“I Thought We Were All in This Together”

Actions taken to address the 2008 financial crisis went well beyond devising resolution strategies for individual large financial institutions. Numerous broad-based programs were developed by the Treasury, the Federal Reserve, and the FDIC to provide liquidity support to the financial sector and guarantee bank debt. As the crisis intensified, these programs expanded to cover broader segments of the population.

The net effect of the federal government's flurry of activity was that Washington temporarily guaranteed virtually all of the debt in the financial system. Almost no one with money deposited at, or invested in, a large financial institution was allowed to lose that money. Then the crisis ended.

The first new broad-based guarantee program was for money market mutual funds (MMMFs), which were in a desperate situation subsequent to Lehman's collapse. MMMFs are large pools of money invested on behalf of individuals, businesses, and governments. The funds are part of the shadow banking system, meaning they are outside of the more heavily regulated commercial banking industry. MMMFs have no capital requirements or federal deposit insurance to protect them against losses—the rationale being that their investors are viewed to be more sophisticated market participants such as pension funds, insurance companies, mutual funds, and other types of “institutional investors” who do not need government regulatory oversight and protection. The investors in MMMFs are highly risk averse, which means they want safe, low-risk assets. The risk-averse nature of these investors also means they withdraw their money at the slightest hint of trouble.

The largest MMMF, the Reserve Fund, had sought higher returns than some other funds and had invested some of its money in riskier, short-term securities issued by Lehman Brothers. In the aftermath of Lehman's bankruptcy, those securities were expected to be worth only a few cents on the dollar. This meant that the Reserve Fund's investors were likely to lose some of their money, in effect “breaking the buck,” as it is called in the MMMF world.

It was no longer clear which MMMFs could be trusted, so investors also began withdrawing their money from other funds. The amount at stake was significant; a total of $3.4 trillion was invested in MMMFs. As these funds met their withdrawals, they were unable to provide loans to banks and corporations in the same manner they had in the past. Many large corporations were dependent on this source of short-term funding, so they also began to have cash shortfalls. The problems in the financial sector were beginning to infect the broader economy.

Fearing a complete financial and economic meltdown, the Treasury sought to calm the situation by guaranteeing that no investors in MMMFs would lose any money. In other words, they were bailed out. These investors weren't just from Wall Street; they were people from across the country who owned mutual funds, brokerage accounts, insurance policies, pensions, and other retirement funds.

The Treasury initially planned to announce an unlimited guarantee for MMMF investors. This would have created an incentive for uninsured depositors to leave commercial banks and savings-and-loans (S&Ls), further exacerbating their problems. After the FDIC and the banking industry pointed this out, the Treasury modified its program to provide unlimited coverage only for existing balances in MMMFs as of September 19, 2008, without providing coverage for any new money. This created the desired effect of stabilizing MMMFs, but without further destabilizing commercial banks and S&Ls.

Subsequent to the Lehman bankruptcy, Paulson put forward a more comprehensive plan for addressing the financial crisis. He wanted Congress to provide $700 billion in taxpayer money that would be used to purchase bad assets from banks. Heated debate quickly broke out over issues such as congressional oversight, the amount of money that should be set aside to help homeowners facing foreclosure, and what controls should be placed on executive pay at participating institutions.

It looked as though the bailout plan would be approved, but on September 29, the House of Representatives voted against it, sending the Dow Jones Industrial Average down by 778 points, the largest one-day point loss ever. As the financial system and the economy continued to plummet, Congress reconsidered and passed the Troubled Asset Relief Program (TARP) on October 3. One of the changes that helped facilitate its passage was a temporary increase in the deposit insurance coverage level from $100,000 to $250,000.

On October 8, the Federal Reserve announced that for the first time it would expand its lending programs to lend money directly to nonfinancial U.S. corporations. The Fed was responding to the fact that financial firms were too weak to sufficiently fulfill their traditional role of lending to major corporations. The Fed had already expanded its lending programs to commercial banks by allowing them to pledge a broader range of collateral against their loans.

These measures didn't slow the panic. Over a seven-day period, ending on October 9, the stock market dropped 20 percent (679 points alone on October 9). Bank stocks were particularly hard hit. The decline continued into the next day as the Dow Jones Average dropped by nearly 700 points before recovering enough to end the day down only 128 points. The market appeared to be in a free fall. Additional steps clearly were necessary.

