Stress Test

by Timothy F. Geithner

ON FRIDAY, July 11, Americans saw an actual bank run—not a metaphorical run, like the digital withdrawals that had crushed Bear, but a physical run on a physical bank, as in It’s a Wonderful Life. That afternoon, the Office of Thrift Supervision and the FDIC shut down and seized IndyMac, a California thrift that was once part of Angelo Mozilo’s Countrywide empire. IndyMac had flourished during the bubble by providing exotic mortgages to buyers without much in the way of income or assets. Its balance sheet was loaded with option adjustable-rate mortgages (ARMs), an almost comically irresponsible product that let borrowers choose their monthly payments, adding to their future obligations if they wanted to pay less at the moment. When the housing bubble popped, IndyMac popped with it, the largest U.S. bank to fail since the savings-and-loan crisis of the 1980s, though not even one-tenth the size of Bear Stearns. In Pasadena, TV cameras filmed tearful depositors lining up outside its locked doors, screaming for their money back.

  The FDIC guaranteed deposits up to $100,000 at thrifts, so most of those panicked account holders had nothing to worry about. But depositors with more than $100,000 in their accounts were legitimately desperate to retrieve their cash. The FDIC paid uninsured depositors only half of what they were owed, in order to limit the losses to its national fund for insured deposits. This was standard FDIC procedure for a bank failure, but it illustrated how in a time of extreme stress, imposing haircuts on unprotected investors can accelerate the very panic you want to contain. The week after the run on IndyMac, depositors pulled more than $1 billion a day out of Washington Mutual, a much larger thrift with similar exposure to risky mortgages. They had seen the IndyMac precedent, and they didn’t want to get stuck with haircuts if WaMu failed.

  The FDIC’s approach to haircuts would inspire fierce debate when WaMu stumbled in the fall. For the time being, though, IndyMac’s indelible images of fear were just more evidence that things were going from bad to worse.

  THE FINANCIAL system could easily absorb the $30 billion collapse of IndyMac. There was no way it could absorb the collapse of Fannie Mae and Freddie Mac. The two government-sponsored enterprises held or guaranteed more than $5 trillion in mortgage debt. They were funding about three of every four new U.S. mortgages, propping up what was left of the housing market. But they were heading for the abyss. Fannie’s stock price plunged to $10.25 the day IndyMac failed, down 90 percent from its peak. Just about everyone except their captured regulator agreed they were woefully undercapitalized. One Wall Street analyst calculated that they had a capital shortfall of $75 billion. Despite Hank’s pleas, Fannie had raised only $7.4 billion in new capital in 2008, while Freddie had failed to raise a dime.

  I had been wary of Fannie and Freddie ever since I had watched the Clinton Treasury and Greenspan try without success to rein in their leverage. At the Fed, I had expressed similarly ineffective concern about their paper-thin capital buffers. Closely entwined with the government since birth, they were the most dangerous example of moral hazard in the financial system. Fannie and Freddie borrowed at artificially low rates, because markets assumed the government would never let them default, and poured the cash into mortgages, mortgage guarantees, and other highly leveraged bets on the U.S. housing market. They enjoyed access to cheap money, like banks, without the tougher constraints applied to banks.

  Now those highly leveraged bets, so profitable for their shareholders and executives during the boom, were threatening to drown Fannie and Freddie in losses. The underwriting for the mortgages they bought and guaranteed was more conservative than the private industry average, but in recent years, under pressure to increase returns to shareholders, their standards had eroded. They had also built up a huge portfolio of mortgage securities, including some backed by much riskier subprime loans. With remarkably thin capital buffers, Fannie and Freddie were acutely vulnerable to a nationwide housing swoon, as well as a recession that was leaving more mortgage holders without jobs and without the ability to make their payments.

  But with private lenders and investors now in full retreat, the housing market was as dependent on the GSEs as the GSEs were dependent on the housing market. As Hank pointed out, when it came to housing finance, Fannie and Freddie, along with the Federal Housing Administration, were essentially the only games in town. If they pulled back in order to reduce their leverage and husband their capital, the death spiral of falling home prices, mounting foreclosures, poisoned mortgage securities, and financial turmoil would intensify. Somehow, Washington needed to restore confidence in Fannie and Freddie. That would require government money.

