4

WHEN THE MUSIC STOPPED

As long as the music is playing, you’ve got to get up and dance. We’re still dancing.

—CHUCK PRINCE, CEO OF CITIGROUP, JULY 2007

By 2006 the United States had built an intricate financial house of cards—a concoction of great complexity, but also of great fragility. Like most houses of cards, this one was constructed slowly and painstakingly. The sheer ingenuity was impressive. But when it fell, it tumbled suddenly and chaotically. All that was necessary to trigger the collapse was the removal of one of its main supporting props. The jig was up when house prices ended their long ascent; after that, the rest of the crumbling followed logically. Unfortunately, not many people had penetrated the tortured logic beforehand; so few were prepared for the devastation that ensued.

The end of the house-price bubble itself could hardly have come as a surprise. By 2006 the “is there a bubble” debate was just about over, and seemingly everyone was wondering how much longer the levitation act could last. The disagreement—and it was a serious one—was over how far house prices would fall. Optimists thought prices would just level off, ending their unsustainable climb, or perhaps decline only a little. Pessimists were talking about price declines of 20 percent, 30 percent, or even more. What about the market? Futures traded on the Chicago Mercantile Exchange on September 16, 2006 indicated that investors expected a 6.4 percent decline in the Case-Shiller ten-city composite index. That proved to be way too small. In the end, the more pessimistic you were, the more prescient you were.

THE CARDS TUMBLE

The bond bubble was far less visible to most people, vastly more complicated, and appreciated by few. It also burst with devastating effect. But the bursting came in stages.

Once house prices stopped rising, subprime mortgages that had been designed to default started doing precisely that. At first, many of us wrongly believed that subprime constituted too small a corner of the financial market to do much damage to the overall economy. We soon learned better. To pick two nonrandom examples, Treasury Secretary Hank Paulson said in an April 2007 speech that the subprime mortgage problems were “largely contained.” A month later, Federal Reserve Chairman Ben Bernanke told a Fed conference that “we do not expect significant spillovers from the subprime market to the rest of the economy or to the financial system.”

Unfortunately, the huge amounts of leverage multiplied the damages manyfold, and the untoward degree of complexity helped spread the ruin far and wide. Financial industry executives—allegedly the smartest guys in the room—had every incentive to keep the party going for as long as they could, and they certainly tried. The regulators, still asleep at their various posts, allowed them to go on for far too long. The day of reckoning was delayed but not avoided. Why did the house of cards tumble so hard and so fast? Most of the reasons were implicit in the last two chapters.

Start by recalling our hypothetical investment bank, FIB, which convinced our hypothetical commercial bank, RBC, and probably itself as well, that mortgage-related securities reduced risk by pooling mortgages from different geographical areas and selling the resulting securities all over the globe. Unfortunately, when the house-price bubble burst, neither type of diversification worked as advertised. Why not?

First, when the national housing bubble burst, home prices actually did fall almost everywhere—an “impossible” event that had not occurred since the Great Depression. In fairness, few observers anticipated this virtually unprecedented collapse. (True confession: I was not one of them.) For decades, Americans had witnessed periodic housing bubbles, which blew up and popped in particular parts of the country. But when home prices fell in, say, Boston, they kept rising in, say, Los Angeles—and vice versa. The period after 2006 was different. House prices fell all over the map, undermining the trumpeted gains from geographical diversification.

That was a forgivable error. The proverbial hundred-year flood actually happened. But when the housing market began to crater, we also learned that many of the MBS were not nearly as well diversified geographically as had been claimed. In fact, it turned out that a distressingly large share of the bad mortgages came from a single state: California. Many of the rest came from Florida, Arizona, and Nevada—collectively known as the “sand states.” For these and other reasons, the MBS turned out to be much riskier than advertised.

Second, the securities were not as widely distributed as had been thought. Yes, there were holders all over the world—from hamlets in Norway to Italian pension funds to billionaires in Singapore. But when the crash came, we learned that many leading financial institutions had apparently found mortgage-related assets so attractive that they still owned large concentrations of them when the bottom fell out. One reason was that there was so much profit in selling the other tranches of MBS, CDOs, and the like that investment banks were willing to hold the lowest-rated (“toxic waste”) tranches themselves. The failures and near failures of such venerable firms as Bear Stearns, Lehman Brothers, Merrill Lynch, Wachovia, Citigroup, Bank of America, and others were all traceable, directly or indirectly, to excessive concentrations of mortgage-related risks.

