14

NO EXIT? GETTING THE FED BACK TO NORMAL

So let it out and let it in, hey Jude, begin . . .

THE BEATLES

What goes up into the stratosphere, such as bank reserves, must presumably come back down to earth at some point. What drops to rock bottom, such as short-term interest rates, must presumably pop back up eventually. When a central bank opens up all the stops to stimulate its economy, as the Federal Reserve did after September 2008, it must at some point throttle back and return to normalcy.

But when and how? Questions like that constitute what has come to be called the Fed’s “exit strategy.” Executing the exit strategy well is important to us all. If the Federal Reserve rushes for the door too soon or too abruptly, we could fall back into recession. If it exits too late or too slowly, we could have an inflation problem on our hands.

EXIT FROM WHAT?

Can’t the Fed just retrace its steps, like a hiker who cuts bark off trees to mark her path and then follows it back to the trailhead? In large measure, the answer is yes. But the pace and order in which the Fed should “let it in” when the time for exit arrives might not match the pace and order in which the Fed “let it out” after September 2008. And there is much from which to exit, such as:

Extraordinary Lending Programs

As we have seen, the Fed cooked up a thick alphabet soup of lending facilities to cope with the ever-changing financial crisis. Remember TAF, TALF, AMLF, CPFF, PDCF, TSLF, and the others? (Maybe you don’t, which is just as well.) Each of these unusual lending programs was designed to meet a particular pressing need for funding or credit. And most of them naturally dwindled away as the need disappeared.

Figure 14.1 is a nice graphical depiction, provided by the Fed, of how the central bank “let it out and let it in” with those extraordinary lending facilities. The ups and downs are impressive. In total, Federal Reserve lending ballooned from roughly zero at the end of 2007 to a peak of about $1.5 trillion at the end of 2008 and then fell back to barely above zero by the end of 2010. It was quite a roller coaster! This part of the Fed’s exit has already been accomplished. As we will see shortly, it’s the only part that has been.

Supersized Balance Sheet

Every loan that underlies figure 14.1 was, of course, an asset on the Federal Reserve’s balance sheet. But as the Fed’s loan volume plummeted after the frightening winter of 2008–2009, the central bank’s total assets did not shrink much. How is that possible, when loans dropped from $1.5 trillion to barely above zero? Figure 14.2 shows the answer: The dwindling volume of loans (the dashed line) was replaced by an even larger volume of securities, mainly Treasuries and debt instruments of Fannie Mae and Freddie Mac that the Fed acquired in the two big asset-purchase programs nicknamed QE1 and QE2 (the dotted line). The Fed’s total asset holdings (the solid line) fluttered for about a year or so in the neighborhood of $2.25 trillion, and then soared again.

FIGURE 14.1 The Rise and Fall of Federal Reserve Lending

SOURCE: Federal Reserve System, MONTHLY REPORT ON CREDIT AND LIQUIDITY PROGRAMS AND THE BALANCE SHEET, December 2011

FIGURE 14.2 The Changing Face of Federal Reserve Assets

SOURCE: Federal Reserve System, MONTHLY REPORT ON CREDIT AND LIQUIDITY PROGRAMS AND THE BALANCE SHEET, December 2011

Figure 14.2 shows the basic story: loans went down, but securities went up. It also shows that the Fed has not even begun to exit from this aspect of its unconventional monetary policy: the huge balance sheet. On the contrary, yet another program of quantitative easing (QE3) was announced in September 2012. So the Fed is still entering rather than exiting.

Mountain of Excess Reserves

Balance sheets must balance. As the Fed’s total assets soared toward the sky, so did its total liabilities. Figure 14.3 shows that almost all the growth of the Federal Reserve’s liabilities during and after the crisis came from “Deposits of Depository Institutions” (the dotted line)—in plainer language, from bank reserves.

FIGURE 14.3 Selected Federal Reserve Liabilities

SOURCE: Federal Reserve System, MONTHLY REPORT ON CREDIT AND LIQUIDITY PROGRAMS AND THE BALANCE SHEET, December 2011

The Fed normally pays for any assets it acquires by “printing money”—that is, by creating new bank reserves, which it deposits into the accounts of the selling banks. A small portion of these new reserves represents the minimum reserves required by law. These required reserves grow in strict proportion to banks’ transactions deposits—which is to say, modestly.

