Stress Test

  The lack of financial prosecutions combined with the revival of financial profits have contributed to the belief that nothing has changed, that if anything the system is even more dangerous. We still do have big banks, some even bigger than they were before the crisis, and the banking system overall is somewhat more concentrated. This in part reflects a long-term industry trend, and in part reflects all the emergency mergers during the crisis. But the U.S. banking system is much smaller relative to the size of our economy, and our economy is much less dependent on banks, than in the other major economies.

  And our reforms have made the entire financial system much safer. Capital and liquidity requirements are much tougher. There is less dependence on runnable short-term funding. Much of the derivatives market has been brought out of the shadows, with margin requirements to reduce the danger of panicked sell-offs. By some measures, the so-called “shadow banking system” has been cut in half since 2007, and much of that Wild West is now subject to stronger oversight. The government also has the authority to wind down systemic firms, so markets are now much less confident the government will save them from future mistakes. The new consumer bureau is already a robust force protecting Americans from financial predation.

  Reform is a “forever war,” to borrow the title of the Dexter Filkins book on Afghanistan. It is an endless process, and we left some unfinished business. We still need stricter rules for money market funds, a new approach to the government’s role in housing finance, a reorganization of our fragmented regulatory system, and stronger emergency authorities, such as a restoration of the FDIC’s ability to provide broader guarantees. There was a lot of messiness in the initial wave of regulations that will have to be refined over time. But the notion that the financial status quo prevailed is absurd. Banking has become more conservative and less profitable. Before the crisis, the largest six banks earned 21 cents for every dollar of common equity; these days it’s less than half that.

  This is not to suggest that public perceptions of our generosity to the financial industry are unfounded. We provided extraordinary support to the financial system in general and some very poorly managed financial firms in particular. We didn’t do it to help their executives buy fancier mansions and sleeker jets. We did it because there was no other way to prevent a financial calamity from crushing the broader economy. When the financial system stops working, credit freezes, savings evaporate, and demand for goods and services disappears, which leads to layoffs and poverty and pain. When investors and creditors start to panic, consumers and businesses follow suit. And to solve a major financial crisis, you have to do things you would never do in normal times or even in a modest crisis.

  This is the central paradox of financial crises: What feels just and fair is often the opposite of what’s required for a just and fair outcome. It’s why policymakers generally tend to make crises worse, and why the politics of crisis management are always untenable.

  The intuitive response after the busting of a credit boom fueled by reckless risk-taking and excessive leverage is to try to rein in credit and risk-taking and leverage. When you’re in a hole, the first thing you’re expected to do is stop digging. The instinctive reaction of the policy wonk as well as the politician to an epic financial crisis is to punish the perpetrators and impose losses on their creditors while limiting taxpayer exposure to catastrophic risk. It seems obvious that government should discourage the risky behavior that created the mess, and certainly shouldn’t reward it. On the fiscal side, it seems just as obvious to many that exploding budget deficits should prompt fiscal restraint, that when families and businesses have to tighten their belts, the government should tighten its belt as well.

  Financial Activity Is Moving Out of the Shadows

  Bank vs. Market Funding

  The decades-long expansion of financial activity in the “shadow banking system” began to reverse itself after the financial crisis and the Dodd-Frank financial reforms of 2010. A smaller percentage of overall financial activity is happening outside traditional banks, and that activity is now subject to additional regulation, as indicated by the lighter shading.

  Sources: Federal Reserve Board and Financial Stability Oversight Council.

  All that sounds right. It usually is right. But in a severe crisis, it’s wrong. And if those natural instincts infect strategy, they can cause a lot of damage.

  In a crisis, when confidence vanishes and risk-taking stops, government measures that further discourage risk-taking further depress confidence and spur runs. And when private demand withers, government austerity only intensifies the problem. What’s needed to avoid a vicious cycle in which the financial system and the broader economy drag each other down are counterintuitive remedies: more private credit, more government borrowing, more confidence even if the mess was created by excessive confidence. Unfortunately, the only way for crisis responders to stop a financial panic is to remove the incentives for panic, which means preventing messy collapses of systemic firms, assuring creditors of financial institutions that their loans will be repaid, and generally reducing uncertainty in financial markets. In these severe cases, government needs to lean against the forces of gloom, borrowing more, spending more, and exposing taxpayers to more short-term risk—even if it looks profligate and immoral, even if it seems to reward incompetence and venality, even if it fuels perceptions of an out-of-control, money-spewing, bailout-crazed Big Government.

  That’s why we stood behind so many mismanaged firms. That’s why we provided support for institutions and financial markets with more than $30 trillion in financial liabilities. And that’s the main reason that people think we cared more about Wall Street than Main Street. The narrative of good-versus-evil was irresistible: We were saving irresponsible bankers while they continued to pay themselves huge bonuses. The conventional wisdom hardened quickly, and with a few honorable exceptions, the media rarely tried to explain that the situation was not so black-and-white.

