Stress Test

  It’s never good when the markets won’t take the word of a Treasury secretary, but I didn’t think there was much I could do to change that except back up my words with actions and give investors a chance to judge what we actually did. The wait for the stress test results would be excruciating. We would have to rely on the patience of the markets at a time when patience seemed almost irrational. And there was a very real possibility that the results would be terribly disappointing. We could talk about leaving banks in private hands, but we wouldn’t be able to do that if it turned out they were truly insolvent. The President wanted us to rip off the Band-Aid, but we knew we might not like what we found underneath.

  “There is the risk that our overall approach to large financial institutions will be overwhelmed by events in the most negative economic scenarios,” we wrote in our memo. “We may, by being proactive, be blamed for causing the problems we are seeking to preempt. Further, there is the risk that … we will pull the ‘Band-Aid’ off a wound that we lack the capacity to sterilize and thus exacerbate problems.”

  In other words, even if we executed our plan properly, a depression remained a serious risk.

  “On balance, though, our judgment is that the risks of alternative paths are even greater.”

  At least that was my judgment. I believed our plan was the least-bad option. We just had to roll it out, piece by piece, and hope it could win the confidence of the markets. The good news, although it certainly didn’t seem that way at the time, was that my early struggles had set expectations about as low as they could go.

  EIGHT

  Plan Beats No Plan

  On February 25, the Fed rolled out more details of the stress test that it would administer to the nineteen largest banks. Expectations were low, and we did not exceed them. The general reaction was that it looked like a whitewash, a bogus exam designed to give a clean bill of health to a gravely ill banking system.

  The stress test was supposed to evaluate whether the banks had enough capital to survive a brutal downturn, and the critics didn’t think the Fed’s assumptions about the trajectory of the economy looked brutal enough. They complained that the Fed’s “stressed” scenario for growth and jobs—the economy shrinking 3.3 percent in 2009, unemployment rising to 10.3 percent in 2010—sounded less like a doom-and-gloom scenario than a likely scenario. They turned out to be right about unemployment. But the Fed’s assumptions for housing prices—down 14 percent in 2009 in a likely scenario, 22 percent in the stressed scenario—turned out to be much darker than reality.

  The Fed’s most important assumptions were the loss rates that bank assets would face in the stressed scenario, because those would determine how much capital the banks would need to raise. And those assumptions would be appropriately dark, even worse than the losses during the Great Depression. But the Fed did not release those assumptions at the time. As a result, many analysts—including all three quoted in the New York Times the next day—suspected the stress test would be too weak, an elaborate ruse to produce reassuring news.

  “It sure sounds to me like they are designing this to make it sound like the banking system is in great shape,” one banking expert told the Times.

  That wasn’t true. The Fed would play it straight. We knew that if the stress test results didn’t seem credible, the markets would ignore them and assume the worst. But few outsiders seemed to have faith in the process. Behind closed doors, even Larry suggested the fix was in, complaining that the Fed was too cozy with banks, and that I was too protective of the Fed.

  In fact, we didn’t know how well the financial system was prepared for a severe downturn, so we designed our program to recapitalize it no matter what the stress test found. After the Fed identified how much capital each bank would need to survive the stressed scenario, the banks would have six months to raise enough capital from private investors to plug their shortfalls. If they couldn’t, the Treasury would invest enough TARP capital to make up the difference. Our investments would again be in preferred shares, but they’d be “convertible” into common equity, so recipients would be better prepared to absorb losses.

  There was one serious problem with this approach. The overarching goal of the stress test was to reduce uncertainty, but there would be tremendous uncertainty until the stress test was complete. For months, markets wouldn’t know which banks would need to raise capital and how much they would need to raise. Shareholders wouldn’t know how much dilution they were going to face, and if the government ended up taking equity stakes at extremely low prices, the dilution could be massive. We were already seeing accelerated flight from bank stocks, which was further depressing their prices and the potential dilution investors would face, which in turn was further accelerating flight from bank stocks.

  Our solution to this death spiral was what Lee dubbed “the Geithner Put.” He later teased me that it would be immortalized as the biggest put in financial history, which is why I prefer to think of it as the Kabaker Put; I’m sure Matt would enjoy having a put named after him. In any case, it effectively created a floor under bank stock prices, stopping the death spiral and encouraging private investments in financial firms that otherwise would have looked much too risky.

  What the put did was fix the conversion price that Treasury would pay for stock in a bank just below the stock’s price in the run-up to my February 10 speech, even if the market price dipped much lower. It did not eliminate the risk for bank investors, but it eliminated a lot of the uncertainty that was frightening them, and it reduced the risk that they would get totally clobbered. They might fall a few feet, but unless their bank was in terrible shape, they wouldn’t fall a few stories. The media paid almost no attention to the put, which was just as well; it probably would have been trashed as a giveaway to bank investors. But our goal was to recapitalize the system, and every dollar of capital that private investors could inject would be a dollar that wouldn’t have to come from taxpayers.