But as figure 14.4 makes clear, the overwhelming majority of the explosion in bank reserves since September 2008 has been in excess (that is, not required) reserves—deposits that banks hold willingly at the Fed even though they are just sitting there doing nothing. Excess reserves don’t lead to a bigger money supply or to more bank credit; and they don’t earn the banks much interest. As you can see in the figure, banks’ holdings of excess reserves were essentially zero from January 2000 until Lehman Day, which is the normal state of affairs. In stark contrast, they topped $1.6 trillion in 2011. That’s not a small discrepancy! An important part of the Fed’s eventual exit strategy will be reducing these excess reserves toward zero again.* It looks like a big job, which they’ve yet to begin.

FIGURE 14.4 Total Reserves, Required Reserves, and Excess Reserves

(in billions of dollars)

SOURCE: Federal Reserve Board 2012

Interest Rates Near Zero

Have you checked the interest rates on your bank accounts lately? They are stunningly low. That’s because the Federal Reserve’s extraordinary monetary policies have been holding the federal funds rate down to a range between zero and 25 basis points since December 2008. To call such a low interest rate abnormal is an understatement. Figure 14.5 displays the history of the federal funds rate from 1990 to 2012. The post-Lehman experience stands out. Except for the period 2002–2004, the funds rate has always been at least 3 percent, and generally higher. Normalizing interest rates is probably the most obvious piece of the Fed’s exit strategy; it will certainly be the most obvious to the general public. It needs to be done deftly.

FIGURE 14.5 The Federal Funds Rate Since 1990

SOURCE: Federal Reserve Board 2012

Forward-Looking Prose

The final unconventional monetary policy from which the Fed will eventually exit is its open-mouth policy—in particular, how it communicates to the market its intention to maintain the superlow federal funds rate for some time to come. As just noted, the funds rate hit virtually zero in December 2008, from where it can fall no further. But how long will fed funds remain in that rock-bottom 0-to-25-basis-points range?

From March 2009 until August 2011, the FOMC told us that the funds rate would remain that low “for an extended period.” These words became almost holy writ in the financial world, even though no one knew exactly what they meant. Market participants were left guessing. Then, at its August 2011 meeting, the FOMC changed the “extended period” wording to “at least through mid-2013.” Now the markets had a date—and a pretty distant date at that. By this change in prose, the Fed was still entering not exiting. It clearly viewed its new wording as an ersatz easing of monetary policy: It will be longer than you thought, folks.

The phrase “at least through mid-2013” was heavily criticized, both outside and inside the FOMC, as misleading and even potentially counterproductive by attaching a date to the Fed’s eventual tightening when exit was actually a function of economic conditions. As Charles Plosser, president of the Federal Reserve Bank of Philadelphia, told the Wall Street Journal, “Policy needs to be contingent on the economy, not the calendar.” He was right. For example, what if the Fed expected to hold the funds rate near zero well into 2014 or beyond? Then might not saying “at least through mid-2013” tacitly make monetary policy a bit tighter than the Fed intended? The Fed’s answer came soon: Yes, it might.

At its January 2012 meeting, the FOMC overhauled the nature of its communications in several ways, yet stuck surprisingly close to the “through mid-2013” concept rather than making the commitment “contingent on the economy, not the calendar.” Their statement observed that “economic conditions . . . are likely to warrant exceptionally low levels for the federal funds rate [read: zero to 25 basis points] at least through late 2014.” The eighteen-month extension was a tacit easing of monetary policy, and long-term interest rates promptly dropped. Here, again, the Fed was still entering, not exiting.

At that meeting, the Fed also began publishing different FOMC members’ forecasts of how the federal funds rate should or would evolve over the coming three years, just as they forecast future GDP, unemployment, and inflation rates. The views on where the federal funds rate should be at the end of 2014 varied enormously—from the 0-to-25-basis-points range that had prevailed since December 2008 (six of the seventeen members) to 2.5 percent or higher (four members). Two members even said that the near-zero interest-rate policy should continue into 2016. They looked like a rather fractious bunch, which they were.

TIMING IS EVERYTHING

That covers the what of the Fed’s exit. How about the when? If the Fed keeps monetary policy too loose for too long, it will overstimulate the economy, leading to higher inflation. Given where we’ve been, a booming economy doesn’t sound so bad. Nonetheless, the fear of accidentally higher inflation has had the FOMC’s hawks clamoring for a faster exit ever since 2009.