  That July weekend, after months of failed efforts to push Fannie and Freddie to raise more capital on their own, Hank decided he needed to act before doubts about their viability became self-fulfilling. On Tuesday, he asked Congress for legislation that would give the Treasury almost unlimited authority to invest in Fannie and Freddie, give their regulator the power to take them over, and give the Fed a consulting role so we could dig into their books. I thought the bill’s substance was excellent, but the politics were brutal. Senator Bunning scoffed that it made the Bear Stearns deal look like “amateur socialism.” Hank was also ridiculed for urging Congress to give him immense and unprecedented powers so that he wouldn’t have to use them.

  “If you’ve got a squirt gun in your pocket, you may have to take it out,” Hank explained to Congress. “If you’ve got a bazooka, and people know you’ve got it, you may not have to take it out.” That’s true, but when you talk about your need for a bazooka, people naturally assume you must face a serious threat to your security.

  Nevertheless, by the end of July, the Democratic-controlled Congress overwhelmingly passed the bill, providing an almost blank check for a Republican Treasury secretary. The legislation also quietly raised the debt ceiling, a routine housekeeping measure that would become anything but routine in the years to come. But while Congress gave Treasury vast financial authority to inject capital into Fannie and Freddie, it denied Treasury any power over the management of the firms. That power went to their regulator, rechristened the Federal Housing Finance Agency. This awkward limitation, designed by Democrats to protect the new agency from political interference by a Republican president, would prove consequential later in the crisis when a Democratic president wanted to limit the damage to homeowners.

  The legislation came close to formalizing the implicit federal guarantee behind Fannie and Freddie, but it couldn’t improve their performance or their balance sheets. They soon announced more than $3 billion in second-quarter losses. Tim Clark, who led the Fed’s dive into their books, concluded that their loan loss projections were half what they should have been. He thought much of their capital was an accounting fiction. His unvarnished assessment was that Fannie and Freddie were functionally insolvent.

  Hank was going to have to use his bazooka after all.

  ONE AFTERNOON that summer, I tried to lighten up the mood at the New York Fed with an impromptu contest for the best metaphor for what was happening to the financial system. “I’ve heard ‘the wheels coming off the bus,’ ” I said. “We’ve talked about the engines falling off the plane.” The usual suspects were wildfires and earthquakes, hundred-year storms and hundred-year floods. We also discussed cancer and contagion, sweaters unraveling and boulders rolling down a hill. I relayed one I had first heard from Goldman Sachs CEO Lloyd Blankfein: “The rivets are coming off the submarine.”

  Whatever the metaphor, things looked bad. Wachovia, the fourth largest U.S. bank, announced an $8.9 billion second-quarter loss. Thanks to some ill-fated financial ventures, the insurer AIG was up to $18.5 billion in losses over nine months. Both of those firms had ousted their CEOs, but neither had regained the confidence of the markets. Overall, financial institutions had already written down $300 billion in losses, and our research division estimated $650 billion in losses still to come, with the possibility of as much as $1.5 trillion.

  Risk aversion was on the rise. The private equity investor Tim Collins told me he was mo
ving all his liquid assets out of financial institutions and into TreasuryDirect, a program that lets individuals hold Treasuries directly in a government account. He didn’t want even the slightest risk that his securities could get stuck in a failed bank, another textbook example of flight-to-safety panic.

  THERE WAS nothing fun about that summer. I got a nasty case of poison ivy, so I worked for weeks with my legs slathered in Calamine lotion and wrapped in gauze. I once walked from my office to a conference room with a long train of Calamine-covered gauze trailing from my pant leg. Carole and I also had a horrible scare when Elise came down with dengue fever in a remote village in northern Thailand. Fortunately, we had family friends from my high school days who lived in Chiang Mai and helped her navigate the local hospital scene. But it was a painful, helpless feeling to know that she was so sick and so far away.

