Federal Reserve documents 28 page

THE SWEARING-IN OF the president indeed marked a new phase in the battle against the crisis. His team brought different ideas, but the bigger transition was that the crisis itself was becoming less purely financial and more economic in nature. Fittingly, President Obama aimed his first major policy initiative at arresting the devastating economic contraction—now more than a year old. As new (and old) Keynesians would predict, collapsing private demand—consumer spending, home purchases, capital investment—had sent production and employment reeling. As Keynes had first suggested in the 1930s, in an economic slump public spending could replace private spending for a time. With the economy still in free fall and with short-term interest rates already near zero, the economy certainly needed fiscal help— increased government spending, tax cuts to promote private spending, or both. I had said so (albeit in my usual cautious central bank speak) during the fall, to the point that the Wall Street Journal editorialized that I had effectively endorsed Obama for president. I wasn’t endorsing a candidate, I was endorsing a program, just as I had supported President Bush’s fiscal stimulus (in the form of tax cuts) that had passed in early 2008.

On February 17, less than a month after his inauguration, Obama signed a major fiscal package, the

American Recovery and Reinvestment Act of 2009. The $787 billion bill included $288 billion in tax reductions to help spur consumption and investment—most notably, a temporary reduction in Social Security payroll taxes—as well as $144 billion in aid to state and local governments, mostly to support spending on education and Medicaid (the principal government program to provide health services to the poor). The remaining $355 billion was spread among diverse federal spending programs, including $40 billion for extended unemployment benefits and $105 billion for infrastructure investment.

I am sure that the Recovery Act helped create jobs and slow the economic contraction—a conclusion shared by our own staff and the nonpartisan Congressional Budget Office. Nevertheless, the recovery would be slow and protracted. In retrospect, some economists (including Obama’s CEA chair, Christy Romer) have said that the stimulus package was too small. Within the Fed, some of our fiscal specialists expressed that concern at the time. Over the next few years, I came to agree that, from a purely economic perspective, the program probably was too small.

I know it’s hard to think of a $787 billion package as small, but its size must be compared with its objective of helping to arrest the deepest recession in seventy years in a $15 trillion U.S. economy. Also, several considerations reduced the impact of the program. First, the headline figure overstated the program’s effective size to some degree. For example, some of the Medicaid spending and certain tax fixes included in the bill likely would have occurred anyway. Second, and importantly, much of the effect was offset by spending cuts and tax increases undertaken by state and local governments. With economic activity contracting, the income, sales, and property tax revenues of state and local governments fell sharply. Many of these governments operate under laws requiring balanced budgets, so they responded to the lost revenue by laying off workers (including thousands of teachers, police, and firefighters), raising tax rates, and canceling capital projects. The federal stimulus package aided state and local governments but not nearly enough to make up for the budget crunch they faced.

Defenders of the package maintain that it was the largest politically feasible option. They may well be right. (Three Republicans supported it in the Senate and none in the House.) Voters usually cheer tax cuts and increased spending on social programs and infrastructure, at least when they benefit directly. But the huge budget deficit generated by the package—and by recession-induced declines in tax revenues and automatic increases in social spending (on unemployment benefits and food stamps, for instance)— worried many Americans. It didn’t help that some voters likely perceived the stimulus as a “Christmas tree” that funded legislators’ pet projects, regardless of their merit. Perhaps if the package had been more clearly focused and sold as a way of strengthening America’s infrastructure and improving the economy’s long-run productive potential, it would have been more broadly supported. But it’s hard to know.

THE OBAMA ADMINISTRATION in its early weeks also focused on rolling out plans for reducing mortgage foreclosures. After extensive conversations and debates during the transition between administrations, Tim and his team had settled on a strategy. On February 18, President Obama unveiled the Making Homes

Affordable program. It had two main components. The first, called the Home Affordable Refinance Program (HARP), would help underwater homeowners who were current on their payments to refinance into mortgages with better terms and lower monthly payments. It echoed some aspects of a proposal by Columbia University professors Christopher Mayer and Glenn Hubbard. (It was Hubbard who, as an adviser to President Bush, had invited me to interview for a seat on the Board of Governors.) With lower payments, an underwater homeowner would have a better chance of staying current. However, only mortgages owned or guaranteed by Fannie and Freddie, which were indirectly controlled by the government, were eligible.