The fretting generally starts with the veritable mountain of excess reserves shown in figure 14.4. As conditions normalize, the hawks reason, banks will not want to hold on to so many excess reserves. Remember, the historical norm is about zero. If the Fed leaves too many reserves hanging around for too long, banks will lend those funds out rather than hoard them, leading to large increases in the money supply and bank credit. And that, economists have taught us since time immemorial, would lead to inflation. In a commonly used metaphor, hawks see the huge accumulation of reserves as dangerous tinder that may one day catch fire. They’d like to clean up the tinder before someone drops a match in it.

But there is also an opposite danger. What if the Fed tightens up too soon, before the economy is strong enough to take it? That potential error worries the doves. A premature tightening of monetary policy could kick our still-weak economy down the stairs again, leading to rising unemployment—and, by the way, to falling inflation. It seems that the Fed’s exit strategy has to be not too fast and not too slow, but just right. The Fed is human; it won’t get the timing exactly right. Rather, it will err in one direction or the other. But magnitudes matter. And unless FOMC members are derelict in their duties, their error should be modest. Why? Because they should be able to see problems coming and make midcourse corrections.

If the Fed starts to exit too late, as the hawks fear, and the economy begins to heat up, the Fed can and will accelerate the exit process. Maybe not quickly enough, so we could wind up with, say, 3 percent or even 4 percent inflation instead of 2 percent—which is the Fed’s announced goal. But inflation much higher than that would be gross incompetence. On the other hand, if the Fed exits prematurely and the economy starts slipping again, which is the danger the doves want to avoid, the Fed can and will slow down its exit. In that case, we might wind up with 1 percent inflation or maybe even zero, rather than 2 percent. But serious deflation seems quite unlikely unless the Fed succumbs to a hawkish feeding frenzy. In short, the risk of exiting either too late or too early is limited by the well-documented inertia that characterizes inflation in America. Unlike, say, stock market prices, core inflation rarely either leaps upward or drops like a stone.* Rather, it rises or falls gradually, thereby giving policy makers plenty of time to react.

Okay. But which of the two bad outcomes worries the financial markets more? Neither. By comparing the nominal interest rate on regular 10-year Treasury notes with the real interest rate on 10-year TIPS (Treasury Inflation-Protected Securities), one can extract a market-based estimate of expected inflation over a ten-year horizon, as explained in the accompanying box.

Look first at figure 14.6, which shows both the nominal interest rate (upper line) and the real interest rate (lower line) on 10-year U.S. government bonds for the years 2007–2011. The vertical distance between the two is a market measure of the expected inflation rate over the decade to come. To improve your vision, figure 14.7 subtracts the real rate from the nominal rate, thereby showing explicitly the vertical distance between the two lines in figure 14.6—that is, the inflation rate that the market expected to prevail.

FIGURE 14.6 Nominal and Real Interest Rate, 2007–2011

SOURCE: Federal Reserve Board

FIGURE 14.7 A Market-Based Measure of Expected Inflation, 2007–2011

There is only one really notable movement in figure 14.7, and it happened during the worst months of the financial crisis: the sharp drop in inflationary expectations in late 2008 and its quick rebound in early 2009. Other than that, expected inflation just bounced around in the 1.5 percent to 2.5 percent range. Clearly, the market expects inflation to remain within shouting distance of the Fed’s 2 percent target.

WHERE’S THE DOOR?

Some of the more hawkish FOMC members were looking for the exit door already by the spring of 2009. In retrospect, and maybe even in prospect, that was astonishingly premature. The worst of the financial crisis was just receding, the economy was still contracting, and quantitative easing—a central part of the Fed’s entrance strategy—had begun only in November 2008. Nonetheless, by July 2009, Chairman Ben Bernanke was already sketching the Fed’s exit strategy to Congress, stating that “it is important to assure the public and the markets that the extraordinary policy measures we have taken in response to the financial crisis and the recession can be withdrawn in a smooth and timely manner as needed, thereby avoiding the risk that policy stimulus could lead to a future rise in inflation.” He explicitly mentioned, as key components of that exit strategy:

• the automatic unwinding of the Fed’s extraordinary lending facilities

• extinguishing bank reserves by selling securities out of its portfolio

• raising the interest rate the Fed pays on excess reserves, which he highlighted as “perhaps the most important such tool.”

The first of these three was mentioned already. It would happen automatically, and it has. The second constitutes the sort of thing the Fed has been doing on a discretionary basis for generations. The third was novel.