The system began to crack in July 2007, the very month in which Chuck Prince made the now-famous statement that opened this chapter, when Bear Stearns told investors in one of its mortgage-related funds that there was “effectively no value left.” The music was stopping. A variety of financial markets started twitching nervously, which should have been taken as an omen. But wishful thinking dies hard.

The real wake-up call didn’t come until August 9, 2007, when BNP Paribas, a huge French bank, halted withdrawals on three of its subprime mortgage funds—citing as its reason that “the complete evaporation of liquidity in certain market segments of the US securitization market has made it impossible to value certain assets fairly.” Loose translation: Dear Customer, you can’t get access to the money you thought was yours, and we have no idea how much money that is. To people acquainted with American history, Paribas’ announcement brought to mind the periodic “suspensions of specie payments” in the nineteenth century—times when some prominent bank precipitated bank runs by refusing to exchange its notes for gold or silver. The big French bank had just refused to exchange its fund shares for cash. Whether you were French or American, the signal was clear: It was time to panic. And markets dutifully did so, all over the world.

At some point, and in this case it didn’t take long, the interplay of falling asset values with high leverage starts calling into question the solvency of heavily exposed financial firms like Bear and Paribas. Thus, market-price risk, which is already acute and getting worse, conjures up visions of counterparty risk: worries that firms that owe you money might not be able to pay up.

Once such seeds of doubt are sown, the scramble for liquidity begins in earnest, because, like it or not, markets are fundamentally built on trust—in particular, on trust that the other guy will pay what he owes you in full and on time. In worst cases, markets seize up. In less severe cases, enormous “flights to quality” are triggered, typically to U.S. Treasury bills. In any case, the bond bubble, which was predicated on blissfully ignoring risk, ended with a bang on August 9, 2007.

That faith in counterparties started to evaporate on that day is evidenced by the sharp rise in interbank lending rates. The key rate that everyone watches is LIBOR—the London Interbank Offer Rate—which indicates what one big bank charges another for short-term lending. This is a market restricted to the big boys: HSBC lending to Citibank, and so on. Was there actually a risk that such august financial giants might fail to repay overnight loans? The risk premium reflected in LIBOR said yes. After being stable for months, the LIBOR spread over Treasuries jumped by 30 basis points in just three business days. In the ultrasafe world of LIBOR, 30 basis points is a big deal. Financial lore has it that markets are alternately dominated by greed and fear. Fear was taking over.

AT THE FED’S ANNUAL WATERING HOLE

In late August of each year, most members of the Federal Open Market Committee (FOMC) doff their gray suits, don their cowboy boots (if they own any), and head off to Jackson Hole, Wyoming. There they meet with a select group of academic economists and a highly select group of bankers and Wall Streeters, for an invitation to the Jackson Hole conference is the hottest ticket in Lower Manhattan. While plenty of time is set aside for hiking and whitewater rafting, the dominant activity at Jackson Hole is shoptalk. It was in great abundance in August 2007.

The annual conclave, which is the Fed’s premier event, is hosted by the Federal Reserve Bank of Kansas City, and its conference planners hit the jackpot in selecting the topic for the 2007 edition: “Housing, Housing Finance, and Monetary Policy.” When the group convened on the evening of August 30, housing was going to the dogs, housing finance was cratering, and a monetary policy response to all this was growing increasingly urgent. Few people wanted to talk about the weather—which, as usual, was gorgeous.

Too bad the conference didn’t take place four weeks earlier. At its August 7, 2007, meeting, the FOMC had concluded that “although the downside risks to growth have increased somewhat, the Committee’s predominant policy concern remains the risk that inflation will fail to moderate as expected.” How’s that again?, many of us thought when we read the statement. The predominant concern is inflation? Many Fed watchers blinked in disbelief. What were those guys thinking?