The second component, the Home Affordable Modification Program (HAMP), would target homeowners who had already missed payments. Unlike HARP, HAMP was financed by TARP funds and included homeowners whose mortgages were not owned by Fannie and Freddie. Borrowers with a monthly debt burden exceeding 31 percent of their gross monthly income would be eligible for a trial loan modification, which could be converted into a permanent modification if they were able to keep up with their payments during the trial period. HAMP would pay private mortgage servicers a fixed sum for each permanent modification they completed. To encourage sustainable modifications, servicers also would receive ongoing payments when borrowers proved able to remain current on their modified mortgages.

I remained perplexed that helping homeowners was not more politically popular. But Americans apparently were no more disposed to bail out their neighbors than they were to bail out Wall Street. Indeed, television personality Rick Santelli’s famous rant about homeowner bailouts on the cable station CNBC is thought by some to have been a trigger of the Tea Party movement. In February 2009, Santelli, commenting from the floor of the Chicago Mercantile Exchange, turned to traders and shouted (ungrammatically), “How many of you people want to pay for your neighbor’s mortgage that has an extra bathroom and can’t pay their bills, raise their hand.” A chorus of “No!” rose from the trading floor. “President Obama, are you listening?” he asked.

IN THE FIRST months of the new administration, completing the stabilization of the banking system remained a top priority. Short-term funding markets had improved noticeably, in large part because of the Fed’s lending programs. But, despite the new TARP capital and the arrangements to prop up Citi and Bank of America, market confidence in banks remained shaky. The share prices of the largest banks nose-dived in January and February—by about 80 percent, in the case of Citi and Bank of America, by half or more for other large banks. The cost of insuring large banks’ debt against default also remained worryingly high, suggesting that market participants saw the failure of another major institution as quite possible. The weakening economy and continuing worries about bank losses were probably the main factors behind market concerns, but uncertainty about what the new administration’s plans would mean for bank investors and creditors didn’t help.

During the transition and in the days after he took office, Tim called frequent meetings with the

Treasury, the Board, the FDIC, and the Office of the Comptroller of the Currency to discuss our options. The New York Fed, which supervised many of the largest bank holding companies, usually joined by phone. Sheila Bair of the FDIC and John Dugan of the OCC were being retained by the new administration. Bill Dudley left his position as manager of the Open Market Trading Desk to replace Tim as the president of the New York Fed.

At first, we and incoming Treasury officials focused on strategies for shoring up the banking system, including further injections of capital, asset purchases, and new forms of guarantees. In late December, Don Kohn had reported on work at the New York Fed in all these areas, and on the day after the inauguration, Board staff walked me through a long list of potential strategies.

Sheila and the FDIC also developed proposals, including the creation of an “aggregator bank”— essentially, a government-owned bank that would buy or guarantee troubled bank assets. Sheila also proposed that banks be allowed to issue so-called covered bonds, with an FDIC guarantee. Covered bonds, common in Europe but rarely seen in the United States, are bank-issued bonds backed by specific, high-quality assets, usually mortgages. If a loan backing a covered bond goes bad, the bank must replace it with another, sound loan. Consequently, covered bonds are safer for investors than standard asset-backed securities, in which defaults of the underlying assets can result in losses to investors.

At the Fed, we supported putting more capital into the banks and were also receptive to covered bonds. We had a concern about Sheila’s plan for an aggregator bank: She wanted the Fed to finance it with 13(3) lending. We certainly didn’t rule out the possibility, but we knew that lending to an aggregator bank would further expand our balance sheet (then at about $2.1 trillion, more than double its pre-crisis level), adding still more reserves to the banking system and possibly further complicating our conduct of monetary policy. To address this issue, we asked Treasury officials whether they would support legislation to allow us to sell our own short-term debt (which we provisionally called Fed bills) directly to the public. Issuing our own short-term debt, as many central banks do, would allow us to finance our lending without creating bank reserves and so give us better control over the federal funds rate. Treasury officials were skeptical. They doubted Congress would approve (I had to agree), and they were reluctant to have a new type of government obligation competing with Treasury securities in the marketplace, which could complicate Treasury’s financing of the national debt. Their concern about competition also made them skeptical of Sheila’s covered bonds, which, under her proposal, would be sold with the “full faith and credit” of the U.S. government and thus would be effectively equivalent to U.S. government debt.

Always overhanging these discussions was a concern that bank losses might yet outstrip the capacity of the TARP, even though Congress had released the second $350 billion installment. Getting the most bang for the buck out of TARP funds was therefore critical. As with the original debate about asset purchases versus capital injections, that consideration seemed to weigh in favor of putting more capital into banks, perhaps on an as-needed basis rather than in the broad-based manner of the Capital Purchase Program.

