by Sheila Bair
Buffett was right. But the truth is, I had no place to turn to find allies to fight these ill-conceived ideas. I had hoped that the new administration would be tougher on the banks. Instead, it was bailing out all the big guys. That meant that neither the OCC nor the Fed would have to suffer the embarrassment of an FDIC resolution of any of the big banks they regulated. The little banks would be subject to our harsh process if they failed, but who cared about them? No one at Treasury, the Fed, or the OCC was going to challenge that. I was alone. There were a few token conversations about moral hazard, and we tried to include language suggesting that the government would support only new investors. But Tim was going to get what he had always wanted—the propping up of all the big banks. Some shareholders would have to accept a reduction in the value of their shares, but big institutional bondholders, who should have been subject to the discipline of the market, would be fully protected.
So I put my hopes into at least getting a bad-bank facility up and running to force the sick banks such as Citi to sell off their assets and take their losses, even if the government would backstop them with capital to keep them open. And I instructed my FDIC troops to do everything they could to keep the OCC and the Fed honest in applying the stress tests rigorously and evenhandedly to all of the nineteen banks. I didn’t want to see the process manipulated to help “favored banks.” Jason Cave was my point person, helped by John Corston and his team of examiners, with Art Murton and his economists providing analytical support. But the stress tests were of the bank holding companies, not the banks, so the Fed—as the primary regulator—had the lead. We didn’t even have backup authority for those holding companies (a situation I remedied later in Dodd-Frank). But Jason et al. pushed anyway. We argued for more aggressive loss rates. We argued for harsher stress scenarios.
When the preliminary results from the stress tests came back in mid-March, they matched what we had expected to see: Citi (even with its $45 billion in government capital) and a few others would be insolvent under the stress test assumptions. Citigroup still needed tens of billions of dollars’ more capital. However, if it raised that amount of additional capital, it would lose a very valuable tax break called a deferred tax asset (DTA). Citi could raise only a few billion more in additional capital before it would lose its DTA. That DTA was worth about $50 billion to Citi.
The preliminary numbers were rough, and everyone agreed that they needed more analysis. But as the Fed continued to review and refine the numbers, an interesting pattern emerged: the amount of capital Citigroup needed to pass the stress test was getting lower and lower. In fact, in some cases, the Fed and the OCC were using numbers regarding likely losses that were more optimistic than those of the better-managed banks. In one situation104, they assumed losses would be only half of what one of the stronger banks was estimating.
The Fed and OCC decided to give the banks such as Citi credit if they planned to sell assets as a way to improve their capital ratios, even though the banks had no firm, legally enforceable agreements to sell them. Citi was also given special credit for its ring-fenced assets. And finally, the Fed and OCC made some very optimistic assumptions about Citi’s and other banks’ future earnings growth. The result105: when the final stress test results were announced on May 7, lo and behold, Citi’s capital need was $5.5 billion—just below the amount it could raise without adverse tax consequences.
The total amount of capital required by the stress tests for all nineteen banks was $75 billion, $33.9 billion of that attributable to Bank of America, which was reeling from its overpriced purchases of Merrill Lynch and Countrywide. But the incongruous results for Citi stood out like a sore thumb. Indeed, Wells Fargo, one of the best-managed banks in the country, was told to raise $11.5 billion, twice as much as Citi. And much to our disappointment, the announcement of the stress test results did not include any firm commitments to review the managements and boards of directors of the institutions that needed to raise capital. I issued a separate statement106 announcing that I looked forward to working with the Fed to review “capital plans and corporate governance structures” at the impacted institutions and put the other regulators on notice that we would continue to push for management changes at the weaker banks.
We did prevail on some important issues. For instance, with Ben Bernanke’s help107, we were able to raise the banks’ common-equity requirement—the highest-quality capital—to 4 percent under stressed conditions, as opposed to the 3 percent favored by the OCC. Again with Ben’s help, the Fed publicly committed to consult with us on the evaluation of each bank’s capital plan. Given our hundreds of billions of dollars of exposure to those holding companies with our debt guarantee program, it was amazing that we even had to fight for that, but we did.
All of the institutions, with the exception of GMAC and Citi, were able to immediately raise additional capital from the private sector to comply with the stress test requirements. In that sense, the stress tests were a success. But I fear the main reason investors were willing to buy more shares of stock in those banks was not that they had confidence in our stress tests; rather, it was that the government pretty much said it wouldn’t let any of the banks fail. It was a reaffirmation of too big to fail, taken all the way down to institutions with $100 billion in assets or more.
And what happened to PPIP, our bad-bank legacy loan program? On March 23, 2009, Tim had announced the program over my objections. As discussions over the structure of the PPIP had progressed, it had become clear that Tim and his senior advisers were primarily interested in setting up programs that would allow banks and other financial institutions to sell securities that they held on their balance sheet at market value. Because the PPIPs would provide favorable government financing to investors to purchase those securities, they could produce prices above the current market values. Thus, selling institutions might be able to book a profit.
