Stress Test

  It’s hard to describe the stress of knowing that the world was depending on our plan to prevent a catastrophe, and that nobody seemed to think it was any good. I remember in the Oval Office the morning after my speech, when the President had asked what the hell had happened, my colleagues kind of turned toward me with body language that implied: Don’t ask us, ask Tim. And it really was all on me. All I could do was keep my head down and try to focus on my work. I had started carrying a Buddhist amulet that my aunt Lydia had given me in my wallet, along with a Unitarian prayer to the Spirit of Truth that I found soothing in turbulent times:

  Grant us to feel thy shadow near

  So may we find the strength to stand face forward,

  Courage to walk toward the dawning day.

  They were comforting words, but honestly, my problem wasn’t really strength or courage. My problem was that there was no way to know for sure if our plan would work.

  On February 18, on the way back to Washington after President Obama announced our housing program in Mesa, Arizona, we met in his office on Air Force One. Financial stocks had plunged 20 percent since my speech. The President was clearly nervous that we were heading for a cliff. He wanted to know how bad the losses in the financial system were going to be, how much government help the banks were going to need, and when we were going to find out.

  My response, unwelcome again, was that we would have to wait for the Fed to finish the stress test. We would move forward with our housing initiatives. We would go ahead and launch our other programs to revive the credit markets and invest in distressed assets. But none of that would dispel the uncertainty.

  “We’ll pressure the Fed to speed this up, but we won’t know how this ends until they’re done,” I told him. “It’s scary. It’s risky. And there’s no way we can know if it’s going to work. But we’re stuck letting this play out.”

  HOUSING WAS at the heart of the crisis, and the housing initiatives we unveiled in Arizona would be especially controversial. The President’s announcement, and the perception that we were bailing out deadbeat homeowners, inspired the CNBC commentator Rick Santelli’s infamous rant that day calling for a new Tea Party, galvanizing an antigovernment movement on the right that would scramble the politics of the Obama years. At the same time, the left would view our efforts to address the foreclosure crisis as woefully inadequate, and as the most damning evidence that we cared more about Wall Street than Main Street.

  We couldn’t legally force private owners of mortgages to give borrowers a break during tough times. The situation got especially complex when loans were sliced and diced into multiple mortgage securities owned by a range of investors, including pension funds, mutual funds, hedge funds, and banks. The complexity of this system and the convoluted web of property rights would make it much harder to implement programs to help homeowners. And while we did commit $50 billion in TARP funds for that purpose, we estimated that about eight million homeowners were at risk of foreclosure, and millions more were in distress. Trying to figure out which ones to help was a thorny policy problem.

  Our goal was not to subsidize borrowers who splurged on overpriced McMansions and vacation homes and investment properties, or took out home equity loans to buy swimming pools and fancy cars. We knew that a few outrageous stories of aid to reckless speculators and scam artists could cripple support for our entire housing program. We also wanted to avoid spending billions of taxpayer dollars to restructure mortgages for families who would lose their homes even with government help; inevitably, some innocent victims of the crisis would have to move into cheaper homes or rental properties. And while we didn’t worry excessively about the moral hazard implications of our programs, we had to be careful not to create perverse incentives for less vulnerable homeowners to stop making payments in order to qualify for help.

  Our biggest debate was whether to try to reduce overall mortgage loans or just monthly payments. The main problems with loan reduction were its huge up-front expense and minimal bang for the buck; the benefit to homeowners would be spread out over the life of their mortgages, when we wanted to maximize the benefits we could provide immediately. Some proposals for broad-based principal reduction for underwater homeowners seemed particularly wasteful, since roughly three-fourths of those families were still current on their mortgages. Homeowners who couldn’t afford their payments were the imminent foreclosure risks. And theoretically, payment reductions could benefit creditors as well as borrowers if they helped get mortgage payments flowing and avoid defaults. Most creditors weren’t eager to get stuck with millions of foreclosed homes after a historic real estate bust, when a glut of inventory was further depressing resale values.

