But the collapse of that famous firm did more than anything else to focus the minds of policy makers on the reality that the risk of another Great Depression loomed. At the end of 2008 they looked into the abyss and got religion. They started deploying all the weapons in their a.r.s.enal. Some tactics, like cutting interest rates, came from the standard playbook. But many others seemed to come from another world, and in some cases another era. To the uninitiated, the names of these tactics-"quant.i.tative easing," "capital injections," "central bank swap lines"-defy definition. But these and many other unorthodox weapons came off the shelf and were mustered into battle. Some had been tried before; others had not. Some worked; some did not.

Nonetheless, their collective effect arguably prevented the Great Recession from turning into another Great Depression. Whether the cure will turn out to be worse than the disease is another matter, and it is to that question-and the risks and rewards of using unconventional policy measures to deal with financial crises-that we turn next.

Chapter 6.

The Last Resort.

When the worst financial crisis in generations. .h.i.t the United States in 2007, Ben Bernanke had just been appointed head of the Federal Reserve a year earlier. It was a remarkable coincidence, for Bernanke was not just any central banker; he was one of the world's leading authorities on the Great Depression. Far more than almost any living economist, Bernanke was acutely aware of the complicated dynamics behind this watershed event. Over the course of his academic career, he had written influential articles that helped untangle the causes and effects of the worst depression in the nation's history.

Bernanke self-consciously built on the pioneering work of monetarists Milton Friedman and Anna Jacobson Schwartz, whose writings he first encountered in grad school. As we saw in chapter 2, these two scholars had broken with earlier interpretations of the Great Depression by arguing that monetary policy-courtesy of the Federal Reserve-was to blame for the disaster. According to this interpretation, the Fed's inaction and inept.i.tude not only failed to prevent the unfolding disaster but even contributed to the problem. Bernanke elaborated on that thesis, showing how the consequent collapse of the financial system threw sand in the gears of the larger economy, dragging the nation into a brutal depression.

Bernanke's keen appreciation of the burdens of history and his debt to Friedman were evident when he attended the venerable economist's ninetieth birthday party in 2002. By then Bernanke was a governor on the board of the Federal Reserve, and when he stood up to give a speech, he famously turned to the elderly man and said, with regard to the Great Depression: "You're right, we did it. We're very sorry. But thanks to you, we won't do it again."

This was the man in charge of monetary policy when the crisis. .h.i.t. Not surprisingly, he saw events through the prism of what had happened nearly eighty years earlier and acted accordingly. Rules would be broken, and new tools tried. There would be no repeat of the Great Depression. As he told a reporter in the summer of 2009, "I was not going to be the Federal Reserve chairman who presided over the second Great Depression."

To that end, Bernanke revolutionized monetary policy, directing a stunning series of interventions into the financial system that even today few people understand. Some of these moves Bernanke had antic.i.p.ated making; others he developed as the months pa.s.sed and the threat of deflation and even a depression increased. They ran the gamut from conventional monetary policy-slashing interest rates to zero, for example-to unprecedented measures heralding a ma.s.sive expansion of the Federal Reserve's power over the economy.

These interventions probably did help avert a twenty-first-century Great Depression, but for the student of crisis economics they raise a host of unsettling issues. Aside from the difficulty of scaling back Bernanke's policies once they're in place, many of them may prove conducive to moral hazard on a grand scale. The Fed, in its rush to prop up the financial system, rescued both illiquid and insolvent financial inst.i.tutions. That precedent may be hard to undo and, over the long run, may lead to a collapse of market discipline, which in turn may sow the seeds of bigger bubbles and even more destructive crises.

No less problematic is the fact that some of Bernanke's monetary policies infringe on the traditional fiscal powers of elected government-namely, the power to spend money. In the recent crisis, the Fed pushed the statutory envelope, a.s.suming various powers, implied and otherwise, to swap safe government bonds for toxic a.s.sets and, more radical, to purchase toxic a.s.sets and hold them on its balance sheet. Such measures, even if they prove effective, amount to an end run around the legislative process.

Bernanke's response, orchestrated by himself and other central bankers, offers a glimpse of the unorthodox ways in which monetary policy can be used-and perhaps abused-to prevent a crisis from spiraling out of control.

Deflation and Its Discontents.

Since the end of the Second World War, the American business cycle has followed a fairly predictable path. The economy would emerge from a recession, grow, and eventually boom; the Federal Reserve would then begin to bring the cycle to a close by hiking interest rates to keep inflation in check, and more broadly, to keep the economy from overheating. Inevitably, the economy would contract; a recession would ensue.

In some cases, most notably in 1973, 1979, and 1990, the recession was set off in part by what economists call an exogenous negative supply-side shock. All three times, a geopolitical crisis in the Middle East triggered a sudden rise in oil prices that sparked inflation. Here too, to control rising prices, the Fed moved interest rates higher, after which the economy started to contract.

Whatever their causes, these various contractions would inevitably moderate inflation, without eliminating it altogether. The fall in output or the gross domestic product-typically a single percentage point or two-led to unpleasant but tolerable increases in unemployment and the familiar hardships of a recession.

