy of its own volition. I understood that it was unusual and awkward for a regulator to push a private-sector firm to declare bankruptcy, but I stressed that he needed to do something to get the process moving for the good of the rest of the system. And although Chris wanted Tim and me to join him on the call, I said that as Lehman’s regulator, he should make the call by himself.

Finally, sharing the line with Tom Baxter, the general counsel of the New York Fed, and other Fed and SEC staffers, Cox called Fuld shortly after 8:00 p.m. to reiterate that there would be no government rescue. Lehman had no alternative to bankruptcy. Fuld connected Cox to Lehman’s board.

“I can’t tell you what to do,” Cox told them. “I can only tell you to make a quick decision.”

As it was, Lehman did not file for bankruptcy until 1:45 a.m. Monday, well after the Asian markets had opened.

While Tim and I waited together for Chris to complete the call with Lehman, I phoned Michele Davis and told her that despite the good news on Merrill Lynch, I was expecting a tough week. As difficult as it was going to be to get fiscal authorities from Congress, we didn’t have much choice, and it was going to take an all-out effort on the Hill. I told her I had alerted the president.

Kevin Fromer had been dealing with the legislative staffs, but I needed to brief the major congressional players and called Chuck Schumer, Barney Frank, Chris Dodd, and Spencer Bachus. “How are all of these free-market people going to feel about letting the markets work?” Barney asked me pointedly. But he clearly understood the ugly ramifications of these developments. He added that he was disappointed not to have heard from me earlier.

Before I left the New York Fed I met a final time with Tim. He had his work cut out for him, navigating the Lehman mess and trying to forestall an even worse one at AIG. Tim was still hoping to fashion a private-sector solution for the insurer. I agreed to have Dan Jester stay in New York to help with AIG, and Jeremiah Norton, deputy assistant secretary for financial institutions policy, would fly up to relieve Steve Shafran. I would return to Washington the next morning, while Tim’s team—with no time to rest after Lehman—tried to determine AIG’s liquidity needs and develop a plan to raise money.

I got back to the Waldorf about 10:00 p.m. Shortly after I arrived, John Mack called me. I could tell that the Morgan Stanley CEO was on edge. In just one day, Wall Street had irrevocably changed: Lehman Brothers was headed for bankruptcy, and Merrill Lynch was about to be bought by Bank of America. Morgan Stanley had held up well so far, but with those two firms gone, John was deeply worried.

“Come tomorrow morning,” he said, “the shorts will be on us with a vengeance.”