Midway through Timothy Geithner’s Stress Test, the former treasury secretary describes a late-2008 conversation with the then president-elect. Obama “wanted to discuss what he should try to accomplish.” Geithner’s reply was that his accomplishment would be “preventing a second Great Depression.” And Obama shot back that he didn’t want to be defined by what he had prevented.

Timothy Geithner
Timothy Geithner; drawing by John Springs

It’s an ironic tale for Geithner to be telling, although it’s not clear whether he himself realizes just how ironic. For Stress Test is meant to be a story of successful policy—but that success is defined not by what happened but by what didn’t. America did indeed manage to avoid a full replay of the Great Depression—an achievement for which Geithner implicitly claims much of the credit, and with some justification. We did not, however, avoid economic disaster. By any plausible accounting, we’ve lost trillions of dollars’ worth of goods and services that we could and should have produced; millions of Americans have lost their jobs, their homes, and their dreams. Call it the Lesser Depression—not as bad as the 1930s, but still a terrible thing. Not to mention the disastrous consequences abroad.

Or to use one of the medical metaphors Geithner likes, we can think of the economy as a patient who was rushed to the emergency room with a life-threatening condition. Thanks to the urgent efforts of the doctors present, the patient’s life was saved. But while the doctors kept him alive, they failed to cure his underlying illness, so he emerged from the procedure partly crippled, and never fully recovered.

How should we think about the economic policy of these past seven or so years? Geithner, while acknowledging the disappointments, would have us view it mainly as a success story, because things could have been much worse. And the middle third of his book, a blow-by-blow account of the acute phase of the financial crisis, carries the implicit and sometimes explicit message that things would indeed have been much worse but for the heroic actions of a handful of high officials, himself included.

But this still leaves open the question of whether things could and should have been considerably better, whether preventing a complete economic meltdown was all that could have been accomplished. Here Geithner implicitly says no—or at least that there was nothing more that he himself could have done.

I’ll return to the questions about Geithner’s role later. First, however, let’s examine the nature of the economic crisis we experienced, and why emergency treatments haven’t produced a full return to health.


Something went very wrong with the US economy in 2008. But what?

Quite early on, two somewhat different stories emerged about the economic crisis. One story, which Geithner clearly preferred, saw it mainly as a financial panic—a supersized version of a classic bank run. And there certainly was a very frightening panic in 2008–2009. But the alternative story, which has grown more persuasive as the economy remains weak, sees the financial panic, while dangerous in its own right, as a symptom of something broader and deeper—mainly a large overhang of private debt, in particular household debt.

What’s the difference? A financial panic is above all about confidence, or rather the lack thereof, and the overriding task of policy is to restore confidence. And one way to think about policy in the crisis is to say that people like Tim Geithner and Ben Bernanke dealt forcefully and effectively with the urgent task of restoring confidence in the financial system.

But confidence in itself is not enough to deal with the broader consequences of a debt overhang. That takes policies that go well beyond saving financial institutions—policies like sustained fiscal stimulus and debt relief for families. Unfortunately, such policies were never forthcoming on a remotely adequate scale, which is why true recovery has remained so elusive. And although Geithner denies it, one contributing factor to the inadequacy of policy was surely the fact that he seemed uninterested in, and maybe even hostile to, the policies we needed after the panic subsided.

So, about the panic: Geithner offers a very good and clear explanation of what financial panics are all about. As he says, they’re basically the bank run from the movie It’s a Wonderful Life writ large. Banks are, more or less by definition, institutions that promise their creditors—depositors if they’re conventional banks—ready access to their funds; but they invest in assets that are relatively illiquid, that is, can’t be converted into cash on short notice. The reason this works is that under normal conditions, only a small fraction of a bank’s depositors will try to pull their money out on any given day.


But the risk of a run is always there. Suppose that for some reason many depositors do decide to demand cash at the same time. The bank won’t have that much cash on hand, and if it tries to raise more cash by selling assets, it will have to sell those assets at fire-sale prices. The result is that mass withdrawals can break a bank, even if it’s fundamentally solvent. And this in turn means that when investors fear that a bank may fail, their actions can produce the very failure they fear: depositors will rush to pull their money out if they believe that other depositors will do the same, and the bank collapses.