  So we decided to focus our program on reducing monthly payments, rather than the overall principal, for at-risk homeowners. It felt like the least-bad option. We thought there were as many as three to four million borrowers who would be likely to face foreclosure without government help but might be able to stay in their homes with more affordable payments. The banks and investors who held their mortgages could cut their losses, too; recovery from foreclosure is typically only half the loan size, and was less than that during the crisis. Larry did point out that debt crises rarely end before governments help reduce excess debt burdens, but nobody had a feasible proposal for a cost-effective, well-targeted, large-scale debt reduction program for homeowners that could get through Congress.

  The big news out of Arizona was our Home Affordable Modification Program, which provided incentives for mortgage servicers to reduce payments on owner-occupied homes down to 31 percent of the borrower’s income. HAMP was modeled in part on work the FDIC had done with IndyMac’s mortgages, but with more of a focus on rewarding successful modifications. We thought it was narrowly targeted for maximum effectiveness at minimum cost. But Santelli and the conservatives who responded to his Tea Party rant thought we were transferring their hard-earned tax money to profligate freeloaders. Meanwhile, liberals would spend the next several years citing our overly optimistic estimate that HAMP would help as many as three to four million homeowners as evidence of the failure of our housing initiatives.

  In addition to HAMP, the President announced the Home Affordable Refinancing Program, or HARP, an effort to help millions of borrowers refinance their mortgages. He also called on Congress to pass “cram-down,” a borrower-friendly bill that would have allowed mortgages to be restructured in bankruptcy court, but Rahm would soon inform him we didn’t have the votes even for the very narrow version being kicked around in Congress, despite Democratic control of the Hill. I didn’t think cram-down was a particularly wise or effective strategy, anyway. The bankruptcy courts were already overwhelmed, and the bill had the potential to push up mortgage costs for all borrowers—though perhaps only slightly given its limited scope—which would have further weakened the recovery.

  The largest and most important housing initiative we unveiled in Arizona, although it wasn’t widely viewed as housing policy at the time, was a new $200 billion capital commitment for Fannie and Freddie, which were burning through the $200 billion they got during Hank’s tenure and were again under fire in the markets. Along with the Federal Housing Administration, they were virtually the only remaining sources of housing finance, essentially keeping the mortgage market afloat. Stabilizing them would be vital to stabilizing the broader housing market—and to keeping mortgage costs low, so that homeowners could enjoy the tax-cut-like benefits of refinancing.

  Our lifeline to Fannie and Freddie would be pilloried as another wildly expensive financial bailout, but it wouldn’t require additional congressional action or TARP money, so it would be relatively easy to execute. At a time when home prices had already fallen more than 30 percent, and futures markets were expecting another 20 percent decline, our $200 billion commitment would help steady a $20 trillion real estate market, while helping millions of homeowners to refinance. Saving Fannie and Freddie not only avoided catastrophe, defusing two of our financial bombs; i
t laid some groundwork for recovery.

  THAT LEFT three bombs still ticking. The most worrisome, yet again, was AIG.

  The Fed and the Treasury had sunk $150 billion into the company, but it was starting to look like a bottomless pit. AIG was about to report a mind-boggling loss of $61.7 billion in the fourth quarter of 2008, about a billion dollars per business day. It was once again on the brink of a ratings downgrade, which would trigger tens of billions of dollars in margin calls, bringing the firm back to the edge of a catastrophic default.

  Government Programs Helped Stabilize Home Prices

  S&P/Case-Shiller 10-City Composite Home Price Index

  In January 2009, home prices had already fallen 30 percent and were projected to plunge 20 percent further. But actions by the government and the Fed to support the economy—and specifically to support Fannie and Freddie, hold down mortgage rates, and help prevent foreclosures—helped stabilize prices in 2009 and facilitated a recovery of prices that began in 2012.

  Sources: Bloomberg and U.S. Department of Housing and Urban Development.