  We didn’t do a good job of communicating the gray, either. That’s partly because I wasn’t a natural communicator. I remember that when I read Chief Justice John Roberts’s decision upholding Obamacare, I felt envious of the clarity of his argument, whatever the quality of the legal reasoning. Nobody ever accused us of that kind of clarity. We were kind of busy, of course, and we faced a different kind of challenge. We didn’t emphasize how awful things were in early 2009, because we didn’t want to depress confidence further, and we didn’t emphasize how much better things were getting later, because we didn’t want to look like we were celebrating when so many Americans were still hurting. But I wasn’t very eloquent—and rescuing banks from failure, protecting creditors from haircuts, and spending borrowed money after a borrowing boom were always going to be hard to explain.

  Herein lies the paradox. In a brutal financial crisis like ours, actions that seem reasonable—letting banks fail, forcing their creditors to absorb losses, balancing government budgets, avoiding moral hazard—only make the crisis worse. And the actions necessary to ease the crisis seem inexplicable and unfair.

  One benefit of our latest brush with catastrophe is that it illuminated as clearly as ever which kind of strategies are helpful in a crisis and which are not. We got to see the effect of our decisions in real time, and I hope the policymakers who have to confront the next crisis can learn from our successes as well as our mistakes. Because there will be a next crisis, despite all we did to improve the resilience of the system. Perhaps my experience can help future policymakers prepare for it, react to it, and try to defuse it before it does too much damage.

  Y. V. REDDY, India’s central banker, gave me a book during the crisis called Complications: Notes from the Life of a Young Surgeon, by Atul Gawande. He told me it was the best book I would ever read about central banking, and the parallels with financial crisis management really are striking. It’s about making life-or-death decisions in a fog of uncertainty, dealing with the constant risk of catastrophic failure. It’s not a co
incidence that after the crisis wound down and I started watching some TV again, I got into House M.D., the series about a misanthropic doctor who leads a team focusing on mysterious medical cases. I could relate—not to the misanthropy, but to the complex problem-solving, the inevitable complications, the team sitting around a table debating diagnosis and treatment.

  That’s financial crisis management, more or less. It’s diagnosis and treatment, prevention and cure, triage and surgery. The stakes are high and the outcomes are uncertain. Every case is different, but some protocols can be applied broadly. Here, then, is a general framework for thinking about the craft of financial crisis management, along with some thoughts about how we applied it. I’ll start with what policymakers need to do in advance to try to prevent and prepare for crises.

  REINHART AND Rogoff titled their history of financial crises This Time Is Different, because experts always have clever reasons why the boom they’re enjoying will avoid the disastrous patterns of the past—until it doesn’t. We actually know what to look out for, especially a sustained rise in private borrowing relative to GDP, the phenomenon captured by the “Mount Fuji” chart that Lee Sachs showed me at the New York Fed. The other big danger sign is a sustained rise in uninsured short-term liabilities issued by the financial sector, the kind of money that can quickly run when good times stop rolling. This is scary when it builds up in highly leveraged banks; it can be even scarier when the risk and leverage migrate outside the traditional banking system to institutions and markets with less supervision and less access to the bank safety net.

  Regulators can discern these warning signs, but they will never be omniscient or omnipotent. There’s no way to be sure exactly when a bubble will pop, no certain way to prevent a mania from becoming a panic. Manias are inherently unpredictable, and regulators are not immune to them. Long periods of success in avoiding financial collapses can actually increase vulnerability to catastrophe, because stability can breed instability. At the New York Fed, I got to see how much power the belief in the “Great Moderation” had over smart people, and to witness its expression in the credit boom. Even though we leaned against the prevailing winds, warning about the growth of leverage and the risks of the shadow banking system, taking some useful steps to encourage stress-testing and limit the danger of derivatives, we were not forceful or creative enough, and we didn’t have the tools necessary to avert disaster.

  Financial crises can’t be reliably anticipated or preempted, because human interactions are inherently unpredictable. But there is a lot that can be done in advance to make crises less damaging. These can be divided into safeguards that help reduce the likelihood and severity of crises, and emergency authorities that help policymakers limit the damage when crises erupt. The goal should not be to prevent the failure of firms that take on too much risk, but to make the system safe for their failure, to prevent their failures from unleashing panics that lead to crashes, to avoid the extreme crises that can lead to depressions when they spiral out of control.

  The most important safeguards are the constraints on risk-taking that I’ve described as “shock absorbers,” starting with strict capital requirements that restrict leverage and ensure that institutions can absorb losses in a downturn. Other shock absorbers include liquidity requirements that limit financial institutions’ reliance on runnable short-term financing, deposit insurance and discount window access for depository institutions, margin requirements for derivatives and other financial instruments, and mortgage down-payment requirements that restrict leverage for ordinary borrowers. Making the risks in banks more transparent, as we did in the stress test, can also help limit the tendency of investors in a panic to run from everyone. To be effective, these safeguards must be applied broadly across the financial system, not just to the legal entities known as “banks” but to any large firm that behaves like a bank, borrowing money short-term in order to lend or invest it in longer-term assets. Otherwise, risk will just migrate to the unconstrained firms, fueling the growth of a parallel system with more leverage and more vulnerability to runs.