Now, we have an answer to this danger: federal deposit insurance, which has made old-fashioned bank runs obsolete by assuring depositors that they won’t lose their money. The problem, it turned out, was that by the mid-2000s much of the US financial system—more than half of it, by Geithner’s reckoning—consisted of “shadow banks,” which didn’t rely on traditional deposits but instead raised money through various forms of short-run borrowing. Lehman Brothers, for example, relied heavily on “repo”—short-term loans, mainly overnight, with assets like mortgage-backed securities as collateral. What became apparent in 2008 was that shadow banks were every bit as vulnerable to runs as conventional banks, but lacked any kind of public safety net.

Worse, runs on shadow banks proved contagious. As investors pulled their funds from shaky firms, those firms were forced into frantic fire sales of assets; such sales depressed the prices of the collateral used by other banks, producing further investor panic.

The point is that a financial panic is very much a case of self-fulfilling prophecy. And there’s a classic way to deal with self-fulfilling panics—namely, for someone to step up as the “lender of last resort,” providing banks facing a run with cash, so that they don’t need to engage in desperate fire sales. When all goes well, the lender of last resort’s intervention can seem almost magical in its effects: in a matter of months or sometimes even days, markets revive, asset prices return to normal, and business as usual resumes.

All this is, as I said, well understood. The lender of last resort’s function was described at length in Walter Bagehot’s Lombard Street: A Description of the Money Market, published in 1873, and cited repeatedly in Stress Test. The Federal Reserve was itself created mainly to serve as a permanent lender of last resort; it was founded after the Panic of 1907, when major disruption was avoided only because J.P. Morgan organized an ad hoc coalition of bankers to act as lenders of last resort, and everyone realized that they couldn’t count on Morgan to be there forever.

So why was it so hard to organize an effective response to the 2008–2009 panic? One answer is that the Fed was set up to deal with conventional banks, and had neither a clear legal mandate for nor much experience in bailing out shadow banks. As a result, the tale of the bailout—which occupies about half of Geithner’s book—is one of frantic institutional creativity. Again and again Geithner, Bernanke, and other officials had to find clever ways to expand their mandate. They rescued Bear Stearns, which ordinarily wouldn’t have qualified for an injection of cash, by funneling the money through JPMorgan Chase. They got Goldman Sachs and Morgan Stanley, which are not now and have never been banks in the ordinary sense of the word, to nonetheless convert themselves officially into bank holding companies, to “create the impression that they were under the umbrella of Fed protection.” They put the Fed itself into the banking business, temporarily taking over most of the market for commercial paper—short-term business loans. And so on, month after month, page after page.

But the centrality of shadow banking wasn’t the only problem. There was also the very real possibility that the financial crisis was more than just a self-fulfilling prophecy, that large parts of the financial system would be bankrupt even if the panic subsided. If that was true, public assistance to institutions facing runs could lead to large taxpayer losses. Then what?

Fear of big bank losses led to a three-sided debate. On one side were the people who warned of “moral hazard”—that bailing out banks would reward bad actors, and encourage future irresponsibility. On another side were the nationalization people, myself among them (Geithner describes me as “the intellectual leader of the nationalization brigade”), who argued that the banks needed to be bailed out—the risks of financial collapse were too great otherwise—but that bank stockholders didn’t. The idea was that the government, in return for taking on big risks, should temporarily acquire ownership of the most troubled banks, so that taxpayers would profit if things went well. This wasn’t an outlandish position, by the way—it was what had happened in many historical bank bailouts, and was in effect what happened in the case of AIG, the giant insurance company, in which the government supplied cash and temporarily acquired a 79.9 percent ownership stake.


Finally, there was Geithner’s position, which was that despite its scale the financial crisis should be treated more or less as an ordinary lender-of-last-resort problem—that temporary nationalization would hurt confidence and was unnecessary, that once the panic subsided banks would be OK. A principal part of Geithner’s argument against nationalization was the belief that a “stress test” of banks would show them to be in fairly decent shape, and that publishing the results of such a test would, in conjunction with promises to shore up banks when necessary, end the crisis. And so it proved. He was right; I was wrong; and the triumph of the stress test gave him the title for his book.