At the Treasury and the Federal Reserve's request, the FDIC looked at developing a program that would guarantee newly issued bank debt. The concept was a radical departure from anything the FDIC had ever done before. The FDIC insured deposits, not other forms of bank debt. The larger banks couldn't borrow money without the government standing behind them, which meant they would not be able to lend money either.

Sheila Bair responded with a proposal that the FDIC guarantee 90 percent of newly issued bank debt. In return, participating banks would pay a fee to the FDIC. Conversely, Tim Geithner wanted to protect 100 percent of newly issued debt, and he was opposed to charging the banks a fee. The compromise was for the FDIC to guarantee 100 percent of newly issued debt, and for participating banks to pay the FDIC a fee for that support.

The compromise produced a better result. There was no reason not to charge the banks a fee for their added protection. To let the banks have added protection for free would simply be providing them with a subsidy at the expense of the federal deposit insurance program. However, the guarantee had to be set at 100 percent because no one would want to lend the banks money if they thought they might lose 10 percent. We had seen what happened the year before at Northern Rock, where British depositors ran from the troubled bank, even though 80 percent of their deposits were insured. That experience showed that co-insurance programs, where the depositor or lender covers part of the losses, don't work in a crisis (the exact time they are needed).

The FDIC's board of directors and the Federal Reserve's Board of Governors recommended to the secretary of the Treasury that a systemic risk exception be granted. Upon consultation with the president, as required by the law, a systemic-risk exception was authorized by Secretary Paulson. Following the Citigroup/Wachovia open-bank assistance transaction that was never consummated, the creation of the Temporary Liquidity Guarantee Program was only the second time a systemic-risk exception had ever been granted.

Some later questioned the legality of the debt guarantee program. The FDIC did not have the authority to provide financial assistance to open banks without there being a finding of systemic risk. Many people both within and outside of the bank regulatory community had assumed that systemic-risk authority was designed to stabilize the financial system solely in situations where the threat was from an individual bank failure. Here, there was no imminent failure precipitating action; instead, the argument for a systemic-risk exception was based on the threat to the entire financial system. Regardless of the controversy, the FDIC and the other agencies were willing to risk pushing the boundaries of their legal authority and take unprecedented steps to save the financial system.

On October 14, the FDIC announced the creation of the Temporary Liquidity Guarantee Program, a voluntary two-part program for banks, thrifts, and their parent companies. The first part of the program was a Transaction Account Guarantee Program, which allowed banks to purchase additional deposit insurance coverage for their non-interest-bearing transaction accounts (these are typically held by small businesses). This part of the program was designed to benefit the smaller community banks around the country, since it would allow them to retain small business customers that might otherwise turn to banks perceived to be too big to fail. The second part of the program was the Debt Guarantee Program, which allowed banks, thrifts, and their parent companies the option to buy insurance coverage for senior unsecured debt issued before June 30, 2009.

The FDIC's Temporary Liquidity Guarantee Program was well received by the marketplace. The day it was announced, along with the Treasury's decision that $250 billion of its TARP money should be used to invest capital in banks rather than purchase assets from them, the stock market rose by 936 points, or 11 percent. It was the fifth-largest percentage gain in history. However, that bounce lasted only a day. The next day, the stock market dropped by 733 points, or 7.9 percent. The panic wasn't over.

Behind the scenes, the FDIC staff focused its attention on the difficult task of transforming the Temporary Liquidity Guarantee Program from a concept into reality. The first issue that arose was the speed at which the agency should make payments to investors if a bank failed and thus defaulted on its debt payments. Insured depositors often wait over a weekend before getting their money back. But this was not how bond markets worked. Investors in bank debt told us that if the program were to work effectively, payments would have to be made on the same day that a bank defaulted.

This created a bit of a dilemma. Bair wanted the agency to protect itself from losses, but that wasn't possible if we were to develop a credible program. Same-day payments would be necessary to provide an acceptable guarantee, but they would expose the FDIC to some risk of losses. Delayed payments could protect the FDIC against losses, but they would be viewed as an inferior program by the marketplace. The program had to be credible, so the arguments for timely payments won out.

Working out the details for same-day payments was complicated, but the staffs of the bank regulatory agencies and the banking industry worked well together. Bill English and others at the Federal Reserve worked closely with Art Murton, Diane Ellis, and others at the FDIC to iron out the details.