Two days later in Paris, the financial world started coming apart at the seams. The next day, the FOMC held a hurriedly arranged telephonic meeting. This time their statement assured the financial world that the Fed was “providing liquidity to facilitate the orderly functioning of financial markets.” (Translation: We are pumping out cash like mad.) But the federal funds rate was kept right where it had been since June 2006, at 5.25 percent. Was the Fed still seeing inflation as the “predominant policy concern”? Okay, give them a break. Only three days had passed since their August 7 meeting. The Fed would fix things soon. Right?

Wrong. The committee met telephonically again six days later, noting correctly that “the downside risks to growth have increased.” That was a healthy step; they demoted inflation from its singular status as the predominant risk. But the FOMC still refused to cut the funds rate. (It did reduce the less-important discount rate.) Looking back, it’s hard to see how this could have been a close call on August 16, and many Fed critics said so at the time. But thirteen days later, as FOMC members from Washington and around the country boarded planes to head to the beautiful Grand Tetons, the funds rate was still stuck at 5.25 percent.

The papers economists presented at Jackson Hole that year ranged from dour to alarmist, thereby matching the corridor conversations. Bob Shiller warned yet again that housing would probably fall hard. He was right. Ed Leamer of UCLA reminded everyone that housing collapses are, historically, central ingredients in recessions. He was also right. A lunchtime address written by the ailing Federal Reserve governor Ned Gramlich warned that “the predictable result [in the subprime lending market] was carnage.” Right again. Even sitting Federal Reserve governor Frederic “Rick” Mishkin, though trying to be as circumspect as he could because the FOMC had just decided to stand pat, strongly hinted that interest rates needed to fall. Summing up the proceedings, a pessimistic Martin Feldstein of Harvard observed that “if the . . . threat from the housing sector materializes with full force, the economy could suffer a very serious downturn.” It did, of course.

But that was just the visible part of the conference. We learned later that Bernanke took the opportunity to cloister away several FOMC members in a small upstairs conference room to figure out what to do next—as their initial efforts were clearly inadequate. There must have been many other interesting sidebar conversations. But still, rates weren’t touched until the FOMC’s next regularly scheduled meeting, which was on September 18—a full forty days after Paribas Day. Yes, a lot of rain can fall in forty days and nights—and it did. The Fed cut the funds rate by 50 basis points on September 18, observing that “the tightening of credit conditions has the potential to intensify the housing correction and to restrain economic growth more generally.”

That last thought was important. Before the cataclysmic failure of Lehman Brothers a year later, there were two competing views of what the crisis was all about. In the narrower, technical view, the financial world was experiencing a liquidity crisis—an acute one, to be sure, but still a liquidity crisis. In plain English, that meant that frightened investors and institutions wanted to get their hands on more cash than was available—partly because of the heightened counterparty risk just mentioned, partly because assets formerly deemed safe now looked risky, and partly because banks and investment funds feared that their customers might show up at the electronic door one day, seeking to make hefty withdrawals. The Paribas approach—just say no—was not an appealing way to cope with such a problem.

The dash for cash was on. The one institution in any country that can provide more cash in a hurry—in principle, in unlimited amounts—is the central bank. Both the Federal Reserve and the European Central Bank (ECB) did so massively, starting on Paribas Day, August 9, 2007. In so doing, they were performing a function that central banks have performed for centuries: serving as the “lender of last resort” in order to get their financial systems through liquidity crises. The volume of new dollars and new euros spewing forth from the world’s two largest central banks was unprecedented. But the actions themselves were time-tested and routine, part of every central banker’s DNA.

Back in 1873, Walter Bagehot, the sage of central banking, had instructed central banks on what to do in a liquidity crisis. His triad was lend freely, against good collateral, but at a penalty rate. Why? Because the acute shortage of liquidity in a panic can push even solvent institutions over the edge. Customers come in demanding their money. If the banks don’t have enough cash on hand, word gets around, and bank runs start sprouting up everywhere. The disease is highly contagious.

By serving as the lender of last resort, the central bank is supposed to stop all that from happening. And every central banker in the world knew Bagehot’s catechism. So that’s basically what most of them did in August 2007. In fact, one can argue that the ECB stuck with the Bagehot script until late 2011. The ECB refused to cut its interest rates until October 2008 (yes, that’s 2008, not 2007), and even then it gave ground grudgingly. The 4 percent European overnight rate that prevailed in August 2007 did not fall to 3.25 percent until November 2008 and did not get as low as 2 percent until January 2009. By contrast, the Fed had virtually hit zero by December 2008.