In some ways, one has to say, the success of the stress test remains something of a mystery—part of the larger mystery of why the US financial bailout was so cheap. Yes, that’s right: in the end, the direct costs of bailing out banks and other financial institutions were remarkably low. Historically, taxpayers have typically been called on to swallow large losses in the course of cleaning up banking crises. Even the savings and loan crisis of the 1980s, which had little effect on the overall economy, ended up costing taxpayers around 5 percent of GDP. But the cleanup from the biggest financial crisis since the 1930s ended up costing less than a tenth of that. Geithner never really explains why.


Whatever the reasons, however, the stress test pretty much marked the end of the panic. The graph on this page shows several key measures of financial disruption—the TED spread, an indicator of perceived risks in lending to banks, the commercial paper spread, a similar indicator for businesses, and the Baa spread, indicating perceptions of corporate risk. All fell sharply over the first half of 2009, returning to more or less normal levels. By the end of 2009 one could reasonably declare the financial crisis over.

But a funny thing happened next: banks and markets recovered, but the real economy, and the job market in particular, didn’t.


According to the independent committee that officially determines such things, the so-called Great Recession ended in June 2009, around the same time that the acute phase of the financial crisis ended. Most Americans, however, disagree. In a March 2014 poll, for example, 57 percent of respondents declared that the nation was still in recession.

It’s not hard to understand where this grim assessment comes from. Output and employment have indeed been growing over the past five years, but slowly. Thanks to the depressed state of the economy, it’s still very hard to find a full-time job—both the number of long-term unemployed workers and the number of people unable to find full-time jobs remain far above pre-crisis levels.

Clearly, restoring confidence in the financial sector, while it may have been necessary to avoid a complete economic meltdown, wasn’t enough to jump-start a strong recovery. Why not?

The best working hypothesis seems to be that the financial crisis was only one manifestation of a broader problem of excessive debt—that it was a so-called “balance sheet recession.” Curiously, while Geithner repeatedly refers to the classic Bagehot analysis of financial panic, he never mentions the almost equally classic analysis of “debt deflation” by the American economist Irving Fisher, who laid out the basics of the balance sheet view back in 1933. Yet a Fisher-type interpretation of our economic troubles has seemed increasingly relevant as financial markets flourish but the real economy remains stubbornly weak.

The logic of a balance sheet recession is straightforward. Imagine that for whatever reason people have grown careless about both borrowing and lending, so that many families and/or firms have taken on high levels of debt. And suppose that at some point people more or less suddenly realize that these high debt levels are risky. At that point debtors will face strong pressures from their creditors to “deleverage,” slashing their spending in an effort to pay down debt. But when many people slash spending at the same time, the result will be a depressed economy. This can turn into a self-reinforcing spiral, as falling incomes make debt seem even less supportable, leading to deeper cuts; but in any case, the overhang of debt can keep the economy depressed for a long time.

Where do bank runs fit into this story? Well, it’s easy to see how fears about excessive debt can raise concerns about the solvency of banks, which in turn can start a process of self-fulfilling panic. The point, however, is that even if the panic can be contained, the problem of excessive debt remains. And that, arguably, is why the bank bailouts of 2008–2009 didn’t lead to a satisfactory economic recovery.

What would it have taken to do better?

Unlike a financial panic, a balance sheet recession can’t be cured simply by restoring confidence: no matter how confident they may be feeling, debtors can’t spend more if their creditors insist they cut back. So offsetting the economic downdraft from a debt overhang requires concrete action, which can in general take two forms: fiscal stimulus and debt relief. That is, the government can step in to spend because the private sector can’t, and it can also reduce private debts to allow the debtors to spend again. Unfortunately, we did too little of the first and almost none of the second.

Yes, there was the American Recovery and Reinvestment Act, aka the Obama stimulus, and it surely helped end the economy’s free fall. But the stimulus was too small and too short-lived given the depth of the slump: stimulus spending peaked at 1.6 percent of GDP in early 2010 and dropped rapidly thereafter, giving way to a regime of destructive fiscal austerity. And the administration’s efforts to help homeowners were so ineffectual as to be risible.

So where was Timothy Geithner in all this?


There’s a curious change in tone about two thirds of the way through Stress Test. Up to that point—basically, up to the stress test itself and its immediate aftermath—Geithner tells a tale of heroic activism, of good men and women pulling out all the stops to save the world. Thereafter, however, Geithner turns apologetic and self-exculpatory. He acknowledges that more stimulus and debt relief would have been good things; he claims that he wanted to do much more, but that practical difficulties and political opposition made stronger action impossible. The can-do hero of the financial crisis, endlessly creative in finding ways to bypass institutional and political obstacles to do what needs to be done, suddenly becomes a passive observer of events.