Another problem arose in dealing with the debt-rating agencies. Even though the FDIC's guarantee was backed by the full faith and credit of the United States government, Standard & Poor's (S&P) wasn't going to give the new program's FDIC-guaranteed debt its top triple-A rating, despite the fact that they had given that rating to all sorts of poor-quality debt that helped create the crisis. S&P said it wasn't sure the FDIC could provide timely payment in the event a bank defaulted on its commercial paper. The lack of a triple-A guarantee on FDIC-guaranteed debt would mean much more than a bruised ego for the agency. We were concerned that the public would take that as a sign that the FDIC could not be trusted to protect their insured deposits. Eventually, S&P acquiesced and granted its triple-A rating to FDIC-guaranteed debt.

These and other key operational details for the Temporary Liquidity Guaranty Program were worked out in about a month. Once under way, the two-part program proved to be popular. Over 7,200 banks—most of them small—joined the Temporary Account Guarantee Program, which added an additional $834 billion to the amount of deposits insured by the FDIC. A similar number of institutions participated in the Debt Guarantee Program. At its peak, there was $345 billion in guaranteed debt, much of it issued by the country's largest banks.

The Temporary Liquidity Guarantee Program was a financial success. By year-end 2012, the FDIC had received $10.4 billion in fees from the Debt Guarantee Program, while incurring only $153 million in losses. The FDIC received $1.2 billion in premium income under the Transaction Account Guarantee Program, but incurred $2.1 billion in losses. On December 31, 2012, the net surplus of $9.3 billion was transferred into the FDIC's deposit insurance fund.

The Temporary Account Guarantee Program was scheduled to expire at year-end 2009, but the FDIC board of directors extended it twice and then Congress extended it for two additional years, through year-end 2012. The FDIC viewed Congress's decision to extend the program as a sign of approval for what originally was viewed to be a controversial decision. (However, in the 2010 Dodd-Frank Act, Congress revoked the FDIC's authority to initiate such programs without congressional approval.)

The FDIC guarantee programs were a critical element that helped bring an end to the financial crisis. Between the increase in the deposit insurance coverage level to $250,000 and the Temporary Liquidity Guarantee Program, the FDIC guaranteed broader segments of the U.S. population that any money they had deposited in, or lent to, banks and thrifts was safe. Insured deposits increased from about $4.7 trillion, or 63 percent of all bank deposits, to $7 trillion, or 80 percent of all deposits. For the smaller community banks, which had far fewer customers with large-dollar accounts, the FDIC insured almost all of their deposits. In addition, for the first time ever, under the Debt Guarantee Program, the FDIC explicitly protected certain types of nondeposit debt.

During the fall of 2008, the Federal Reserve was providing a tremendous amount of support to the financial system. It flooded the market with liquidity. The Fed had learned from its mistakes during the Great Depression. It failed miserably in that role during the 1930s, which in the eyes of most historians contributed significantly to the severity of the Depression. Federal Reserve Chairman Ben Bernanke understood this history lesson well and was determined to ensure that his agency did not make the same mistake.

The severity of the financial crisis was forcing the federal government to do more and more to provide liquidity to the banking system and to guarantee the public that money they deposited, invested, or otherwise lent to the banking system would remain safe.

Throughout the financial crisis, the well-publicized infighting among Bair and the other principals created problems. Over the course of 2007 through early 2009, I accompanied Bair to a number of meetings with, among others, Paulson, Geithner, Bernanke, and Dugan. Initially, Bair had a hard time getting a word in edgewise. At first, I attributed that to the FDIC's still being perceived by the Treasury and the Federal Reserve as minor leaguers who didn't need to be consulted, even though we were expected to provide financial and organizational support for their ideas. While I believe that remained part of the reason for poor communication, it soon became clear that the personal relationships between Bair and the other principals were not very good.

Treasury and Federal Reserve officials believed Bair was difficult to deal with and was overly focused on the FDIC at the expense of the overall financial system and couldn't be trusted to keep internal discussions confidential. In his book, Geithner states that the FDIC was tested by being given false information to see if it was leaked to the press—which it was.

Bair created her fair share of discord. She often pushed for additional concessions from the other principals late in the discussions. More than once, an FDIC staff member was directed to call the Treasury or the Federal Reserve with a new negotiating point, just before a new program was about to be announced, and after the other principals thought we had reached agreement.

As a result, the FDIC continued to be brought late into many discussions. That hindered our ability to execute our responsibilities. We saw the problems that arose at IndyMac where, for different reasons, there was not enough time to engage in proper advanced planning. The urgency of the moment didn't always allow for advance notice and discussion, but other times it was avoidable.