During the crisis, Congress gave the Fed authority to pay interest on bank reserves—and to set the rate thereon. Bernanke argued that once the exit started, the Fed could induce banks to shed their idle reserves more slowly by offering them a higher interest rate, thereby wrapping a kind of fire-resistant blanket around some of the inflationary tinder. Elaborating a few months later, Bernanke observed that while the Fed has “two broad means of tightening monetary policy at the appropriate time—paying interest on reserve balances and taking various actions that reduce the stock of reserves,” it would “likely would use both in combination.”

The Fed chairman did not, of course, indicate anything about exit dates, nor even about what facts on the ground might trigger exit. But he certainly suggested that he had a plan in mind and was waiting for the right moment to put it into effect. An interesting historical question, which Bernanke may never answer, is whether he actually felt the need to have an exit plan ready as early as mid-2009 or was just going through the motions to placate the Fed’s hawks—who were itching to exit.

EXIT BECOMES REENTRY

Both Bernanke and the hawks are still waiting. Most of the Fed’s attention since July 2009 has been focused instead on whether and how to “let it out” further. Why? Because the aftershocks weren’t over, and the Fed worried about a renewed earthquake. The first temblor came from Greece—or did it?

The U.S. economy was beginning to show clear signs of new life by late 2009 and early 2010. Real GDP grew at a 4 percent annual rate in the final quarter of 2009. Payroll employment gains rose sharply to an average of 315,000 per month over the March–May 2010 period, even reaching an eye-popping 516,000 in May. Although those numbers were bloated by the government’s hiring of over 500,000 temporary workers to conduct the 2010 Census (which peaked in May), the economy appeared to be reaching “escape velocity.”

But then GDP growth dropped off to a modest 2.4 percent average pace in the four quarters of 2010, and to virtually zero in the first quarter of 2011. Following suit, job losses reemerged over the four months after May 2010’s Census-induced hiring surge. Something went wrong badly that spring, and as figure 14.8 shows, the change was startlingly abrupt. What happened? The timing certainly points toward Greece. But was it just a coincidence?

FIGURE 14.8 The Bottom Drops Out—Again

That Greece had a serious deficit and debt problem was certainly no secret in 2009 and into 2010. Yet markets seemed not to be terribly concerned until late March 2010. Then, as is so often the case, they became obsessed by it, reacting violently, as if someone had shouted “fire” in a crowded theater. A series of promises of fiscal austerity by the Greek government (quickly broken) and of international assistance from the EU and the IMF followed. But the panic was on. The yield on 10-year Greek government bonds skyrocketed from barely over 6 percent in late March to over 12 percent by early May—doubling in about six weeks. It’s hard to hold back a tidal wave.

Twelve percent interest rates when inflation is running about 2 percent are crippling, and so raised serious questions about Greece’s financial viability. Could an advanced European country, the cradle of Western civilization no less, actually default on its debt? Unthinkable. Or was it? And since Greece shared the euro with sixteen other countries, markets soon cast a jaundiced eye on other heavily indebted eurozone nations, principally Portugal and Ireland.*

That said, it is hard to understand how the Greek sovereign debt crisis could have brought hiring in the United States to the screeching halt depicted in figure 14.8. After all, fears of recession in Greece, with a GDP roughly the size of Minnestota’s, certainly could not precipitate a U.S. recession. While the U.S. government had a huge fiscal deficit, similar to Greece’s, the financial ructions emanating from Athens sent nervous money flocking to the United States, not away. Interest rates on U.S. Treasury securities therefore dropped, and the dollar rose. Had worldwide investors worried that Greece’s debt problems foreshadowed our own, U.S. Treasury rates would have risen instead, and the dollar would have fallen. While financial contagion clearly afflicted Portugal and Ireland, it did not hop the Atlantic. Instead, the U.S. emerged as a major safe haven for frightened capital.

Still, the coincidence in time is striking. Our economy slipped a gear almost immediately after the financial panic hit Greece. Of course, other things were happening at the same time. The Fed’s QE1 asset-buying program was drawing to a close, and the Obama administration’s health care reform bill became law on March 30. The baseball season also started. But it is hard to blame any of these events, or a variety of others, for the abrupt slowdown in the U.S. economy. Perhaps the ructions in Greece (and Portugal and Ireland) just awakened latent fears that we were all headed back into the soup, spooking potential hirers. In any case, what happened happened. And as one corollary, the Federal Reserve stopped talking about exit. Instead, the main question at FOMC meetings became, What else can we do to spur growth?