  The hedge fund manager David Tepper later told the press that in February 2009, while the world was mocking my speech, his Appaloosa Management fund began buying bank stocks, because he read our public statements and thought our strategy sounded sensible. He continued to increase his investments in Bank of America and other troubled firms after we announced the put, and in 2009 his fund reportedly enjoyed some of the best returns ever recorded on Wall Street.

  “You seemed to be keeping your word,” Tepper told me after I’d left Treasury. “And I figured if you didn’t keep your word, I could sue you.”

  Of course, our goal wasn’t to help Tepper make billions of dollars for himself and his investors. Our goal was to get the economy growing again, and that required stabilizing the banks so they could start lending again. Tepper simply listened to what we said, watched what we did, and bet that we would succeed.

  OUR BASIC strategy was to follow through on what we said we would do, and hope the system would hold together until the stress test was complete and our other rescue programs had time to get traction. But Citigroup and AIG were still on the brink of failure, and the collapse of either one could have made the fall of Lehman look mild.

  With Citi, we faced a dual challenge. The bank clearly needed more capital to survive—and not just any capital, but common equity. Treasury was the most likely source of new capital. But rumors that we intended to nationalize the bank were shaking confidence in the entire banking system, because nationalization would mean steep losses for investors. We didn’t want to let Citi fail, but we also didn’t want to nationalize banks unless absolutely necessary; we had to dispel rumors of a government takeover if we didn’t want a broader run on the system.

  Instead, we settled on a solution where Citi’s private preferred stockholders would convert their shares into common equity, and we would convert an equal amount of the government’s TARP preferred shares. This conversion would enhance Citi’s ability to absorb losses without additional TARP funds. It would also demonstrate that our initial TARP injections of p
referred stock into other firms could be a source of greater firepower in the future. And while it would increase the government’s stake in Citi to as much as 36 percent, we would not take a majority position, reassuring the markets that we wanted to avoid nationalization.

  On February 26, the day after we rolled out the stress test, my exhausted team pulled an all-nighter to finalize the terms of the conversion, while Citi worked with investors such as Singapore’s government fund and a Saudi prince to ensure their participation. By 5 a.m. on the 27th, all that was missing was a signature from Treasury—and since I was still the President’s only Senate-confirmed official at the Treasury, the signature had to be mine. But nobody knew where I was. A young staffer who had been assigned to catch me on my way into the office had gone to sleep on Kabaker’s couch after setting his alarm to go off early. But I had arrived even earlier to work out in the Treasury gym.

  Lee finally found me on a treadmill in the department’s basement. I paused the machine and signed, adding $50 billion in common equity to Citi’s buffer against losses.

  As vital as it was to show we weren’t looking to nationalize systemic firms, it was even more vital to show we wouldn’t let them fail. We had to demonstrate that we were serious about no-more-Lehmans. Otherwise, private investors wouldn’t put cash at risk to help recapitalize the financial system, and we weren’t sure we had enough resources to do it through TARP. When it came to AIG, even after our extraordinary commitments, the markets weren’t convinced we were willing to stand behind the company, and the rating agencies, with their exquisite timing, were threatening new downgrades. On March 2, we announced another $30 billion commitment to help AIG meet its obligations, and the rating agencies agreed to hold off. We were all in.

  The political reaction was predictably fierce. AIG had explored the frontier of recklessness, and taxpayers were outraged that they were repeatedly paying to keep it afloat. That outrage was about to explode.

  ON TUESDAY afternoon, March 10, I got some ugly news. Our TARP team had learned that AIG was about to pay $165 million in bonuses to employees in the Financial Products unit that had helped blow up the firm. The bonuses amounted to less than 0.1 percent of the U.S. commitment to AIG, but it was hard to imagine worse optics.

  We went back and forth in my office trying to figure out whether we could legally block the bonuses, and whether it even made sense to try. Our lawyers thought AIG was contractually committed to paying the bonuses, and they did not believe we had the legal means to block them. I was instinctively skittish about the U.S. government breaking contracts, especially at a time when all sorts of commitments had been called into question. We weren’t Venezuela. And I worried that if the U.S. government tried to impose a solution in one case, even this egregious case, we would send a message that no contract was safe, a destabilizing message during a crisis. As Lee Sachs liked to say: “In a storm, the world needs anchors.” The rule of law was arguably our most important anchor, especially during this limbo period when fears of nationalization and federal interference were pervasive.

  We had dealt with this bonus issue in a more general and theoretical form in early February, before we knew about AIG’s situation. Senate Banking Committee Chairman Chris Dodd, facing a tough reelection battle in Connecticut, had inserted a populist executive compensation amendment into the Recovery Act, capping bonus payments for TARP recipients and empowering Treasury to “claw back” existing bonus arrangements. In discussions with Dodd’s staff, my team had proposed a variety of changes, including a suggestion to remove the retroactive claw-back provision. We thought it would infringe the sanctity of contracts and change the terms of the TARP rescue. Senator Dodd agreed and stripped the provision out of the bill, while rejecting most of our other recommended fixes.