In some instances, the economy would grow again of its own accord; in others, policy makers facilitated a recovery by resorting to a time-honored tool: they would cut interest rates, effectively making it cheaper for households and firms to borrow money. This would nudge people to spend more, driving up demand for everything from houses to factory equipment. Cutting interest rates often had the added effect of driving down the value of the dollar, making exports more attractive, making imports more expensive, generating demand for domestic goods, and contributing to an eventual recovery. Fiscal stimulus was also used to restore growth.

The first ten recessions in the postwar United States largely followed this script. Most lasted less than a year, save for a nasty recession in the wake of the oil shock of 1973, which was triggered by the Yom Kippur War; and after a second oil shock in 1979 caused by the Iranian Islamic Revolution, the Federal Reserve used high interest rates to slay inflation, resulting in a far more unpleasant recession. While brutal, that campaign proved successful and set the stage for the much-celebrated Great Moderation. As a consequence, recessions in 1991 and 2001 lasted a mere eight months each, and while these downturns brought pain aplenty, they ended with renewed growth and optimism, thanks in part to varying doses of monetary easing, fiscal stimulus, and tax cuts.

The twelfth postwar recession, which took hold in the wake of the recent financial crisis, has been different. Prices not only moderated but in some cases registered declines for the first time in fifty or sixty years. This was deflation, a phenomenon that unnerved policy makers across the ideological spectrum. Its recurrence "gives economists chills," reported The New York Times in the fall of 2008. The following spring Bernanke explained, "We are currently being very aggressive because we are trying to avoid . . . deflation."

To the uninitiated, the fuss seemed a bit mystifying. After all, aren't falling prices a good thing? Consumer goods cost less; people can buy more with every dollar they own; what's not to like? In fact, in a handful of episodes small, steady rates of deflation have gone hand in hand with robust economic growth, as technological advances drove down the price of goods. Between 1869 and 1896, for example, the spread of railroads and new manufacturing techniques helped push down prices by some 2.9 percent a year. At the same time, despite recurrent crises, the economy grew at an average annual rate of 4.6 percent.

This episode remains something of a curiosity for economic historians because deflation is generally not compatible with economic growth. Why? In most cases, deflation isn't caused by a technological revolution; it's caused by a sharp fall of aggregate demand relative to the supply of goods and the productive capacity of the economy.

This more common kind of deflation can have all sorts of peculiar effects on the day-to-day functioning of the economy. It can deter consumers from spending on big-ticket items: buying a car or a house, for example, becomes a bit like catching a falling knife. Similarly, a factory contemplating some capital investments may prefer to remain on the sidelines until prices stop falling. Unfortunately, postponing spending, far from stimulating economic growth, does precisely the opposite.

A bout of deflation born of a financial crisis is of a different order altogether and may be far more dangerous and destructive. Such bouts were relatively common in the wake of the perennial crises of the nineteenth century, then became much rarer in the twentieth. While deflation accompanied the global depression of the 1930s, it largely disappeared after that watershed event. Only in the 1990s did it resurface, first after the collapse of j.a.pan's a.s.set bubble, and then during the brutal recession that hit Argentina in 1998-2001.

During the recent crisis, the prospect of this kind of deflation was what gave economists the chills. They knew well that its ill effects could ramify throughout the economy. Even if it doesn't end in an outright depression, deflation can suffocate growth for years, leading to a condition that might best be described as stag-deflation, in which economic stagnation and even recession are combined with deflation. In such a condition, the usual tools of monetary policy cease to have much effect.

Irving Fisher was one of the first economists to understand the dynamics of deflation. While Fisher remains infamous today for claiming, shortly before the market crashed in 1929, that stock prices would remain on a "permanently high plateau," he redeemed himself by subsequently articulating a compelling theory of the connection between financial crises, deflation, and depression, or what he called the "debt-deflation theory of great depressions." Put simply, Fisher believed that depressions became great because of two factors: too much debt in advance of a crisis, and too much deflation in its wake.

Fisher began by observing that some of the worst crises in American history-1837, 1857, 1893, and 1929-followed on the heels of an excessive acc.u.mulation of debt throughout the economy. When the shock came-the stock market crash of 1929, for example-margin calls led to frenzied attempts to pay down debt. Fisher believed that this rush to liquidate debt and stockpile liquid reserves, while rational, damaged the health of the larger economy. As he explained in 1933, "The very effort of individuals to lessen their burden of debts increases it, because of the ma.s.s effect of the stampede to liquidate . . . the more debtors pay, the more they owe." Fisher famously noted that from October 1929 to March 1933, while debtors frantically reduced the nominal value of their debt by 20 percent, deflation actually increased their remaining debt burden by 40 percent.

Why? The rush to liquidate a.s.sets at fire-sale prices, Fisher argued, would lead to falling prices for everything from securities to commodities. Supply would far outstrip demand, and prices would fall. At the same time, people would tap money deposited in banks in order to liquidate debts or as a precaution against bank failures. These withdrawals would lead to a reduction of what economists call "deposit currency" and, by extension, a contraction of the overall money supply. This contraction would depress prices still further. As prices continued to fall, the value of a.s.sets across the board would drift downward, triggering a commensurate decline in the net worth of banks and businesses holding those a.s.sets. More fire sales and more deflation would result, leading to less liquidity in the markets, more gloom and pessimism, more h.o.a.rding of cash, and more fire sales.