Well, not quite. That was certainly the view of Chairman Bernanke and the FOMC majority. But the committee’s hawks, as always, were unconvinced. The one big hawk who had a vote in 2010 was Thomas Hoenig, then the president of the Federal Reserve Bank of Kansas City (and now an FDIC director). He dissented at every single meeting in 2010—thereby establishing a new FOMC record—fretting out loud about both inflationary tinder and the wisdom of the Fed’s commitment to keep rates near zero for “an extended period.” More or less the same hawk-dove debate was proceeding outside the Fed, too, with hawkish critics of the Fed arguing for exit and dovish critics goading the Fed to further expansionary action. But as the economy slowed after Greece, the FOMC hawks gradually lost the internal debate to Bernanke and the doves. By the time of its November 2010 meeting, the FOMC was announcing what came to be called QE2—a program of large-scale purchases of longer-dated Treasury securities, designed to push down long-term interest rates.

QE2 kicked off a firestorm of protest, most of it from the political Right. (Ironic reminder: Ben Bernanke is a Republican who described his own political views as libertarian!) But some came from within the FOMC itself. Perhaps the most stunning criticism was that of Fed governor Kevin Warsh, the young Bush appointee who had been one of Bernanke’s trusted confidants in the dark days of 2008. On November 3, Warsh voted in favor of QE2 at the FOMC meeting. Five days later, he was on the op-ed page of the Wall Street Journal objecting to it. My eyes popped when I opened my Wall Street Journal that morning. I guess a lot had changed in five days! Warsh warned that the “Fed’s increased presence in the market for long-term Treasury securities poses nontrivial risks that bear watching. . . . Responsible monetary policy . . . requires attention not only to near-term macroeconomic conditions, but also to corollary risks with long-term effects.” Bernanke must have had an Et tu, Brute moment.

But despite incessant clamoring for its early termination, QE2 ran through its scheduled completion in June 2011, during which time the Fed acquired an additional $600 billion in Treasury securities, mostly with 3- to 10-year terms. Its balance sheet was getting bigger, not smaller. QE2 appears to have achieved its aims, with modest downward effects on medium- to long-term interest rates, though smaller than those from QE1. The Fed was getting down to the light artillery. Was there anything left in its arsenal? Check that: Was there anything left that the fractious FOMC could agree on? There certainly were other options. For example, I was urging the Fed to cut the interest rate it paid on excess reserves and to conduct QE in non-Treasury securities.

THE FRACTIOUS FED

Then Greece struck again in the April–July 2011 period. Under pressure from more bad budget news (Greece would fail to meet its deficit-reduction targets) and a sagging Greek economy (which made budget targets much harder to meet), interest rates on Greek sovereign debt spiked anew—from about 12.7 percent on 10-year bonds at the beginning of April to about 17.8 percent by the middle of June. By mid-August Greek rates were higher yet.

As if in step, the U.S. economy foundered again. GDP grew by a paltry 1.3 percent annual rate over the first three quarters of 2011. The economy was no longer earning gentlemen’s Cs; it was getting Fs again. These adverse developments strengthened the hands of the FOMC doves—led by Vice Chair Janet Yellen, the New York Fed’s Bill Dudley, Boston’s Eric Rosengren, and Chicago’s Charles Evans. You might have thought such a weak economy would silence the hawks, too. But it didn’t.

Instead, a trio of Reserve Bank presidents—Richard Fisher of Dallas, Narayana Kocherlakota of Minneapolis, and Charles Plosser of Philadelphia—dissented loudly in August when the FOMC decided to replace its “extended period” language by the aforementioned pledge to hold the federal funds rate unchanged “at least through mid-2013.” Their dissents raised eyebrows because three no votes at a single meeting is virtually unheard of on the consensus-bound FOMC. The dissenters had a point: Economic conditions should trump the calendar. But casting a dissenting vote against wording—as opposed to against a policy action—is quite unusual.

The FOMC’s reentry was not over. With the economy sputtering, the doves ascendant, and the hawkish trio still dissenting, the Fed announced in September 2011 that it would embark on what the media termed “Operation Twist.” The objective of Twist was the same as that of QE2: to reduce long-term interest rates, this time by buying longer-dated bonds (as QE2 did) while selling shorter-dated bonds (which QE2 did not).

Operation Twist could be seen as a kind of compromise. On the one hand, unlike QE1 and QE2, it did not expand the Fed’s balance sheet—which was the main bugaboo of inflation hawks both inside and outside the Fed. Largely for this reason, Twist was a weaker stimulus than QE2. On the other hand, it presumably constituted stronger medicine than doing nothing. If Operation Twist succeeded in nudging down longer-term Treasury rates, it would probably nudge down longer-term private borrowing rates as well. The Fed hawks, however, were in no mood to compromise. Messrs. Fisher, Kocherlakota, and Plosser dissented against starting Operation Twist in September. The normally collegial FOMC was getting testy.