  On the one hand, we needed to salvage AIG to prevent a return of the panic. On the other hand, AIG increasingly looked unsalvageable. We had a “blue team” analyzing the best way to pump more money into the firm to stop the bleeding for good. But we had just tried to fashion a permanent solution for the company three months ago. So we also set up a “red team” to study if it would be possible for AIG to go into bankruptcy without mass contagion. When I polled both teams at a February meeting in my small conference room, there was growing support for letting AIG go bankrupt, especially if AIG could separate its rogue Financial Products unit from its healthier insurance subsidiaries—or, as somebody suggested, “take FP out back and shoot it.”

  One of the investment bankers whom Hank had hired at the end of 2008 to run TARP suggested the world might not end if AIG collapsed.

  “I remember when Lehman filed, and the next week, life went on,” he said. “Joe Schmo went to the ATM and didn’t even notice it.”

  I was a bit stunned to hear that version of history. Lehman’s filing had sent the worst shock through global markets in generations. It had taken trillions of dollars in guarantees by the U.S. government to break the panic, to make sure average Americans could still withdraw cash from ATMs when they wanted. And I thought AIG, the insurer of choice for troubled financial institutions and financial instruments, still posed even greater systemic risks. Perhaps I had some post-traumatic stress disorder from Lehman, but only because Lehman had been truly traumatic.

  “You may not have felt anything,” I replied. “But for those of us who saw it up close, believe me, we’re not doing that again.”

  I had just pledged in my speech that the failure of systemic firms was no longer an option. Now we had to prove we intended to keep our commitments. Some red-team members suggested we could argue that AIG was an exceptional case, so letting it go would not reflect on our commitments to other firms. I asked if they were absolutely certain the world would believe us, because otherwise an AIG collapse could tear down the system. If we pulled the plug on AIG after investing such extraordinary resources in its survival, we would send a powerful message that our words didn’t mean much, that we couldn’t or wouldn’t act to save systemic firms, most likely sparking a run on the remaining weak links.

  “If we renege on AIG, Citi and Bank of America are next,” I said. “Until someone can show me there’s a firewall high enough to protect the rest of the system from AIG’s failure, we’re not going there.”

  This is a classic problem in crisis response. The overwhelming temptation is to let the most egregious firms fail, to put them through a bankruptcy-type process like the FDIC had for community banks and then haircut their bondholders. But unless you have the ability to backstop every other systemic firm that’s in a similar position, you’ll just intensify fears of additional failures and haircuts. We didn’t have that ability. Lee used to say it would be nice if we were the Adam Sandler character in the movie Click who could freeze time with a remote control. Then perhaps we could wind down AIG in an orderly fashion, and persuade the markets there was no need to panic. But in the real world, people react in real time, in dynamic and uncontrollable ways that reflect their perceptions and fears. We could hope our decisions about AIG would have minimal impact on other firms, but we couldn’t be sure its failure wouldn’t topple the entire financial system.

  Citi was already under siege, as was Bank of America. On Friday, February 20, Citi’s stock price dropped below $2. The cost of insuring BofA’s bonds rose 15 percent. The markets had lost confidence in the quality of their capital, a slightly technical but extremely serious problem. Citi had enough “regulatory capital” to meet the requirements we had imposed at the Fed, but less than 20 percent of that capital was common equity from shareholders, the highest-quality capital, the permanent investments that can absorb losses and protect a troubled firm from insolvency. If you counted only common equity, Citi and BofA were now two of the most highly leveraged banks. Citi had more than $60 in assets for every $1 in top-flight capital, while BofA’s ratio was nearly 40:1. And markets had stopped trusting lower-quality capital, even the preferred investments we made through TARP.

  This was mostly foreseeable, and I should have paid much more attention to Citi’s lack of common equity while I was at the Fed. Even the capital we pushed Citi to raise in late 2007 and 2008 came in the lower-quality forms that investors now found meaningless. But now we had a more immediate problem. Rumors were swirling that we were going to nationalize Citi and BofA, so investors, fearful of getting wiped out, were running from them.