The resulting deflation would have perverse consequences. As borrowers moved to pay off their debts (and as aggregate demand for goods started to fall in a severe recession), the lowered prices of goods and services would paradoxically increase the purchasing power of the dollar, and by extension, the real burden of their remaining debt. In other words, deflation increases the real value of nominal debts. Instead of getting ahead of their debts, people fell behind. Fisher called this the "great paradox"-the more people pay, the more their debts weigh them down.

This is debt deflation. To understand it better, let's consider its counterpart, what might be called "debt inflation." Imagine that you are a firm or a household, and you take out a ten-year loan for $100,000 at an interest rate of 5 percent. At the time, inflation hovers around 3 percent. If inflation stays at this rate, you'll really be paying interest at 2 percent per year-that's what's left after inflation eats away at the nominal, or original, rate of interest. If inflation goes up to 5 percent a year, it will effectively wipe out the interest rate entirely, and you will have the equivalent of an interest-free loan. But if inflation runs out of control, hitting 10 percent, you're not only getting an interest-free loan; your princ.i.p.al is eroding as well. These examples show you how to calculate the "real interest rate"-the difference between the nominal interest rate and the inflation rate.

Confused? Let's think about a more extreme example. Imagine that you take out that same $100,000 loan-and inflation runs completely out of control. Prices and wages soar to astonishing levels. It used to cost a dollar to buy a loaf of bread; now it costs a thousand dollars. At the same time, a minimum-wage job that once paid peanuts now pays several million dollars a year; a "good" job pays a hundred million. Now go back to that $100,000 debt you incurred. It's still sitting there, denominated in those older, more valuable dollars. The amount of the princ.i.p.al has not changed with inflation. It's now much easier to pay off your loan. Heck, it's nothing more than a month's worth of groceries.

The key here is that the dollars you're using to pay off the debt are worth less than when you incurred the debt in the first place. For this simple reason, inflation is the debtor's friend: it effectively erodes the value of the original debt.

Deflation, however, is not the debtor's friend. Let's go back to our original example of a ten-year loan at an interest rate of 5 percent. Contrary to expectations, the economy experiences deflation of 2 percent. That means you're effectively paying 7 percent interest a year. If deflation hits 5 percent, your real borrowing costs have doubled to 10 percent a year. In other words, the dollars you're using to pay off your debt are worth more than they were when you incurred the debt in the first place. Unfortunately, even though each dollar is worth more, you now have fewer of them because your wages have declined.

The upshot of debt deflation is that debtors-households, firms, banks, and others-see their borrowing costs rise above and beyond what they originally antic.i.p.ated. And during a major financial crisis-with rising unemployment, growing panic, and a general unwillingness to lend-anyone who owes money has much more difficulty making good on his debt or, alternatively, refinancing it on less onerous terms. Investors shun risky a.s.sets, seeking liquid and safe a.s.sets like cash and government bonds. People h.o.a.rd cash and refuse to lend it, which only exacerbates the liquidity crunch. As credit dries up, more and more people default, feeding the original cycle of deflation, debt deflation, and further defaults.

The end result is a depression: a brutal economic collapse in which a nation's economy can contract by 10 percent or more. In the Great Depression that both traumatized and inspired Irving Fisher, the collapse was unprecedented. From peak to trough, the stock market lost 90 percent of its value, the economy contracted by close to 30 percent, and 40 percent of the nation's banks failed. Unemployment surged to close to 25 percent. And deflation? Prices fell off the cliff. A dozen eggs that cost $0.53 in 1929 cost $0.29 in 1933, a drop of some 45 percent. Comparable declines. .h.i.t everything from people's wages to the price of gas.

It's no surprise that Fisher's vision was a dark one. As he wrote from the depths of the crisis in 1933, "Unless some counteracting cause comes along to prevent the fall in the price level, such a depression . . . tends to continue, going deeper, in a vicious spiral, for many years. There is then no tendency of the boat to stop tipping until it has capsized." While Fisher acknowledged that things might ultimately stabilize-after "almost universal bankruptcy"-he thought this to be "needless and cruel." Instead, he counseled that policy makers "reflate" prices up to precrash levels. As he put it, "If the debt-deflation theory of great depressions is essentially correct, the question of controlling the price level a.s.sumes a new importance; and those in the drivers' seats-the Federal Reserve Board and the Secretary of the Treasury-will in [the] future be held to a new accountability."

Those words likely haunted Ben Bernanke, Henry Paulson, and Timothy Geithner as they confronted what looked like a reprise of the Great Depression. Unfortunately, like almost everything else with financial crises, engineering a reflation-or to put it more baldly, creating inflation-is not as simple as it seems. Once a deflationary spiral has gained momentum, conventional monetary policy tends not to work. Nor does it work against other ills that accompany financial crises. Other weapons must be developed and thrown into battle.

The Liquidity Trap.

When economists talk about the futility of ordinary monetary policy, they refer to a "liquidity trap." Policy makers dread this state of affairs, and to understand why, we must examine how central banks exercise control over the money supply, interest rates, and inflation.