  My main financial-world anxiety—and the market’s, too—was still Lehman Brothers. Fuld kept asking me what I was hearing, but institutions were not lining up for the privilege of investing in his firm. Hank and I tried to encourage some interest from Bank of America, but CEO Ken Lewis, who had bought Countrywide despite its glut of toxic mortgages, told us he wasn’t interested in Lehman. Fuld thought he could raise capital from the Korea Development Bank, but we had a hard time imagining the Koreans pulling the trigger. Meanwhile, banks, pension funds, and other institutions kept reducing exposure to Lehman.

  That summer, Lehman and other weak firms frequently complained to me that stronger creditors were preying on them, demanding more collateral when they could least afford it. Fuld was particularly incensed about JPMorgan’s margin calls. AIG’s new CEO, Robert Willumstad, accused Goldman Sachs of being too aggressive in marking down the value of bonds it held that AIG had insured. I called Dimon and Blankfein to tell them not to overdo it, but I had no evidence that they were overdoing it. They seemed to be responding sensibly to the declining value of securities that Lehman and AIG held or guaranteed, and to the market’s loss of confidence in those firms. I couldn’t blame them for acting to protect their own firms.

  Another common request I got that summer—from Fuld and others running or representing the weaker firms—was to reassure the public that there was no cause for alarm, that the financial system was safe and well-capitalized. But neither of those statements was true. I thought pretending otherwise to try to jawbone the markets would damage the Fed’s credibility as well as mine. At that point, happy talk would have seemed so defensive and outlandish that I feared it would just convince the few remaining optimists that all hope must be lost.

  I felt like I was watching a disaster unfold in slow motion, with no ability to prevent it and weak tools to limit the damage. Neither the Fed nor the Treasury had authority to inject capital into troubled institutions, except Hank’s new power to invest in Fannie and Freddie. We had only limited tools to defend against a run on firms outside the commercial banking system, at a time when running seemed increasingly rational. The flood had already breached the levees, and all we could do was pile up more sandbags.

  Throughout the crisis I often thought about the Serenity Prayer: God, grant me the serenity to accept the things I cannot change, the courage to change the things I can, and the wisdom to know the difference. It helped me to focus on what we could do, rather than obsess about what was beyond our powers. But I also thought about what I had said to Rubin during the Asian crises: Just because a problem has no apparent solution doesn’t mean it isn’t a problem. The summer of 2008 did not feel like a time for serenity about the things we could not change. And I was not serene.

  I remember in August, when I was driving home after visiting my parents, I pulled off Interstate 95 at an exit in Warwick, Rhode Island, to finish a call to Rubin about the perilous state of the financial system. I don’t remember the conversation itself—it’s lost in the fog of war—but whenever I drive past that Warwick exit, I get a wave of the same crushing fear and nausea I felt that summer.

  I WAS in the Adirondacks over Labor Day weekend, spending time with my family and trout fishing with Paul Volcker and Tim Collins. I took up fly-fishing late in life, and I don’t do it much, but it’s the most calming activity I know. It requires total focus. It blocks out the rest of the world, which was definitely a bonus that weekend.

  I felt a bit guilty that I had gone fishing, because Hank had asked me to come to Washington to help him plan a resolution for Fannie and Freddie. They were dead men walking, struggling just to finance themselves. Foreign governments and other investors who had assumed their paper was as safe as Treasuries were screaming for U.S. government protection. But this was a Treasury operation, and I didn’t think Hank needed me in the war room. He sounded worried that I was distancing myself for political reasons, but that wasn’t it. I knew we were all in this together. I just wanted some downtime to hang out with my family and enjoy the rhythm of casting for trout before the crisis consumed everything.

  Hank and his team did a fantastic job. By Monday morning, September 8, the Federal Housing Finance Agency had forced Fannie and Freddie into conservatorship and replaced the CEOs. Treasury also committed to backstop the firms with up to $200 billion in government capital, easing fears that Fannie and Freddie would default. That meant lower borrowing costs for the firms, lower mortgage rates for the public, and the removal of an existential threat to global finance. Hank took a lot of grief for firing the bazooka so soon after telling Congress he wouldn’t have to, and his reversal did give the impression that we were lurching from emergency to emergency without a comprehensive plan. But he did the right and courageous thing, heedless of the political costs. And President Bush backed him all the way. I was deeply troubled by many of the Bush administration’s economic policies, particularly its legacy of fiscal profligacy, which would complicate our later efforts to defuse the crisis and revive the economy. But I admired the President’s willingness to support Hank’s strategy when it wasn’t popular.