Is that really how it was? I’m sorry to say this, but Geithner doesn’t appear to be a reliable narrator here.

Take the question of stimulus. “We all felt the stimulus should be as big as possible,” Geithner tells us. But the memo on economic stimulus that was presented to the president-elect in December 2008—a memo that reflected a lot of input from Geithner—warned against a big plan: “An excessive recovery package could spook markets or the public and be counterproductive.” And Geithner reportedly snapped at Christina Romer, the chair of the president’s Council of Economic Advisers, that stimulus is “sugar.”

Geithner also makes some demonstrably false statements about the public debate over stimulus. “At the time,” he declares, “$800 billion over two years was considered extraordinarily aggressive, twice as much as a group of 387 mostly left-leaning economists had just recommended in a public letter.” Um, no. A number of economists, including Columbia’s Joseph Stiglitz and myself, were warning that the package was too small; so was Romer, internally. And that economists’ letter called for $300 to $400 billion per year. The Recovery Act never reached that level of spending; even if you include tax cuts of dubious effectiveness, it only briefly grazed that target in 2010, before rapidly fading away.

And then there’s the issue of debt relief. Geithner would have us believe that he was all for it, but that the technical and political obstacles were too difficult for him to do very much. This claim has been met with derision from Republicans as well as Democrats. For example, Glenn Hubbard, who was chief economic adviser under George W. Bush, says that Geithner “personally and actively opposed mortgage refinancing.”

Furthermore, Geithner seems to want it both ways—to portray himself as a frustrated advocate of more debt relief, while at the same time asserting that such relief would have made little difference. And his eagerness to make the latter assertion leads him to engage in some demonstrably bad arithmetic. The economists Atif Mian and Amir Sufi are our leading experts on the problems created by debt overhang (and the authors of an important new book on the subject, House of Debt); they looked at Geithner’s claims about the benefits of debt relief to the economy and showed that they are absurdly low, far below anything current research suggests.

The best guess is that Geithner was in fact unenthusiastic about stimulus and more or less hostile to mortgage debt relief. But did this matter? You can argue that a bigger stimulus plan would have failed to pass Congress; you can argue that mortgage refinancing would either have proved impossible to implement or have provoked a huge political backlash. The truth is that we’ll never know, because the Obama administration never really tried to push the envelope on either fiscal policy or debt relief. And Geithner’s influence was probably an important reason for this caution. Geithner saw the economic crisis as more or less entirely a matter of lost confidence; he believed that restoring that confidence by saving the banks was enough, that once financial stability was back the rest of the economy would take care of itself. And he was very wrong.


Stress Test concludes on a note of celebration. Yes, mistakes were made, Geithner concedes, but on the whole, he tells us, Washington rose to the occasion, doing what was necessary to prevent another depression.

To the rest of us, however, the victory over financial crisis looks awfully Pyrrhic. Before the crisis, most analysts expected the US economy to keep growing at around 2.5 percent per year; in fact it has barely managed 1 percent, so that our annual national income at this point is around $1.7 trillion less than expected. Headline unemployment is down, but that’s largely because many workers, despairing of ever finding a job, have stopped looking. Median family income is still far below its pre-crisis level. And there’s a growing consensus among economists that much of the damage to the economy is permanent, that we’ll never get back to our old path of growth.

The only way you can consider this record a success story is by comparing it with the Great Depression. And that’s a pretty low bar—after all, aren’t we supposed to know more about economic management than our grandfathers did?

In fact, we did have both the knowledge and the tools to fight this disaster. We know a lot about how fiscal policy works, and the United States clearly had the borrowing capacity to spend more on fighting unemployment. Whatever Geithner may say, it’s clear that a lot more could also have been done to reduce the burden of mortgage debt. Yet we didn’t do what needed to be done.

I like Geithner’s metaphor of a stress test—and his book is very much worth reading, especially for its account of the crisis. But he’s wrong about the outcome of that test. We can argue about how much of the blame rests with the Obama team, how much with the crazies in Congress who met every administration initiative, no matter how reasonable, with scorched-earth opposition. But the overall grade seems clear. We didn’t pass the test—we failed, badly.