In the United States, the Federal Reserve primarily controls the money supply through "open market operations": that is, it can wade into the secondary market and buy or sell short-term government debt. When it does so, it effectively adds or removes money from the nation's banking system. It thereby changes what is known as the "Federal funds rate," the interest rate banks charge each other for overnight loans for funds on deposit at the Federal Reserve. In normal times, the Federal funds rate is a proxy for the cost of borrowing at any number of levels of the economy, and manipulating it is one of the most effective tools at the disposal of the Fed.

Here's how it works. Let's say that the Fed is worried about inflation and wants to keep the economy from overheating. The Fed therefore goes out and sells $10 billion worth of short-term government debt. By doing so, it effectively removes money from the banking system. Why? Because the purchasers of the debt have to write checks drawn on their respective banks, which the Fed then cashes and keeps. The banking system and the larger economy are now out $10 billion. Moreover, because banks use every dollar on deposit to create many more dollars' worth of loans, the real hit to the banking system-and by extension, the money supply-is something approaching $25 billion or $30 billion.

In this way, the Fed has tightened the money supply and made credit harder to obtain: it has effectively raised the cost of borrowing. Money, like any other commodity, responds to the laws of supply and demand, and now that the supply is lower, borrowing money costs more. Interest rates, in other words, go up because lenders can now command a higher rate. Whenever the media report that the Federal Reserve has "raised" interest rates, it hasn't literally done so; rather, it has "targeted" a higher interest rate-the Federal funds rate-via these open market operations.

Now let's imagine that the Fed is no longer worried about inflation; in fact, it's worried about the fact that the economy, instead of overheating, is headed toward a recession. The Fed therefore sets a lower target for the Federal funds rate and floods the economy with money, buying up short-term government debt. Where does it get the money? It creates it out of thin air. The Federal Reserve effectively writes a check for $10 billion and gives it to the sellers of government debt. These sellers deposit the money they've received from the Fed in various banks. Now those banks can use it to make loans worth several times that amount. Money is suddenly more available, and as a consequence, credit is easier to obtain. More to the point, it's cheaper: the net effect of adding money to the economy is that the Federal funds rate will fall, as will interest rates generally.

This is what takes place during normal times. A liquidity trap, by contrast, is not normal. It's what happens when the Fed has exhausted the power of open market operations. That dreaded moment arrives when the Federal Reserve has driven the Federal funds rate down to zero. In normal times setting that rate would pump plenty of easy money and liquidity into the economy and spur wild growth. But in the wake of a financial crisis, cutting interest rates to zero may not be enough to restore confidence and compel banks to lend money to one another. The banks are so worried about their liquidity needs-and so mutually distrustful-that they will h.o.a.rd any liquid cash rather than lend it out. In this climate of fear, the policy rate may be zero, but the actual market rates at which banks are willing to lend will be much, much higher, keeping the cost of borrowing expensive. Because it's almost impossible to drive policy rates below zero-you can't make banks lend money if they'll be penalized for doing so-policy makers find themselves in a serious quandary. They're in the dreaded liquidity trap.

During the recent crisis, central banks around the world found themselves in precisely this position. As the crisis worsened, they slashed interest rates, and by late 2008 and 2009 the Federal Reserve, the Bank of England, the Bank of j.a.pan, the Swiss National Bank, the Bank of Israel, the Bank of Canada, and even the European Central Bank had pushed interest rates close to zero. Compared to previous financial crises, this exercise of monetary policy was remarkably swift and partially coordinated. But the collective cuts did little to stimulate loans, much less consumption, investment, or capital expenditures, as market rates remained very high given the fear and uncertainty that gripped banks, households, and firms. Nor did these cuts arrest the slide toward deflation. Conventional monetary policy ceased to have sway over the markets. The metaphor of choice was that exercising monetary policy was like "pushing on a string." It was useless.

The reason was simple: the cuts in the Federal funds rate (or its equivalent in other countries) did not percolate throughout the wider financial system. Banks had money, but they didn't want to lend it: uncertainty bred by the crisis, and concerns that many of their existing loans and investments would eventually sour, made them risk averse. This failure of conventional monetary policy nicely ill.u.s.trated an old adage: you can lead a horse to water, but you can't make it drink. The Fed could pump plenty of water or liquidity into the banks, but it could not make them lend. If they did anything with their excess reserves, they sank them into the closest thing to cash: risk-free government debt.

We can glimpse the liquidity trap in the gap or "spread" between interest rates paid on supersafe or otherwise solid investments and those paid on riskier investments. There are many ways of measuring this spread. For example, the "TED spread" is the difference between the interest rate on the short-term government debt of the United States and the three-month LIBOR (see chapter 1), the interest rate that banks charge one another for three-month loans. During normal times, the TED spread hovers around 30 basis points, reflecting the fact that the market deems bank-to-bank loans as only slightly riskier than loans to the government.