In early 2009, the Treasury, the Federal Reserve, and the FDIC discussed whether to revive the idea that bad assets should be removed from the banking system through a Public-Private Investment Program. Bair strongly supported an FDIC-managed program that would allow private-sector investors the opportunity to partner with the federal government and purchase poor-quality loans from banks. The Treasury and the Federal Reserve favored a Fed-managed program that would allow private-sector investors the opportunity to partner with the federal government to purchase poor-quality mortgage-backed securities. After some back and forth, the Treasury agreed to support both programs.

The concept called for private-sector buyers to have the federal government (Treasury) as their equity partner. The public- and private-sector partners would share in the ownership of any purchased assets. The partnership would allow taxpayers the opportunity to gain in any upside along with the private-sector investors. In practice, banks would identify the loans they wanted to sell, pools of loans would be auctioned off to the highest bidder, and the private-sector partners would manage the assets.

The major, and perhaps insurmountable, obstacle would be finding a sales price that would be agreeable to both buyers and sellers. Potential buyers would want to discount the price of the assets to something less than their current market value so they could make a reasonable profit on the transaction. Sellers viewed their loans as worth much more than current market value since they believed that if the assets were held to maturity, market prices would improve dramatically. There was a large gap between these two views. Helping to bridge that gap was the FDIC's willingness to allow buyers to finance part of the purchase price by issuing debt guaranteed by the FDIC. The FDIC guarantee would lower the partnership's borrowing costs, which would encourage potential investors to raise their bids closer to the banks' expectations. But unless the federal government was going to force banks to sell their bad assets at market value, causing them to incur significant losses, the program was going to be hamstrung.

The investment community did not want the federal government as its partner. Many felt that if they profited from the partnership, Congress might view those profits as coming at taxpayer expense and retroactively raise their taxes as a result. While of less significance, private-sector investors also were leery of the many operational requirements that would be placed on them by the government.

Bair told me that she wanted me to run the FDIC-managed program. By now, Geithner had replaced Paulson at the Treasury, and relations between the two agencies were not at their high point. At the first large staff meeting I scheduled at the FDIC, the senior Treasury official working on the program excused himself after a couple of minutes. Apparently, he had something more important to do than trying to develop a program to help save the financial system. We set about working on the program with his staff.

Early on, we decided it would enhance the likelihood for success if the Treasury did not participate as an equity partner. This didn't have anything to do with the degree of cooperation we were getting; rather, we believed it would enhance the program's likelihood of success. The private sector would have fewer concerns partnering with the FDIC, since our funding came from the banking industry rather than from taxpayers.

Eventually, work on the FDIC-managed program was discontinued because another interagency program was developed (stress tests) that eliminated its need. However, the FDIC reverted back to what had worked well during the S&L crisis and successfully used the public-private joint investment concept to sell assets from closed, rather than open, banks. The closed-bank program was easier to implement, since there no longer was an unwilling seller. It would prove to be a financial and operational success.

The concept for stress tests began in April 2008 subsequent to the collapse of Bear Stearns. Tim Geithner, then president of the Federal Reserve Bank of New York, was concerned about the condition of the four remaining large investment banks: Goldman Sachs, Morgan Stanley, Merrill Lynch, and Lehman Brothers. There was less concern at that time about their capital adequacy than about their potential lack of liquidity and whether they, too, could be susceptible to runs or continued demands for additional collateral by anxious lenders.

The derivatives market was a major concern. Bear Stearns had been bailed out in part because regulators did not know the systemic risk associated with the investment banks' interrelationships with, and exposure to, other large firms. This lack of transparency was frustrating to the bank regulators, who weren't directly responsible for supervising investment banks and other parts of the shadow banking system, since the bank regulators knew that spillover effects from these relationships potentially could harm the commercial banking system and create systemic problems.

The Federal Reserve Bank of New York wanted to know if the remaining large investment banks could survive a “Bear Stearns” scenario. Testing started taking place over the summer of 2008, but time ran out. By September, all four of the remaining large investment banks were in trouble. Lehman Brothers went into bankruptcy. Bank of America purchased Merrill Lynch. Goldman Sachs and Morgan Stanley were granted bank holding company charters, which gave them access to the Federal Reserve's emergency lending facilities.

By late fall of 2008, Citigroup needed a bailout. By now, bank regulators were pretty frustrated with Citigroup. A year earlier, the Federal Reserve and the Office of the Comptroller of the Currency (OCC) had asked the company to report on its overall exposure to subprime mortgages. The answer they received was that the exposure was modest, but Citigroup had only accounted for subprime mortgages on the bank's books. It did not include the exposure it had elsewhere. That exposure, which included so-called super senior collateralized debt obligations (CDOs), was enormous and far less safe than what was suggested by the CDOs' triple-A ratings.