Was this mess nothing more than a big liquidity event, as the ECB’s actions suggested? Perhaps not. An alternative, and darker, view of the crisis conceptualized what was happening as a serious impairment of the economy’s normal credit-granting mechanisms. On this broader view, the scarcity of liquidity was just the tip of the iceberg. The real problems lurked down the road—in gigantic losses of wealth; in massive deleveraging and possible insolvencies of major institutions; and, as just mentioned, in severe damage to the banking system, the shadow banking system, and other credit-granting mechanisms. If all that happened—and in August 2007, it hadn’t happened yet—the whole economy would be in big trouble. Economies that are starved of credit fall into recessions, or worse. Businesses decline and fail. Workers lose their jobs.

THE FED SPRINGS INTO ACTION

Well, maybe not exactly “springs.” Bernanke, who was a noted scholar of the Great Depression,* was slowly bringing his rather hawkish committee around to the view that this was something big—not just a major liquidity event, but potentially the cause for a big recession. But old habits die hard, and while the Fed was way ahead of the ECB, it was not quite there yet. At its September 18 meeting, the FOMC qualified its view that “the tightening of credit conditions has the potential to . . . restrain economic growth” by adding that “some inflation risks remain.” It was a finely balanced assessment of risks—far too balanced, given the emerging realities. Just five days earlier, the Bank of England had intervened massively to save Northern Rock, a huge savings institution, from the first bank run in Britain since 1866.* Things were coming unglued in England. Our problems here were strikingly similar. Could we be far behind?

While the Fed’s speed made the ECB look like the proverbial tortoise watching the hare, this particular hare wasn’t actually running that fast. After its 50-basis-point rate cut on September 18, 2007, the Fed waited another six weeks—until its next regularly scheduled meeting—to move again. By that time, many mortgage-lending companies had failed, and Citigroup and others had announced major write-downs on subprime mortgages. But the Fed chipped in with only another 25 basis points on October 31—a baby step that it repeated at its next regular meeting on December 11. A number of FOMC members were less than convinced of the need for easier money.

After that, however, the Fed seemed to step it up a notch. The next day it announced two new liquidity-providing facilities. The first was a series of currency swap lines with foreign central banks, which were finding themselves seriously short of dollars. In a currency swap, the Fed, say, lends dollars to the ECB in return for euros. When the dollar liquidity crisis in Europe passes, the ECB pays back the dollars and gets back the euros. The initial announcement was for just $24 billion, which was considered sizable at the time. But the swap lines eventually topped out at a whopping $583 billion in December 2008.

The second facility was the Term Auction Facility (TAF), designed to do Bagehot-type lending to banks, though for periods longer than normal—up to four weeks. The Fed’s earlier attempts to lend to banks had been stymied by bankers’ fears of being stigmatized by asking the central bank for a loan. Didn’t that mean you were on the ropes? The TAF sought to overcome the stigma problem in two ways: It was set up as an auction in which any bank could show up to bid; the bank didn’t need to be in bad shape. And since the bank wouldn’t receive the cash for a few days, a TAF loan would not save a bank that was on the brink of disaster. TAF was important both for what it accomplished—at its peak in March 2009, it was lending $493 billion—and for its influence on subsequent policy. As we shall see, the Fed’s memory of the stigma problem that led to TAF lingered on, affecting future policy decisions. TAF itself was shut down in March 2010. It was no longer needed.

But the Fed was just warming up. Over the Christmas–New Year holidays, Bernanke must have got to thinking—or to having nightmares—about the 1930s. On January 9, 2008, he convened an FOMC conference call—ostensibly to review recent developments but perhaps actually to shake the committee out of its lethargy. The minutes of that meeting noted that “the downside risks to growth had increased significantly since the time of the December FOMC meeting,” but the committee was not yet ready to cut interest rates. Undaunted, Bernanke got them all on the phone again on January 21. The minutes of that call observed that “incoming information since the conference call on January 9 had reinforced the view that the outlook for economic activity was weakening.” Only twelve days had elapsed between the two calls. How much could have changed?