At the height of the crisis, the TED spread hit 465 basis points, because banks no longer trusted one another enough to lend money on a three-month horizon, except at exorbitant rates. At the same time, risk-averse investors fled to the haven of the safest a.s.set of all: the debt of the U.S. government. These forces conspired to simultaneously drive up the cost of borrowing for banks and drive down the cost of borrowing for the U.S. government. The widening spread was a reflection of this dynamic, and the higher the spread, the greater the stress in the markets. So while the Fed was willing to lend money at low rates, the actual market rates at which banks lent to one another-the LIBOR-remained very high. Worse, because the rates of many other kinds of short-term loans and of variable-rate mortgages are pegged in part to the LIBOR, borrowing remained very high for private firms and households.

Measurements like the TED spread are a bit like blood pressure readings: they reflect the underlying health of the economy's circulatory system. They reveal how readily money flows through the economy, or how "liquid" markets are at a given moment. When conditions are normal, markets are relatively liquid and trust rules; people lend money to one another with ease, and borrowing costs remain at normal levels. In a time of crisis, when the patient (the financial system) is very sick indeed, the lifeblood of the system (money) isn't flowing, despite the usual measures used to keep it healthy: namely, pursuing open market operations to achieve lower interest rates. Deflation becomes a very real possibility.

How does one deal with this sort of problem? Back in 2002, when Bernanke spoke about the perils of deflation, he alluded to a number of possible interventions. As he recognized at that time, these experimental measures carried significant risks, given "our relative lack of experience with such policies," as he rightly characterized it. The j.a.panese had experimented with some of these policies in the 1990s, but they remained highly controversial.

When the crisis. .h.i.t, Bernanke inst.i.tuted a series of such measures, aimed at cutting the spreads between the short-term-and subsequently, the long-term-rates set by the market and the short-term rates set by policy makers. To accomplish this feat, the Fed set up a series of new "liquidity" facilities that made low-cost loans available to anyone who needed them. In effect, the government jumped directly into the market, reaching far beyond the usual mechanisms of injecting liquidity-cutting the overnight Federal funds rate-and made loans directly to ailing financial inst.i.tutions. It became the quintessential lender of last resort, making loans and liquidity available to an ever-widening cross section of the financial system.

Initially, the Fed aimed these maneuvers at inst.i.tutions-depository inst.i.tutions or banks-that already had some rights to borrow overnight funds directly from the Federal Reserve, from the "discount window" (the term refers to an earlier era, when cash-strapped banks would literally go to a teller window at the Fed). Few banks exercised this right, simply because in normal times the Fed imposed a penalty rate on anyone who approached the discount window. The window was designed to make small, emergency loans; it wasn't designed for a crisis. As conditions worsened, however, the Fed cut the borrowing penalty and allowed banks to obtain loans for longer periods of time. By March 2008, banks could borrow for up to ninety days from the discount window, with almost no penalty.

Yet the crisis worsened, whereupon the Fed then introduced new liquidity facilities. The Term Auction Facility (TAF) targeted depository inst.i.tutions, giving them another means of securing ready cash for periods much longer than overnight. But it did little to stop the liquidity crunch or the ugly cycle of fire sales, forced liquidations, and declining a.s.set prices that Fisher had predicted. The Fed had to adopt other tools aimed at the parts of the financial system that had no existing access to its resources.

Accordingly, the Federal Reserve established the Primary Dealer Credit Facility (PDCF), which made overnight loans to "primary dealers," the banks and broker dealers with whom the Fed trades when it conducts open market operations. Another facility, the Term Securities Lending Facility (TSLF), made loans of medium-term maturity to the same group, in exchange for illiquid securities held by such inst.i.tutions. Thus, for the first time since the Great Depression, the Fed used its emergency powers to lend to nondepository inst.i.tutions. From there the facilities multiplied, with acronyms to rival anything devised during the New Deal: the Commercial Paper Funding Facility (CPFF), the Money Market Investor Funding Facility (MMIFF), and most unp.r.o.nounceable of all, the a.s.set-Backed Commercial Paper Money Market Mutual Fund Liquidity Fund (ABCPMMMFLF), better known simply as the AMLF.

This alphabet soup of lending facilities operated in a variety of different ways and had different objectives or targets. Sometimes the facilities permitted financial inst.i.tutions to borrow directly from the Fed. In other cases, they enabled financial inst.i.tutions to swap illiquid a.s.sets-higher-quality a.s.set-backed securities, corporate bonds, commercial paper-for supersafe and liquid government debt. In still other cases, the facilities directly or indirectly financed the purchase of illiquid short-term debt. Whatever the mechanism, the objective was the same: inject liquidity into specific markets that showed signs of trouble and stress. This unprecedented intervention was not as indiscriminate as it might seem. The Federal Reserve did not accept junk bonds or other low-grade debt as collateral; it accepted only what was, in theory, higher-quality debt.

These efforts eventually bore some fruit: at the end of 2008, in the aftermath of the Lehman collapse, the Fed and other central banks flooded the financial markets with hundreds of billions of dollars' worth of liquidity, and the spreads between short-term market rates and safe government a.s.sets started to decline. As c.u.mbersome and radical as these measures were, they successfully injected a measure of liquidity into the short-term credit markets. Nonetheless, it was arguably a Pyrrhic victory. The Federal Reserve and other central banks that inst.i.tuted comparable programs had gone from being lenders of last resort to lenders of first, last, and only resort. In the process, they crossed the proverbial Rubicon not once or twice but many times.