The capital requirements enacted in the early 2000s in the aftermath of the Keystone, BestBank, and Superior failures, which focused on raising capital for subprime loans held by banks, but not by nonbank affiliates or subsidiaries, had come back to bite. Subprime lending had simply shifted from banks, where the capital requirements were relatively high and the risks could be easily observed, to nonbank subsidiaries or special-purpose vehicles, where the capital requirements were relatively low and the risks could less easily be detected.

As a result, much uncertainty remained as to the true condition of most of the country's largest financial institutions. The market needed to know which institutions could withstand a continuation of severe market conditions. Without such an understanding, all of the large banks were suspect and the crisis would not end.

The view that emerged among bank regulators was that the stress scenarios would have to be severe enough to leave no doubt about whether the largest banks could survive the worst of market conditions. The tests would have to be much tougher than originally planned and include more banks. The tests needed to be harsh, doable, and logically consistent.

In February 2009, the federal banking agencies began their work anew. Til Schuermann, a senior official at the Federal Reserve Bank of New York, was placed in charge of the interagency effort. Til argued that the first step toward developing credible stress tests was to ensure that the regulators took control of the modeling efforts, something that had not been done in the past. The FDIC supported the Federal Reserve in taking that position. The OCC argued against it, claiming that it was an unnecessary complication and expense. Til's position won. The mistakes of Basel II, where the banks' own models were exclusively relied upon, were not going to be repeated here.

The interagency task force began developing the loss rates that would be applied to various types of bank assets under a severely distressed economic scenario. Lisa Ryu and Robert Burns were key FDIC officials involved in that effort. They and about 150 other examiners and analysts from the Federal Reserve, OCC, and the FDIC worked countless hours over a 10-week time frame to develop the economic assumptions that would be used to indicate severe stress and then to assess the banks' responses and determine the results.

The stress test they developed imposed tougher loss rates on participating banks than what was experienced each and every year since 1922, a period that included the Great Depression. As a result, there would be no doubt about the severity of the tests. Nineteen U.S. banking organizations with more than $100 billion in total assets, together holding two thirds of the assets in the entire banking industry, were required to determine whether they had sufficient capital to withstand these depression-like conditions.

Overall, the agencies worked well together, although there was some tension. At one interagency meeting, a dispute arose over whether there should be separate press releases announcing the process and then later the results, or just one public announcement. Bair threatened to have the FDIC drop out as a participant. Bernanke looked up at her and calmly said: “I thought we were all in this together.” Ultimately, we were.

Surprisingly, some staff members did not want to reveal individual bank results, since bank supervisory ratings have always been confidential. However, the whole point of the exercise was to provide enough information to gain public credibility. It quickly was decided that individual results would be made public.

The results were released on May 7, 2009. They indicated that 10 of the 19 firms needed a total of $75 billion in additional capital to withstand the most distressed scenario. These firms were required to develop plans to raise that capital. To ensure the public that none of the 19 firms would be closed under any scenario, the Treasury made a commitment to provide the capital, if necessary. In effect, the federal government was assuring the public that none of these institutions would be allowed to fail.

The severity of the stress tests, combined with the federal government's guarantee that the country's 19 largest financial institutions with two thirds of the assets in the entire banking system would not be allowed to fail under any circumstances, was the final step that ended the financial crisis. Nine of the 10 banks required to raise additional capital had no problem doing so. Only GMAC required capital from the federal government. It would be a long slow road to economic recovery, but the panic had subsided because the federal government temporarily stood behind virtually the entire banking system.

Looking back on 2008 and early 2009, the U.S. financial system came extraordinarily close to collapse. The consequences of the financial crisis, while painful, could have been much worse had regulators not acted to stabilize the financial system. After the fact, the government's actions were widely criticized. No one likes bailouts. They can be unfair and harm long-term economic growth by planting the seeds for future crises and even larger bailouts. I believe that if the government thought it could let a major financial institution fail without destabilizing the entire financial system, it would have done so. Indeed, that was the FDIC's preference and the rationale behind closing Washington Mutual.

The Treasury, the Federal Reserve, and the FDIC reached a point where everyone either sank or swam together. We could have stood on principle and watched the financial system collapse, or we could use the power of the federal government to calm the situation by providing liquidity and temporarily guaranteeing everyone that their money was safe. We chose the latter course. At a time when there was a crisis of confidence, and the entire financial system was teetering on the verge of collapse, bailouts were necessary.