Once confirmed as Treasury secretary, Tim became the final decision maker when it came to disbursing TARP funds. He encouraged robust interagency discussion and creative thinking, but ultimately he discarded many of the options we had evaluated in favor of a relatively simple package. He proposed putting the largest, most systemically important banks through what he originally called a “valuation exercise”—later called the “stress test.” It would estimate how much capital those banks would need to withstand a deep recession, one even deeper than what we were experiencing, together with significant further deterioration in financial markets. If market analysts found the estimates credible, the stress test would increase confidence in the banks that passed. Banks that came up short would get a chance to raise capital privately; if they couldn’t, Treasury would fill the hole with TARP funds. Either way, we hoped, customers, counterparties, and potential investors would know the banks were viable.

In addition to the stress tests, Tim proposed vastly expanding the capacity of the Fed’s TALF program —created to unfreeze the market for asset-backed securities—to up to $1 trillion (from $200 billion), and broadening it to accommodate more types of assets, such as loans to finance business equipment. The new ceiling, which seemed unlikely to be reached, was intended to convey the strength of the government’s commitment to unfreezing the important market for asset-backed securities. We agreed to the expansion of TALF so long as Treasury provided additional capital from the TARP to protect the Fed in the event of losses.

Tim also wanted the TALF to include older, “legacy” assets, such as existing private-label mortgage securities, as a way of increasing prices and improving liquidity for those assets. We resisted that step, worried about the effects on the size of our balance sheet and because we wanted the program to focus primarily on new credit extension. Eventually, we did make a limited exception for highly rated legacy securities backed by commercial real estate mortgages. To ensure that the valuations of those mortgages were up to date, we required them to be re-rated and repackaged into new securities. We made the exception because we believed that supporting the market for existing commercial MBS would help revive that market, which at the time was moribund, and thus help restart the flow of credit for new commercial projects.

Additionally, Tim proposed a new Treasury program, without Fed involvement, targeting other legacy assets. Under the PPIP—for Public-Private Investment Program—private investors would receive loans from TARP to finance their purchase of existing assets, such as private-label residential MBS and structured credit products created before the crisis. The investors would also put their own money at risk, and they and the government would share any profits made from later reselling the assets. The program gave private investors, rather than the government, the responsibility for deciding which assets to buy and how much to pay, thereby avoiding having the government determine prices for toxic assets. The investors had strong incentives to make smart choices, since their own returns depended on the assets they chose and their ability to keep down the costs of acquiring them. It had been a long time coming, but the Treasury had finally found a way to fulfill Paulson’s original vision of using TARP funds to buy troubled

assets, without overpaying for them.

Tim publicly introduced the stress-test idea, along with the expansion of the TALF and the creation of the Public-Private Investment Program, in a speech delivered February 10 in front of a phalanx of American flags in his department’s marble-walled Cash Room. The strategy was more work-in-progress than a full-fledged plan, and financial markets reacted badly to the lack of detail. The Dow plunged 382 points that day. In a call with the FOMC to brief members in advance, I had predicted that the markets wouldn’t like the lack of specificity. But, with confidence in banks diminishing by the day, I understood the urgency of announcing something. It didn’t take long for the details to emerge: Two weeks after Tim’s speech, the Fed and other bank supervisory agencies released plans for conducting the stress tests. On March 3, we provided more information about the expansion of the TALF, and, in late March, Treasury outlined the specifics of PPIP.

Conducting the stress tests, however, would take some time, and while we waited for the results, doubts about the banking system persisted. An ongoing question, actively debated by Tim and Larry Summers in the White House, was what to do if the test revealed a capital hole deeper than could be filled by the remaining TARP funds. Summers was pessimistic and presumed that, if the stress tests were to be credible, they would have to show catastrophic losses that would overwhelm the TARP. He accordingly favored nationalizing some troubled banks—that is, having the government take them over, lock, stock, and barrel. The idea was less outlandish than it would have seemed six or eight months earlier. A week after Tim announced the stress tests, no less a free-market champion than Alan Greenspan raised the possibility of temporarily nationalizing some banks—an event that might need to occur once in a hundred years, he said. Other prominent commentators, like Paul Krugman of the New York Times, agreed that it might be necessary. But Tim wanted to avoid nationalization, if at all possible, and I sided with him. “We don’t plan anything like that,” I told the House Financial Services Committee on February

25. I repeated my view in meetings at the Treasury and the White House.

I knew that nationalization might have some political appeal. It would look less like a bailout, and we

could implement reforms at nationalized institutions without having to deal with private boards and shareholders. But, based on our recent experiences with the seminationalizations of Fannie, Freddie, and AIG, I believed in practice it would be a nightmare. Once nationalized, banks could be wards of the state for many years. Governments do not have the expertise to run banks effectively, and private investors are hardly likely to want to put money in banks that are government-controlled. And politics would almost certainly intrude—for example, nationalized banks might be pressured to extend credit to government-favored groups, irrespective of creditworthiness, which could lead to more losses and bailouts.