  Markets Were Very Concerned About Large Financial Institutions Failing

  Average Credit Default Swap Spreads of Six Large Institutions

  This market measure of the default risk of large financial institutions’ senior debt increased sharply after Lehman’s failure. While it receded temporarily following the initial TARP capital injections in the fall of 2008, confidence in the biggest banks only began to recover after the stress test results and the cumulative effects of our other policy initiatives in 2009.

  Source: Bloomberg (represents the unweighted average of five-year CDS spreads of Bank of America, Citigroup, Goldman Sachs, JPMorgan, Morgan Stanley, and Wells Fargo).

  We were, in fact, preparing to take a larger stake in Citi to address its capital problems, but we were hoping to keep government ownership below a majority stake if at all possible. It was also true that inside the administration, we were discussing the possibility that bad stress test results could force us to take substantial equity stakes in some banks, perhaps amounting to effective nationalization. In our internal strategy discussions, some participants—at times including Larry and his adviser Jeremy Stein, his former Harvard colleague—suggested that widespread nationalization was inevitable and even necessary. Nationalization was a threatening word for investors in a market economy, implying that shareholders could be wiped out, creditors could be haircut, and politicians could be taking control of private firms.

  The proponents of nationalization inside the administration were indeed pushing to impose losses on bank creditors, arguing that otherwise we would never have enough TARP dollars to recapitalize the system. External voices were no more reassuring. Even the world’s most famous free market advocate, Alan Greenspan, declared in February that “it may be necessary to temporarily nationalize some banks in order to facilitate a swift and orderly restructuring.” As the markets started hearing rumblings that we too were talking about nationalization and haircuts—sometimes from White House officials who brain-stormed with academics and market participants—bank stocks fell and default risks increased.

  This was deeply damaging at an exceptionally fragile moment. After outlining our programs before we could spell out the details, we were extremely vulnerable to speculation about our true intentions. The fear that we might re-create the conditions that unleashed the panic
of the fall was undermining everything we were trying to do to put out the remaining fires. Lee kept saying he could feel the markets tremble with every exploratory call to New York from a White House staffer. There was an element of self-fulfilling prophecy at play. If the markets thought the administration thought the banks were doomed, the markets could doom the banks.

  Of course, there were still a lot of good reasons to be worried about the world, no matter what was leaking out of our meetings. That Sunday, Larry and I sent the President an unbelievably dark memo. We noted that of the nation’s four largest banking institutions—Citi, BofA, Wells Fargo, and JPMorgan—JPMorgan was the only one that wasn’t “in distress.” We said there was “a significant chance” that Citi and BofA would end up in the hands of the government. We warned that AIG was not only a devastating problem in its own right, it reflected deeper problems in its industry that imperiled similarly massive life insurers such as MetLife and Prudential. And we calculated that after financing the TALF credit facility, the PPIP investment funds, our foreclosure relief programs, the auto industry rescue, and whatever emergency interventions we’d need while the stress test was under way, we would likely have less than $100 billion left in TARP to recapitalize the banks.

  “That is surely not enough,” we wrote.

  We would end up including a $750 billion placeholder for additional financial rescues in our first budget, more than we would request for the Pentagon.

  But our first task, we told the President, was “immediate reassurance.” The world had listened to my speech, and it hadn’t believed my no-more-Lehmans promise. Our pledges to try to keep the banking system private hadn’t seemed credible, either. We decided that before the markets opened Monday morning, the regulators would need to sign in blood on another joint statement, basically repeating the never-again message I had failed to get across in my speech two weeks earlier. “We reiterate our determination to preserve the viability of systemically important financial institutions,” it said. In other words: No, seriously, we mean it. “The strong presumption … is that banks should remain in private hands,” the statement concluded. In other words: please stop running away from Citi and BofA.