  The reaction to Fannie and Freddie quickly made the backlash over Bear look mild. Senator Bunning, who had said the Bear deal’s assault on free enterprise made him feel like he lived in France, now said he felt like he lived in China. Senator Obama and the Republican presidential nominee, John McCain, both expressed outrage about public rescues of private firms, although they didn’t directly criticize what Hank had done. McCain and his running mate, Sarah Palin, wrote a Wall Street Journal op-ed titled “We’ll Protect Taxpayers from More Bailouts.” Obama’s campaign put out word that he didn’t want a taxpayer-financed rescue of Lehman, which was also the emphatic consensus of both parties in Congress.

  The economy was clearly deteriorating, with unemployment up to 6.1 percent, and no politician wanted to get on the wrong side of rising populist anger. I still had the luxury of laboring in relative obscurity, but as the public faces of the crisis response, Hank and Ben could not avoid the political arena. Hank’s aides were pressing him to draw a line in the sand against bailouts. Some of Ben’s advisers also wanted him to correct impressions that the Fed’s money store was open.

  That was a problem, because there was no chance a crisis this huge would be solved without putting more public money at risk. And Lehman was on the edge of the abyss. On Tuesday, September 9, after word leaked that the Koreans had lost interest in investing, Lehman’s stock price dropped another 45 percent, while the cost of insuring its debt increased almost 50 percent. We had hoped the Fannie and Freddie rescues would buy Lehman time, but they clearly spooked some investors who hadn’t realized the full gravity of the situation, accelerating the flight from Lehman. I interrupted Hank’s lunch that day to tell him Lehman looked doomed. He asked if I thought it could last the week. I said probably, but the markets would need to see that we were working on a solution.

  Hank said he’d reach out again to Bank of America, even though Ken Lewis had previously sounded dubious about Lehman. I mentioned that Merrill’s John Thain and a few other market types had raised the possibility of a Long
-Term Capital Management–style consortium of private firms helping out, even though I was doubtful that could work. Lehman was much larger than LTCM, and it would need much more money than LTCM had needed a decade earlier; the firms that would have to step up were also in much worse shape in a much worse economy. But dubious options were better than no options.

  The stock market took its biggest hit since the start of the crisis that Tuesday, with financial firms such as Merrill, WaMu, and AIG getting pounded. In July, Bob Willumstad had visited the Fed and danced around the issue of whether we might be able to help if AIG’s liquidity ever dried up; I had seen no reason to extend the privilege of Fed liquidity to an insurer. He came back that Tuesday to make his request for Fed help more explicit, this time with much more urgency in his voice. He emphasized that major Wall Street institutions hedged their risks through credit default swaps and other insurance contracts with AIG. At the time, I still thought it was almost inconceivable that the Fed would ever help out a troubled insurance company.

  LEHMAN WAS beyond troubled. On Wednesday the 10th, Fuld tried to reassure the markets by preannouncing the firm’s third-quarter earnings, but its $3.9 billion loss reassured no one. Markets especially hated Lehman’s skeletal plan to spin off its real estate holdings, as if it could just stick its overvalued investments into a SpinCo box and proceed with business as usual. Fuld didn’t seem to realize the endgame had begun. We still hoped to find a last-minute buyer, but my team began drawing up a Lehman liquidation game plan, drawing on our foam-on-the-runway work over the summer on how to cushion the damage from a failed investment bank.

  Bank of America had agreed to give Lehman another look, but it hadn’t even sent a due diligence team. Ken Lewis was in a dispute with the Fed related to his Countrywide purchase, and even after Ben promised to deal with it, Lewis told me he wouldn’t even look at Lehman without assurances in writing. That seemed like an obnoxious demand at a time like this. “If you don’t believe the word of the chairman of the Fed, we have a larger problem,” I told him. Lewis agreed to send his team.