Treasury wanted to exclude the FDIC from any involvement in the PPIP for securities; they would be run by the Fed and Treasury. Tim wanted the FDIC to handle only PPIPs for loans. As previously discussed, accounting rules allow banks to hold loans at book value (essentially the unpaid principal balance). That value is typically much higher than a loan’s value if it is sold on the open market. Because the PPIP would require banks to sell their loans at market prices, as established by a competitive bidding process, the banks would most certainly take big losses in selling them. It would take real strength of will for the government to force the banks to sell the troubled loans; Tim had no stomach for it. The PPIP for securities was a potentially lucrative opportunity for banks and investors, and Wall Street was licking its chops to get in on the action. I was fearful108 that if the loan program and these now-separate securities programs were announced together, the public reaction would be adverse.
I was right. Commentators and liberal pundits such as Paul Krugman attacked the programs as more government giveaways, while more thoughtful analysts such as Mark Zandi were favorable, recognizing that if we were to avoid our own “lost decades,” à la the Japanese experience, we would have to find a way to get the toxic loans off banks’ books. But the worst attack on PPIPs came from New York Times columnist Andrew Ross Sorkin.
In early April, Sorkin contacted our press office to say he was doing a story on the Public-Private Investment Program and wanted to interview me. Consistent with our open-door policy, I agreed. The PPIP was the subject of much misunderstanding, and Sorkin had a reputation for understanding market mechanics. I was hopeful that we could get a dispassionate piece from him explaining the program.
Our staff took substantial time to conduct a briefing for him on the PPIP. Then he interviewed me by phone on the afternoon of April 6. I thought the interview had gone fine. He hadn’t seemed hostile to the program, and I expected to get a fair and balanced story. However, when Sorkin’s column came out the next day, it was very negative—not just about the program but about me personally. (Notably, it said nothing about the Treasury-run progra
m for securities; he bashed only our program for loans.) The column started snidely109 with the notion that the FDIC was supposed to do only the “simple job” of insuring deposits, ignoring our long record of cleaning up troubled loans. It accused me of saying that the PPIP had “no risk”—which I had most certainly never said—and, in a particularly low blow, compared me to Joseph Cassano, the head of the Transaction Development Group at AIG. It cited “unnamed sources” attacking the program and repeatedly and erroneously said I was putting taxpayers at risk, giving only passing reference to the fact that we were funded by the industry.
I had flown to Kansas City immediately after the Sorkin interview to hold a town hall meeting the next day on deposit insurance. Because of personal threats I had received after the WaMu failure, I traveled with a security detail provided by the FDIC inspector general (usually Agents Matt Alessandrino and Gary Sherrill, who I owe for keeping me safe and secure throughout my term). Two IG agents picked me up at my hotel at 5:30 A.M. to take me to some local morning news shows. Andrew Gray was sitting in the backseat of their SUV as I climbed in. He handed me a coffee and a sheaf of newspapers, with The New York Times on top. He looked grim. I read the Sorkin piece and couldn’t believe it. I felt as if we had been misled. Sorkin had not shown his hand during the interview. Usually when reporters are going to take a negative angle, they tell you up front.
I was stressed and sleep-deprived. Yes, I’ve been known to use an expletive now and again, though rarely and then only in front of close aides. However, I let ’er rip after reading Sorkin’s piece, and let me take this public opportunity to apologize to those two IG agents who were sitting in the front seat of the car when I unleashed my four-letter-word diatribe.
Andrew, as upset as I was, emailed a somewhat stern protest to Sorkin, who read it on his BlackBerry while he was appearing on Morning Joe and announced on air that Andrew was threatening him (he wasn’t). Later that morning, I got on the phone with Sorkin and his editor, Larry Ingrassia, and I didn’t hold back. We pointed out the blatant inaccuracy that I had said that the program was “no-risk” as well as the over-the-top comparison of me to the guy who had driven AIG into the ground. Sorkin didn’t have any good responses, but he also didn’t acknowledge that he had done anything wrong. Exhausted, I went from my heated exchange with Sorkin and Ingrassia to sitting down for a previously scheduled interview with Alison Vekshin of Bloomberg News, and then appearing at a public roundtable discussion on deposit insurance. It was a challenging morning, to say the least.
After a few months of fits and starts, the Treasury and Fed launched the securities purchase program, but it never really took off. It was easier for the banks, if they wanted to sell, to simply do so directly through the Fed. Also, the few banks that did sell securities in the PPIP took losses, suggesting that they were not carrying the securities at true market prices.
Our loan program—which required both Fed and Treasury approval—languished. Though Ben was supportive and the Fed had approved the launch of our program, Tim twiddled his thumbs. In fact, despite repeated attempts on my part, he never approved the launch of the program as required under the statutory process. By the fall of 2009, in the face of growing congressional opposition to PPIPs (fed in part by the Sorkin piece), I gave up. We did use a PPIP structure to sell loans we acquired from failed banks, with good results. But we were never given the opportunity to launch a wide-scale program to remove bad loans from the balance sheets of the major banks. Our staff, led by Mike Krimminger and Joe Jiampietro, had worked countless hours to design the program. But it was just another example of Tim refusing to cede authority to us over cleanup programs, even though we had the best experts in the world on the subject. I think the recovery would have been much stronger if he had just let us do our job.