Whether there were feasible alternatives to nationalization, though, depended on the results of the stress tests. As the regulator of the large bank holding companies, the Fed took the lead. Coryann Stefansson, an associate director of the Board’s supervision division, organized a comprehensive review. From February through May of 2009, our staff and the staff of other agencies bore an extraordinary

burden. More than 150 Fed examiners, analysts, and economists worked evenings and weekends for ten weeks. We focused on the nineteen largest U.S.-owned bank holding companies, those with assets of $100 billion or more. Collectively, they held about two-thirds of the assets and half the loans in the U.S. banking system.

Stress testing itself was not a new idea. For years, both banks and their examiners had used the technique to analyze how, say, a given portfolio of assets might perform under adverse conditions. But, far more ambitiously, we aimed to conduct a single, rigorous test that covered all the big banks and all their assets simultaneously and using the same criteria. That way, we and the markets could both assess the overall health of each institution and compare each institution to its peers. (David Wilcox and his team had proposed a stress test of the banking system as part of the co-investment plan they presented to the Paulson Treasury in October 2008.) We asked each bank to provide detailed estimates of its likely losses and earnings over the next two years under two hypothetical economic scenarios: a baseline scenario corresponding to the consensus of private-sector forecasters, and an adverse scenario that assumed significantly worse economic and financial conditions. Our supervisors and economists intensively reviewed the reported results for consistency and plausibility, using statistical and economic models to analyze the data. When we were comfortable with the estimated revenues and losses, we then calculated how much capital a given bank would need under each of the two scenarios.

Then, as previously arranged with the Treasury, we told the banks that they had six months to raise enough capital to allow them to remain viable and continue to lend normally, even in the adverse scenario. If they were unable to raise the required capital from private markets within sixth months, they would have to take capital from the TARP under conditions imposed by the Treasury.

We also decided to make public, in considerable detail, the stress test results for each bank, including each bank’s projected losses for each type of asset. The banks objected strongly, and some of our veteran supervisors were uneasy. The release of this information would contravene the practices of generations of bank examiners at the Fed and every other bank regulatory agency, where “supervisory confidentiality” was sacrosanct. In normal times, assurances of confidentiality increase banks’ willingness to cooperate with examiners by allaying any concern that their proprietary information would be obtained by competitors. In the atmosphere of fear and uncertainty that prevailed in early 2009, we could not dismiss the possibility that disclosing banks’ weaknesses could further erode confidence, possibly leading to new runs and further sharp declines in bank stock prices. Fed Board members agreed, however, that releasing as much information as possible was the best way to reduce the paralyzing uncertainty about banks’ financial health.

Our test was tough as well as transparent, and markets judged the results to be highly credible—in part because we reported loss estimates more severe than those of many outside analysts. For example, we projected, in the hypothetical adverse scenario, that banks would suffer loan losses of 9 percent over the next two years, higher than the actual losses in any two-year period since 1920, including the years of

the Great Depression. But equally important, I think, was that government capital from the TARP could be tapped to help any bank in serious trouble. The availability of backstop capital gave the regulators the right incentives: Without it, we might have been suspected of going easy on weaker banks, for fear of inducing runs. With the backstop, investors could see that we had every reason to be tough, to ensure that troubled banks would be forced to take all the capital they needed to remain stable.

Because most banks tested were found to be either adequately capitalized or reasonably near to being adequately capitalized, the test substantially increased confidence in the banking system. After the release of the results in May, the private sector became willing once again to invest in U.S. banks. By November, the tested banks would increase their collective capital by $77 billion. Ten of the nineteen institutions needed more capital, but only GMAC, General Motors’ financial arm, couldn’t raise it on its own. Treasury injected $3.8 billion in TARP capital into GMAC (later renamed Ally Financial); this was in addition to $12.5 billion in previous injections. The cost of insuring against defaults by large financial institutions fell sharply as confidence returned.

The stress test was a decisive turning point. From then on, the U.S. banking system would strengthen steadily—and, eventually, the economy would follow.

AS I WRITE, six years after the end of the most intense period of the financial crisis, politicians, journalists, and scholars continue to debate its causes and consequences. Why did the crisis happen? What made it so bad? Were the policy responses the right ones? What would have happened if governments around the globe hadn’t ultimately contained the crisis?