In normal times, the lender of last resort helps individual banks with liquidity problems. But in this particular crisis, central banks ended up providing support to virtually every bank. And they did so not simply in the form of overnight loans, as is usually the case; this time the liquidity crunch was so severe that the Fed lent money for weeks or even months. In addition, it lent to inst.i.tutions that had never before been recipients of such aid: the primary dealers, which included many firms that weren't banks in any sense of the word, and the money market funds. The Fed even effectively lent money to corporations via the CPFF. It also provided "liquidity support"-special low-cost lines of credit-to a host of inst.i.tutions considered too big to fail: AIG, Fannie Mae and Freddie Mac, and Citigroup. Central bankers in Europe adopted similar measures.

These interventions had little or no precedent in the history of central banking. They amounted to a ma.s.sive expansion of government support of the financial system. But they were only the beginning.

Last Lender Standing.

As a typical crisis gathers steam, runs against a nation's banks and other financial inst.i.tutions take place. Depositors in Mexico demand their pesos back; investors in j.a.pan demand the return of the yen they've lent out. It's an unpleasant scene, but the central bank in each of those nations can save the day because it can print money to meet the demands. The domestic currency is in demand, and to quell the panic, the central banks can provide it.

But when the liabilities of financial inst.i.tutions, corporations, households, or even the government are denominated in a foreign currency, the situation can unravel. Emerging-market economies may end up getting much of their financing from banks and other financial inst.i.tutions in other countries. The foreign currency in question is most often the dollar, but it could also be the euro or any number of different currencies.

If for some reason the creditors of an emerging-market economy decide not to roll over its debt when it comes due, then anyone who owes dollars has to pay off the debt. That puts debtors in a tight spot: they don't have the dollars. They can go to the central bank, but it is unlikely to have stockpiled ma.s.sive foreign currency reserves, and it can't help out. Nor can it print dollars: that would be counterfeiting. So these debtors are extraordinarily vulnerable. Their predicament has been at the heart of a number of recent emerging-market crises: Mexico in 1994, East Asia in 1997 and 1998, Russia and Brazil in 1998, and Turkey and Argentina in 2001.

Enter the International Monetary Fund. The IMF was born at the end of World War II; one of its princ.i.p.al responsibilities has been to act as an international lender of last resort to governments and central banks who find themselves in the position so many countries did in the 1990s. The IMF was busy that decade, but in the 2000s the world's emergency-room doctor had little to do-until the crisis. .h.i.t. Then the IMF once again became the world's lender of last resort to a host of emerging-market countries.

It gave this support in two forms. It extended the more traditional lifeline, a Stand-By Arrangement (SBA), to fourteen countries, with Hungary, Ukraine, and Pakistan among the biggest recipients. As with the support given to emerging markets in the 1990s, the IMF made these foreign-currency loans only if the recipients adopted economic reforms that would in theory put them on more stable ground in the future. Other more stable countries with a stronger track record of inst.i.tuting financial reforms-Mexico, Poland, and Colombia-tapped unconditional lines of liquidity known as Flexible Credit Lines (FCLs). Unlike SBAs, FCLs served as precautionary or prophylactic lines of credit: the IMF effectively pledged to help out but did not immediately disburse any money.

The scale of all this lending was remarkable. By the summer of 2009, the IMF had authorized over $50 billion in SBAs and $78 billion in FCLs. Many of these lifelines overshadowed the rescue packages put together a decade earlier. In 1997, for example, South Korea received a loan of under $10 billion to tide it through the crisis that was then sweeping Asia. By contrast, Ukraine, a country with an economy a fraction of the size of South Korea's, received a whopping $16.4 billion in 2008.

The IMF was not the only lender of last resort. In addition to its myriad domestic interventions, the Federal Reserve played this important international role, by providing "swap lines." Under these agreements, the Fed "swaps" dollars for some other central bank's currency. It thereby enables the central banks to lend out dollars to anyone needing them in their home countries. For example, in April 2009, Mexico activated a $30 billion swap line with the Fed. This infusion of money injected liquidity into the market for dollars and helped anyone who owed dollars to pay off or roll over his debt.

These actions alone were remarkable, but in one of the strange and unprecedented features of the recent crisis, even the most stable, advanced economies faced liquidity crises comparable to the ones suffered by emerging markets. Many financial inst.i.tutions in Europe had borrowed enormous quant.i.ties of dollars in short-term loans to underwrite various speculations. When the interbank market froze up at the peak of the crisis, they were unable to roll over their dollar-denominated debts. Everyone needed dollars, and as a consequence, the value of the dollar went through the roof. This fact was terribly ironic: the country that was the ground zero of the financial crisis-the United States-saw its currency appreciate sharply in 2008.