Things got worse at the next meeting. Oddly enough, the FOMC’s three voting hawks did not repeat their dissents against Operation Twist in November. To further cloud the picture, the Chicago Fed’s Evans, who had been arguing for stronger expansionary medicine for months, dissented on the dovish side instead. Chairman Bernanke now faced an almost unprecedented spectacle: He was getting dissents from both sides.

While the Federal Reserve’s equivalent of a civil war was going on, what was happening to expected inflation—the presumed fear of the hawks? Figure 14.7 told the tale earlier in this chapter. As measured by the difference between the nominal and real interest rates on 10-year Treasuries, the expected inflation rate was 2.3 percent at the close of 2010, rose to as high as 2.6 percent, and then fell to as low as 1.8 percent during 2011, and closed the year at 2 percent. The hawks may have been harboring inflation fears, but the financial markets weren’t.

IS UNCONVENTIONAL MONETARY POLICY THE NEW NORMAL?

The Fed ran out of conventional monetary policy ammunition on December 16, 2008, the day it reduced the federal funds rate to approximately zero. From that point on, it had two basic choices. It could fold its tent, leaving the economic recovery to the fiscal authorities and the economy’s natural recuperative powers. But Bernanke was not about to let that happen. Years earlier he had promised that Federal Reserve passivity would never again cause a depression. Alternatively, the FOMC could resort to a variety of second- and third-best unconventional monetary policies, such as quantitative easing and verbal commitments. Which, of course, is exactly what they did—and are still doing. Exit is nowhere in sight.

The likely outlook for the next year or two is for unemployment well above the Fed’s target of around 5.6 percent, inflation at or below the Fed’s 2 percent target, and extremely low nominal interest rates. If that scenario comes true, the Fed will have to stick with unconventional monetary policies (UMPs) for several more years. Even when things return to normal, we are likely to be living in a world of 1 percent to 2 percent inflation, not 4 percent to 5 percent inflation. In that sort of environment, random events are far more likely to push the federal funds rate down to near zero in the future than in the past—calling again for UMPs. Let’s recall briefly what some of those are:

Two classes of UMPs have been discussed extensively in this chapter because the Federal Reserve has deployed them time and again. One is making verbal commitments to keep the fed funds rate low for “an extended period,” or “at least through late 2014,” or until certain interim targets for inflation and unemployment are achieved. The other is conducting large-scale purchases of assets, as in QE1, QE2, and Operation Twist. We may see more of both of these UMPs from the Fed in the coming years, as we did in September 2012.

But there are other options, and if the economy falters, we may see those, too. As a rule, central bankers disdain these other options because they are so out of the ordinary. In fact, they don’t much like even making long-run verbal commitments or conducting large-scale asset purchases—the two UMPs the Fed has already employed. But given the way the U.S., European, and Japanese economies look right now, things may remain unconventional for a while longer yet. As discussed earlier, the Fed can:

Peg one or more bond prices: Instead of purchasing a stated dollar volume of bonds, as in a typical QE program, the Fed can pledge to buy as many bonds as necessary to drive the corresponding interest rates down to prescribed target levels. Example: The Fed could purchase as many 10-year Treasuries as it takes to drive their yield down to 1.5 percent.

Reduce the interest rate it pays on excess reserves: Instead of inducing banks to hang on to idle reserves by paying them to do so, the Fed could pay banks nothing, or even charge them a modest fee, to encourage bankers to do something more useful with their money. Example: The Fed could charge banks 25 basis points instead of paying banks 25 basis points.

ARE WE WAITING FOR GODOT?

Godot took his time. Similarly, the Fed may have to wait a long time for its opportunity to exit.

In 2012 the U.S. economy just plodded along. The Fed started a new round of experiments with a different sort of verbal commitment—publishing forecasts of its own behavior—in January. Those forecasts included widely disperse predictions from FOMC members about when the exit from near-zero interest rates would start. Then, in September, it extended its super-low interest rate pledge to “at least through mid-2015” and surprised markets by making QE3 an open-ended commitment to buy mortgage-backed securities at a rate of $40 billion per month.

As this book went to press, there was not yet much sign that exit is in sight. Indeed, few people even clamor for it any longer. The time for Federal Reserve exit will come one day. But it’s not here yet.