CHAPTER 15
The Care and Feeding of Citigroup: Bailout Number Three
Though Tim successfully stymied our efforts to launch a bad-bank program for the purchase of toxic loans, he did for a time show a lot of interest in how the structure would work. In fact, he invited me to two separate meetings in his office to discuss the subject. However, both of those meetings quickly digressed into how the bad-bank structure might work for Citigroup.
Throughout the early months of 2009, as the stress tests were being launched, there was more behind-the-scenes maneuvering on Citigroup. The bank had a terrible fourth quarter in 2008, with losses of $8.3 billion. As I had feared, the November bailout, which had provided a second capital injection and a ring fence agreement for the government to absorb potential losses on $300 billion in assets, was not enough to calm market fears.
The institution continued110 to lose deposits and was under severe pressure from key trading partners. To try to calm the markets, Citi’s management was eager to formally execute the legal documents regarding the ring fence. That would give it another vehicle for making an announcement to remind nervous depositors and others of the bank’s government support. But they still hadn’t decided on the assets they wanted to include in the ring fence, which made closing the deal difficult. Daniel Frye of the FDIC staff pushed hard to include in the final deal a ban on bonuses if the government took any losses on the ring fence, but the other regulators would not support us. On January 15, 2009, we announced the execution of the deal.
The announcement didn’t do much good. The ratings agencies111 were threatening further downgrades, citing Citi’s low levels of high-quality capital—tangible common equity—and not giving much credence to the government’s preferred stock investments. They also cited Citi’s poor earnings prospects and the fact that it hadn’t put its most toxic securities into the ring fence.
Then, on January 16, without any consultation or advance warning to us, Citi’s management announced a “restructuring” of Citi into a “good bank” and a “bad bank.” That was nothing more than a publicity stunt. It identified about $864 billion in “bad” assets that Citi said it eventually planned to sell off. There was no fundamental restructuring, not even a concrete plan on how to sell those assets without raising additional capital to absorb the losses, just a happy-face announcement that it was going to get rid of all the stupid loans and investments it had made over the past several years. The market took it for what it was.
Clearly, more needed to be done. As I had predicted in November, the second round of capital investments and the ring fence were not enough. We wanted Citi really to sell off its bad assets—not just announce that it planned to do so—and major management changes were also clearly needed. I was particularly concerned about Citi’s CEO, Vikram Pandit. He had no commercial banking experience, yet he was running one of the world’s largest commercial banks. In addition, Citi management’s performance during the crisis had not been impressive. They had had a very difficult time making decisions and then executing once the decisions were made. Their dithering and ultimate failure to quickly close the Wachovia deal had given Wells the unexpected opening to come in with a competing bid. They frequently couldn’t answer even basic questions that I had—for instance, it took them weeks to tell me how much of their foreign deposits were covered by foreign deposit insurance schemes. They had a terrible time deciding what assets to put into the ring fence. They had a reputation in the market of being the gang that couldn’t shoot straight. From what I saw, it was well deserved. They had brought in Richard Parsons to be the new chairman of the board on January 21. I was skeptical when Parsons’s selection as chairman was announced. I viewed him as a politically connected insider, not someone who knew much about running banks. But I give him credit; on the surface at least, he seemed to want to have a good working relationship and once appointed, he actively reached out to us and the other regulators to try to get on top of the management problems and correct them.
Indeed, I was surprised when Tim started reaching out to me directly on the possibility of doing a good-bank/bad-bank structure for Citi. Initially, he raised the idea of the FDIC setting up and funding a bad bank
, without imposing any loss absorption on shareholders and bondholders. I was flabbergasted. Why in the world would the FDIC take all of the losses and let Citi’s private stakeholders take all of the upside with the good bank? During the second meeting, we discussed a proposal to have the common equity and some of the preferred shareholders help absorb losses. Our view was that all of the private preferred shareholders should convert and that the bondholders should take some losses as well. That was only good policy—the private sector should take losses before the government—and it was clear that given the size of Citi’s toxic assets, both shareholders and bondholders would have to pony up to keep the bad bank solvent.
That was a nonstarter for Tim. He wanted the FDIC to take a hit. I reminded him that our statute clearly gave us priority over bondholders. I wasn’t going to budge. Given Citi’s highly unstable condition, Tim felt that he needed to go ahead with something, so he decided to convert $25 billion of the government’s preferred shares into common equity to raise Citi’s all-important tangible common-equity ratio. We felt strongly that Treasury should require all of the preferred securities holders to convert to common shares before Treasury converted, but Tim thought that was too harsh. He argued that the preferred shareholders would never go along.
We had also hoped112 that Tim would impose conditions on Citi related to disposing of its toxic assets as well as seek strong commitments on management changes. Though language along those lines was included in early drafts of the press release announcing the third bailout, it was dropped in the final release. Treasury agreed to convert113 its preferred shares to match private preferred conversions up to $25 billion. The release also stated that a majority of Citi’s board would be replaced. There was no mention of dealing with Citi’s troubled assets, nor was there a hint of management changes.