Bernanke's solution was yet another bit of lender-of-last-resort legerdemain. The Federal Reserve can't lend directly to financial inst.i.tutions outside the United States, but it can lend dollars to foreign central banks, who can in turn lend them to the financial inst.i.tutions that need them so desperately. In return, the Fed gets an equivalent sum of whatever currency is the stock in trade of the central bank receiving the dollars. In this way, vast quant.i.ties of dollars traveled from the Federal Reserve to the European Central Bank, the Swiss National Bank, and the Bank of England, as well as the central banks of Sweden, Denmark, and Norway. In return, the Fed took custody of an equivalent amount of euros, pounds, francs, and other currencies. By late 2008 these swap lines totaled half a trillion dollars, and they started to decline only in the spring of 2009.

The crisis subsided because of these and many other extraordinary efforts undertaken to bring liquidity and stability back to the markets. But as policy makers found out, arresting the more immediate and dramatic crisis in short-term lending was one thing; getting banks to stop the larger drift toward deflation and depression was quite another.

Nuclear Options.

One of the more remarkable weapons that the Fed and other central banks brought to bear on the crisis was "quant.i.tative easing," though Ben Bernanke advocates calling it "credit easing"; economist Paul Krugman argues that it should be called "qualitative easing." Whatever its name, a modest version of this particular strategy had been tested in j.a.pan in the 1990s. The basic idea is to have the central bank intervene in markets for long-term debt in the same way that it does in markets for short-term debt.

Why go down the path of credit easing? The measures adopted so far hadn't worked their magic. Thanks to cuts in the overnight Federal funds rate, banks had access to plenty of cash; and thanks to the host of new liquidity facilities, financial inst.i.tutions of all stripes had access to cash as well, eventually driving down the cost of short-term borrowing, as measured by the LIBOR rate. Yet for all that largesse, banks continued to refuse to make longer-term loans to the many firms and businesses that needed credit to stay alive. Banks were getting no-interest loans from the Fed, but market rates for everyone else remained high. Financial inst.i.tutions continued to h.o.a.rd cash in antic.i.p.ation of future losses, or they sank it into the safest investments around: government debt, or "agency debt," the obligations of Fannie Mae and Freddie Mac.

Banks' propensity to park money in government or agency debt-particularly long-term debt-was understandable. By borrowing money from the Fed at policy rates approaching zero, then plowing it into a ten-year or thirty-year Treasury bond paying 3 to 4 percent, they could make a reliable profit and steer clear of all the risky borrowers who were clamoring for loans. While this strategy did nothing to ease the credit crunch, it made eminent sense from the standpoint of self-preservation.

Using quant.i.tative easing, the Federal Reserve would attack this problem on multiple fronts. It would wade into the financial system and start buying up long-term government debt: ten-year and thirty-year Treasury bonds. That would immediately inject ma.s.sive amounts of liquidity into the market because the Fed would pay for those bonds by creating money out of thin air. As it purchased hundreds of billions of dollars' worth of bonds, cash would flow to the banks that sold them. Now the banks would have even more cash, and presumably, they would be tempted to lend it.

The Fed's actions were designed to have the additional positive consequence of reducing the attractiveness of those bonds as a future investment. Why? Because bond prices and bond yields move in opposite directions. If the price goes up, the yield goes down. When the government created a demand for the bonds by buying them up, their price went up, and their yield went down. That meant they became less attractive as a place for banks to park money. In theory, banks would therefore look for other places to sink their money and therefore would consider making loans to those starving for credit.

This policy, announced in March 2009, went hand in hand with ma.s.sive purchases of other a.s.sets. On the same day the Fed announced that it would purchase upwards of $300 billion in long-term Treasury bonds, it also announced that it would buy a trillion dollars' worth of mortgage-backed securities and $55 billion worth of agency debt. As was the case with the proposed purchase of government bonds, the Federal Reserve had already made forays into these markets the previous fall. Still, the scale and scope of these interventions-particularly in the MBS market-was astonishing. So too was the announcement that the Fed would commit a trillion dollars to the Term a.s.set-Backed Securities Loan Facility (TALF), to support with Fed loans the private securitization of credit card debt and auto loans.

By broadening the range of a.s.sets it held, the Fed sought to prop up markets for various kinds of long-tem debt. Its intervention via the TALF program was a relatively modest attempt to revive the market for securitization. But by wading into the housing market, the Fed had bigger ambitions. Its purchases of mortgage-backed securities effectively gave Fannie Mae and Freddie Mac breathing room to guarantee more mortgages or bundles of mortgages. That strategy went hand in hand with the Fed's campaign to drive down the yield on ten- and thirty-year government bonds. Because long-term interest rates tend to move in tandem with one another, this intervention would have the effect of lowering mortgage rates, thereby jump-starting the mortgage market. It would also help drive down the costs of borrowing for corporations.

The Federal Reserve was not alone in its use of quant.i.tative easing. In Britain, the Bank of England was caught in a liquidity trap as well. It had cut its benchmark rates close to zero, the lowest since it was founded in 1694, and it had created liquidity facilities similar to those devised in the United States. But these moves failed to halt the prospect of debt deflation, and so in March 2009, in a bit of quant.i.tative easing of its own, the Bank of England pledged to buy some 150 billion worth of government debt and corporate bonds. The European Central Bank followed suit two months later, pledging 60 billion to purchase "covered bonds," a form of mortgage debt.