What had changed was that the FOMC was now ready to act—dramatically. With just one dissenter, the members agreed to announce an almost-unprecedented 75-basis-point cut in the federal funds rate early the next morning.* In the entire eighteen and a half years of the Greenspan Fed, the FOMC had moved the funds rate by 75 basis points only once—and that was an increase. Furthermore, federal funds rate announcements always come at exactly 2:15 p.m. This one came at 8:30 a.m. The Fed clearly wanted to be heard on January 22—and it was. Eight days later, at its regularly scheduled meeting, and again with one dissenter,* the FOMC dropped the funds rate another 50 basis points, down to 3 percent, leaving federal funds trading 125 basis points lower than they were only nine days previously. The Fed was on DEFCON 1.

The FOMC majority now clearly saw the task ahead of them as a two-front war, and it was gearing up for battle on both fronts. It needed to provide massive amounts of liquidity, for well-known reasons that Bagehot had articulated 135 years earlier. But it also needed to cut interest rates to fight an imminent recession, as Keynes had prescribed 72 years earlier. Chairman Bernanke did not want to preside over another episode like the 1930s. In Europe, however, overnight rates were going nowhere. The ECB was fighting the shortage of liquidity hard, maybe even harder than the Fed. But it was far from convinced that recession was in its future. At the ECB’s headquarters in Frankfurt, there was lots of Bagehot but not much Keynes.

MISSING PERSONS

Meanwhile, one might ask—as some people did—where was the United States Treasury? The approximate answer: working hard, but mostly behind the curtains. To provide some much-needed stimulus for the sluggish economy, President George W. Bush had gotten a temporary tax cut through Congress in February 2008—tax cuts were his remedy for everything. After that, however, he more or less checked out, delegating management of the economy to his secretary of the Treasury, Hank Paulson.

The Bush administration looked to be a spent force by then. But Paulson seemed to be the perfect man for the job. A complex, multifaceted financial crisis, centered on Wall Street, was brewing. President Bush’s two previous Treasury secretaries had been an aluminum executive and a railroad man, and the chairman of the Fed had come from the cloistered halls of academia. But Paulson had been the head of Goldman Sachs, a lion of Wall Street, a King of the Universe, a man who had spent his entire adult life in the rough-and-tumble of the financial world. (He collected birds of prey as a hobby, which, some people claimed, told you something.) Who could have been better prepared for the difficult job he now faced? Or so we thought.

Throughout this period, while the Fed was working feverishly to douse the flames that threatened to engulf Paulson’s old neighborhood (and, later, even his old firm), the secretary of the Treasury was pretty quiet in public. Other than shepherding the February 2008 tax cut through Congress, the Treasury secretary seemed content to work behind the scenes, with the Fed in the visible lead chair. One major reason, of course, is that he lacked something the Fed had: money.

Fortunately, the Bernanke-led Federal Reserve had plenty of that, and it kept up the fight. On March 11, 2008, the Fed announced yet another novel way in which to serve as the lender of last resort. (There would be more to come.) The nervous markets were craving Treasury securities, especially T-bills, as a kind of security blanket. And the Fed owned lots of them. The Term Securities Lending Facility (TSLF) was designed to lend Treasury securities to the so-called primary dealers—a bunch of Wall Street firms and their foreign counterparts that helped make a market in Treasuries. As security for the loans, the Fed would take what Bagehot would have called “good collateral.” Well, not all of it was supergood. Let’s just say, pretty decent collateral.

The idea was straightforward. The markets were suffering from a shortage of highly liquid Treasury securities, but they had a huge surplus of other stuff. If the Fed was willing to swap T-bills for some of that other stuff, it might alleviate the liquidity squeeze. After all, T-bills can be converted into cash in seconds. And the Fed could make the swaps without expanding its balance sheet or bank reserves, the latter being the usual fodder for money-supply expansion. Hence, the TSLF was born—or, rather, announced; its first actual loan was not made until March 27. That date was exquisitely bad timing for one of the primary dealers that was fighting for its life at the time: an investment bank named Bear Stearns.