  When the AIG news arrived a month later, I still worried that unless we had a clear legal basis, we could not force a financial firm to violate a contractual obligation without unleashing a new wave of panic and uncertainty. It would send a message that the protections of the law might not apply when politically inconvenient. It could make customers and counterparties wonder whether their contracts might be violated next, sparking runs on other financial institutions, and it could encourage talented employees of TARP-funded firms to flee. It could also make firms less willing to participate in TARP out of fear that we would retroactively change the conditions, weakening the power of our crisis-fighting tools.

  But these concerns had no public constituency. The mess at AIG took the bonus issue out of the realm of theory. It was going to be impossible to explain.

  The public was understandably incredulous that major TARP recipients had expressed their appreciation for their taxpayer-financed rescues by paying out boom-level bonuses in January 2009. And of course, I had helped design the original AIG rescue when I was still at the Fed. At the time, my staff and I had been too consumed with trying to contain the post-Lehman panic to even consider whether we could do anything about executive compensation. We had structured the rescue to distance government officials from managerial and commercial decisions inside the firm; AIG had promised those bonuses well before it began receiving government assistance. But now AIG’s independent decisions were ending up on our doorstep anyway.

  The AIG bonus scandal would take the media frenzy and the public anger to new levels, and I knew it would be another devastating blow to confidence in our crisis response. The next day, I called Ed Liddy, the new CEO of AIG, to say, essentially: What the fuck?

  “This is going to kill us—and you!” I said.

  Liddy said he had tried to renegotiate the bonuses, but his employees refused to accept anything less than they were promised. His attorneys believed that if the bonuses weren’t paid in full, the employees would sue for breach of contract and the firm would be on the hook for as much as triple the amount, including damages. He said he could only try to renegotiate future payouts. I knew that wouldn’t mollify anyone, but I did not believe I had any legal basis to stop contractually promised payments, and I thought the public relations value of trying to intervene would pale in comparison to the damaging specter of the U.S. government trying to break a private contract.

  On the way to a meeting in the West Wing, I stopped by Rahm’s office to tell him about the AIG mess. Rahm was the center of energy and activity in the West Wing, and we had a great relationship. He was constantly checking in with me to hear how the markets were doing, to plot strategy, to get a jump on what might go wrong next. He did not try to micromanage me on policy, and he was very creative about protecting me from other potential sources of interference in the White House. He wanted to keep close track of what we were doing, weigh in on our communications strategy, and make sure there were no unexpected surprises. I wanted that, too. I wanted to expose the White House to our unpalatable choices, and make sure it had a measure of ownership of our decisions, but without dragging the President too far into the muck of the details.

  But there was no way I could protect the White House from this. It was hard to imagine a more politically damaging news story than taxpayer-funded bonuses for the arsonists who set the system on fire. And I told Rahm we didn’t think we had any legal basis to intervene.

  “We don’t really have a choice,” I said. “It’s going to be a nightmare.”

  It was already hard to fathom the mess that had been dropped in President Obama’s lap. Rahm called it “the gift bag.” February was the worst jobs month yet, with unemployment rising to 8.1 percent. The S&P 500 had fallen to new lows the day before I learned about the bonuses, down 57 percent from its peak. The President publicly said it looked like a good time to buy stocks, which struck me as an unwise thing for him to say, although investors who took the financial-adviser-in-chief’s advice would have more than doubled their money in four years. Meanwhile, car sales at Chrysler and GM had fallen in half, a disaster for their workers and suppliers, and an ominous sign for the entire manufacturing sector.

  I was far away that
Saturday when the AIG bonus news became public, attending a meeting of G-20 finance ministers in Horsham, an out-of-the-way town in the English countryside. It fell to Larry to go on the Sunday television shows to defend our position, repeating the word “outrageous” again and again, while trying to point out that we’re a nation of laws. It didn’t matter. The words “AIG” and “bonuses” in the same sentence were enough to convince most Americans that we were on the side of the bad guys.

  HORSHAM WAS as cold as one would expect of England in March, and the chill invaded the stone-walled lodge where the ministers met. I was already run down after months without much sleep, and I caught some kind of flu. On the military plane that took us home Saturday night, I was coughing, aching, and unhappy. The summit had not been a great success; the headlines were mostly about international tensions. And I had to review materials for a meeting the next day with the President and his economic and political advisers, many of whom were deeply skeptical about our financial repair plans. We would have to make the case for staying the course.

  I’m not a screamer, but I got a bit loud that night on the phone with Lee and Kabaker, and not just because the reception over the Atlantic was spotty. I was irked by a PowerPoint presentation that they had negotiated with Larry while I was away. The offending slides confronted the hypothetical of a bank that couldn’t raise enough private capital to fill the hole identified by its stress test: What would we do if we had to inject so much TARP capital that the bank was effectively nationalized? Lee and Larry had agreed to present two options to the President. One was the Treasury plan for a “conservatorship approach,” which would allow us to handle a bank in a way similar to the government’s handling of Fannie and Freddie, so we’d have the discretion to wind down our investments gradually to minimize shocks to the system and losses for the taxpayers. The other was Larry’s “rapid resolution exit,” which would require an immediate restructuring of the bank, including immediate liquidation of bad assets.