All these interventions const.i.tuted a dramatic shift in the role of central banks. In previous crises, central banks restricted their efforts to acting as lenders of last resort. This time, however, in a series of incremental steps, central banks around the world adopted a new role: as investor of last resort. They began by creating liquidity facilities that enabled financial inst.i.tutions to swap toxic a.s.sets for supersafe government debt; they thereby effectively created an artificial market for unwanted a.s.sets. At the same time, when they made outright loans, they accepted a remarkable range of collateral, everything from corporate bonds to commercial real estate loans to commercial paper. This too helped prop up the value of a range of a.s.sets.

The policy of quant.i.tative easing, adopted by the Fed and other central banks, marked the culmination of this process: outright purchases of long-term debt in the open market. As a consequence, the balance sheets of central banks underwent a profound transformation. In 2007, for example, the Federal Reserve held approximately $900 billion worth of a.s.sets, consisting almost entirely of its stock in trade: the debt of the U.S. government. By the summer of 2009, the Fed's balance sheet had ballooned to approximately $2.3 trillion or $2.4 trillion, the overwhelming majority of which consisted of a.s.sets acc.u.mulated during the crisis. Some of these a.s.sets, such as the debt of Fannie Mae and Freddie Mac, were somewhat safe. Others were less safe, particularly those derived from home mortgages, credit card debt, and auto loans.

Most dodgy of all were the collateralized debt obligations and other potentially toxic a.s.sets acquired during the bailout of Bear Stearns and AIG. These a.s.sets, Fed staffers reported in February 2009, represented "some of the most esoteric components of the Federal Reserve's balance sheet." It was a serious understatement. Unlike most of the a.s.sets it holds at this writing, the Fed "owns" these a.s.sets via its control of three limited-liability corporations known as Maiden Lane I, II, and III. Each is privately administered by BlackRock Financial Management. This highly unusual arrangement has attracted considerable criticism-and skepticism. It is also without precedent in the history of the Federal Reserve.

Taken together, all these actions const.i.tuted a ma.s.sive and unprecedented intervention in the financial system, using conventional and unconventional monetary policy. Over the course of the crisis, Bernanke (and to a lesser extent, other central bankers) sought to counter the effects of the financial crisis with three kinds of tools. Most traditional was the provision of liquidity (lender-of-last-resort support) to a host of financial inst.i.tutions, including banks, broker dealers, and even foreign central banks. Less conventional was the creation of the special facilities that purchased (or financed the purchase of) specific kinds of short-term debt-commercial paper, for example. Then the Fed began to play the role of investor of last resort, which culminated in the most radical programs of all: its commitment to intervene in markets for long-term debt (various a.s.set-backed securities and long-term government debt).

While these measures are somewhat staggering to contemplate, they were not as crazy as some of the other options that had been contemplated during the crisis. For example, the Federal Reserve could have intervened directly in the stock markets, buying up unwanted equities. This tactic had been deployed during the Asian financial crisis of 1998, when monetary authorities in Hong Kong purchased 5 percent of the shares being traded on the local stock exchange. The measure was widely criticized at the time, but it managed to forestall a foreign exchange crisis by frustrating the attempts of some large hedge funds to pull off a "double play," shorting both the currency and the stock market. Indeed, the government went on to make a tidy profit from its investment. Likewise, the Bank of j.a.pan adopted a similar policy in 2002, though its intervention paled in comparison to Hong Kong's and aimed merely to prop up the prices of certain bank stocks and, by extension, the banks themselves. In 2009, it repeated these measures for much the same reason.

The Fed did not go down this road, and with good reason: it would have raised the understandable concern that the government was manipulating markets in the world's biggest economy, thereby endangering its already fragile credibility. That same concern explains why the Fed set certain limits on its other interventions. It accepted only investment-grade a.s.sets as collateral for making loans and refused to purchase low-grade commercial paper when it waded into that particular market. There were limits to how far the Fed would go to stop the crisis.

Nor did the Fed ever deploy several other extremely controversial weapons. It might have used quant.i.tative easing on a far more ma.s.sive scale, manipulating the foreign exchange markets to weaken the value of the dollar, or even employed some version of a strategy half-seriously proposed by Milton Friedman: having the government print money and scatter it on the population from helicopters. Friedman never intended that policy makers actually distribute money like manna from heaven, but there were functional equivalents of doing this: giving people tax cuts financed entirely by printing money, for example. Bernanke embraced this idea back in 2002 but never pursued it during the crisis.

Nevertheless, Bernanke and other central bankers did employ some highly unconventional measures in their efforts to put a stop to the crisis. Unfortunately, a radical remedy administered in a crisis is bound to have unintended consequences. For starters, the Fed has sent a clear message to the financial markets that it will do almost anything and everything to prevent a financial crisis from spinning out of control. That's wonderfully rea.s.suring, but it creates moral hazard on a grand scale. The next time a crisis. .h.i.ts, banks and other financial firms could be forgiven for believing that the Fed will rescue them once again. In fact, now that there's a precedent for setting up special liquidity facilities and extending lender-of-last-resort support to broad swaths of the global financial system, firms may reasonably expect them to be resurrected at the slightest sign of trouble down the line.