  There are some economic costs to shock absorbers that require firms to operate with more conservative funding and thicker cushions of loss-absorbing capital, and the financial industry will always complain about them. But those costs are small compared to the economic benefits of less frequent and severe crises. Strong shock absorbers not only reduce the probability that any bank will fail, they reduce the risk that the failure of a bank—or a stock market crash, a real estate slump, or a recession—will create a panic that could threaten the rest of the system. They also reduce the danger that if a crisis does erupt, banks will have to rein in their lending to conserve capital, depriving the economy of financial oxygen and offsetting the power of crisis-fighting tools like fiscal and monetary stimulus.

  Shock absorbers for borrowers are helpful, too. Higher down-payment requirements for homeowners provide a cushion in case of job losses, medical emergencies, or housing downturns, protecting the market as well as the individual borrowers. Initial margin requirements for investors not only limit their leverage and potential losses if prices fall, they limit the risk of forced fire sales to meet margin calls, and the vicious dynamics of margin spirals. Those kinds of tools can take some of the momentum out of a mania, reining in excessive risk-taking.

  When firms and investors and ordinary families have strong appetites for risk, money tends to find its way around all those safeguards, the way water in a river flows over and around stones, the way risk migrated into the shadow banking system before our crisis. Regulators should try to expand the scope of those safeguards over time, to capture the changing sources of financial and economic risk, even though they will always lag behind the markets. Financial regulation will never keep up with financial innovation, but regulators should regard it as a constant challenge, and keep at it.

  DURING THE boom before our crisis, U.S. capital and liquidity requirements were too weak, but they were strong enough to fuel a huge buildup of risk outside our traditional banking system—in Fannie and Freddie, in investment banks like Bear and Lehman, in other nonbanks like AIG and GMAC. The problem with our rules was not just that they were insufficiently conservative, but that they were too narrowly applied. At the New York Fed, I was worried that the major banks we helped supervise didn’t have enough capital—as it turned out, some of them didn’t—but I was even more worried about the thinly capitalized nonbanks outside our purview. I tried to raise awareness about these problems through my speeches and our horizontal risk reviews, pushing for more rigorous stress-testing and more conservative risk management. We managed to fix some of the plumbing problems in the derivatives market through our work with the Fourteen Families. And we did push to limit leverage in unregulated hedge funds through the “indirect channel” of our supervision of the banks that lent them money.

  With the knowledge we have today, it’s clear we didn’t do enough. Before the crisis, I didn’t push for the Fed in Washington to strengthen the safeguards for banks, nor did I push for legislation in Congress to extend the safeguards to nonbanks. I also could have tried to use the indirect channel of supervision more aggressively to rein in other parts of the shadow banking system that were connected to traditional banks, like tri-party repo, asset-backed commercial paper, and the offbalance-sheet vehicles that ended up doing so much damage.

  Our financial reforms have dramatically improved the safeguards in the system, forcing banks to hold more capital and rely less on short-term funding. The requirements are even tougher for larger banks, whose failure can cause even more damage. And the new financial stability council is now responsible for monitoring risks throughout the system, with the power to extend enhanced regulation where it’s needed. Our oversight system is still flawed, but it now has more authority to respond to threats in the shadows. The Fed is also subjecting the largest banks to genuinely stressful stress tests, forcing them to imagine the unimaginable and maintain enough capital to survive
it.

  Our crisis, after all, was largely a failure of imagination. Every crisis is. For all my talk about tail risk, for all our concern about “froth,” we didn’t foresee how a nationwide decline in home prices could induce panic in the financial system sufficient to drag down the broader economy. But good crisis prevention does not depend on imagining the precise form of the unimaginable. You can’t expect to preempt surprises. You just have to recognize that surprises will come, and force the system to build stronger defenses that can help it withstand the extreme ones.

  THERE’S NO foolproof strategy to prevent financial fires, so policymakers also need to make sure they have adequate firefighting tools before fires start. If they don’t have the authority to act, their crisis-management skills won’t matter.

  Two of the most vital tools, fiscal stimulus and monetary stimulus, don’t require emergency authorities, but their effectiveness can still be enhanced or diminished in advance. A government in a strong fiscal position—with a modest budget deficit, if not a surplus, and a sustainable ratio of debt to GDP—will be in a better position to borrow at low rates to finance a massive temporary stimulus and maintain the stimulus as long as is necessary to cushion the effects of deleveraging. Similarly, on the monetary side, a central bank with a strong record of maintaining low inflation in the past will have more room to expand the money supply dramatically during a crisis without inviting too much fear of inflation in the future.