New insights will certainly emerge in coming years, just as Milton Friedman and Anna Schwartz, writing in the 1960s, fundamentally changed our understanding of the Great Depression. However, as we battled an extraordinarily complicated crisis, we didn’t have the luxury of waiting for the academic debates to play out. We needed a coherent framework to guide our responses.

For me, as a student of monetary and financial history, the crisis of 2007–2009 was best understood as a descendant of the classic financial panics of the nineteenth and early twentieth centuries. Of course, the recent crisis emerged in a global financial system that had become much more complex and integrated, and our regulatory system, for the most part, had not kept up with the changes. That made the analogies to history harder to discern and effective responses more difficult to devise. But understanding what was happening in the context of history proved invaluable.

Based on the historical parallels, I believed then and believe now that the severity of the panic itself —as much as or more so than its immediate triggers (most prominently, subprime mortgage lending abuses and the house price bubble)—was responsible for the enormous financial and economic costs of the crisis. Despite the feeling at times that we were working with chewing gum and baling wire, our policies (and the Treasury’s and the FDIC’s) drew heavily on classic prescriptions for fighting financial panics, and they ultimately eased the crisis. If they hadn’t, historical experience suggests that the nation

would have experienced an economic collapse far worse than the very severe slump we endured.

THE DETAILS OF earlier banking panics in the United States differed substantially, but major panics tended to follow a consistent story line. Many were preceded by a credit boom that made both lenders and borrowers more vulnerable to financial shocks. And most began with one or more triggering events that led depositors to worry about their banks, such as the failure of the stock market speculation scheme that touched off the Panic of 1907.

In a panic, runs on a few institutions soon become contagious. Contagion can occur through several channels. When bad news about one institution emerges, for example, depositors naturally wonder whether other institutions with similar asset holdings or business models might be in trouble as well. Also, financial institutions are interconnected, regularly lending to each other and transacting through a variety of business relationships. Consequently, like a row of dominoes toppling, one institution’s failure may cause others.

Perhaps the most dangerous channel of contagion, however, is a fire sale of assets. Financial institutions facing runs must quickly obtain cash to satisfy their depositors or other creditors. If they cannot borrow the necessary cash, they must sell assets. First they jettison the easy-to-sell assets, such as government bonds. Then they try to liquidate the difficult-to-sell assets—like loans to individual businesses. If many institutions are trying to unload difficult-to-sell assets at the same time, then the market prices of those assets will plunge. As asset values fall, institutions’ financial conditions deteriorate further, increasing their creditors’ fears and possibly leading to even more widespread runs.

A firm that does not have the cash to meet its current obligations is said to be illiquid. An illiquid firm need not be insolvent; that is, the value of its assets may still exceed the value of its liabilities, even if it lacks ready cash. However, in a panic, the distinction between illiquidity and insolvency quickly blurs. On the one hand, depositors and other short-term lenders likely would not run if they did not suspect their bank might be insolvent and thus likely to default. On the other hand, in a panic, even initially sound firms may be forced into insolvency, as fire sales and any economic slump resulting from the panic depress the value of their assets. Major panics involve both illiquidity and insolvency, and so both short-term lending and injections of capital may be required to end them.

When a serious panic occurs, significant damage to the broader economy is almost inevitable. Amid fear and uncertainty, investors want to hold only the safest and most liquid assets. Lenders become ultraconservative, so credit disappears or remains available only to the best borrowers at high cost and under stringent conditions. The prices of riskier assets, like stocks and corporate bonds, may also fall sharply, reducing household wealth and companies’ access to new capital. As credit tightens and asset prices fall, firms and households hit the pause button. Hiring, investing, and spending fall precipitously, pushing the economy into recession.

This basic scenario was repeated many times in the United States until the reforms of the Great

Depression—especially the institution of deposit insurance. The U.S. financial system then entered a lengthy period of relative calm, but significant financial crises did occur in Japan, the Nordic countries, and emerging markets in Latin America and East Asia. Economists intensively studied the Asian and Latin American crises of the 1980s and 1990s but did not think that those countries’ experiences were particularly relevant to the United States. The emerging-market countries had underdeveloped financial systems and, as small economies dependent on international trade and investment, were much more vulnerable to so-called external shocks, such as sharp changes in international capital flows. Economists, including me, also studied the experiences of the Nordic countries and Japan, but we concluded that institutional, economic, and political differences made those countries special cases. We should have listened to Mark Twain, who is reputed to have said that history does not repeat itself, but it rhymes. Although the recent crisis took place in a radically different financial and economic context, it rhymed with past panics.

Date: 2016-04-22; view: 790