Toward a New Consensus on Monetary Policy
If you’ve ever experienced high inflation, you’re unlikely to forget it. In the decades between the end of World War II and the creation of Europe’s new currency, Germany’s central bank set the global standard for sound finance and monetary conservatism. Germany’s folk-memory tied the hyperinflation of the 1920s to the destruction of German society and the rise of Adolf Hitler. “Never again” was the idea that motivated, and to some degree still motivates, German monetary policy.
Compared with that, the global stagflation of the 1970s seems hardly worth mentioning -- but it still made quite an impression on those who lived through it. Inflation in the U.S. was more than 12 percent in 1974. It subsided, but under the influence of the decade’s second great oil-price shock, it surged to almost 15 percent in 1980.
Inflation on that scale makes a mockery of prudent saving, overthrows carefully planned investments and destroys fixed incomes. It’s a wrenching, miserable experience. It shaped an intellectual consensus, which in turn formed the views of today’s top central bankers. The question now is whether, and how far, those views ought to change again.
Central bankers dread inflation not just because of the immediate harm but also because of the damage they have to inflict to quell it. Once high inflation takes hold (wages chasing prices, causing prices to rise further, causing wages to rise further, and so on), a central bank has to prove it means business. This can mean provoking a recession -- like the one Federal Reserve Chairman Paul Volcker forced on the U.S. economy in 1981-82.
Recall that in 1980 U.S. inflation was intolerably high. Recall, too, that the cause wasn’t an “overheating” economy. After the oil-induced recession of the mid-1970s, the U.S. still hadn’t returned to full employment. By 1981, Volcker had pushed the Fed’s policy rate up to 20 percent. The cost of funds for prime borrowers rose to almost 22 percent. Unemployment soared -- as a deliberate act of policy -- to almost 10 percent.
The controversy over this assault on inflation makes today’s debate about quantitative easing seem tepid. Volcker was widely reviled, and popular protest over the monetary squeeze spilled onto the streets of Washington.
And yet: The policy succeeded. Inflation came down, and the economy recovered. Soon Volcker was revered for his determination and discipline. The experience, replicated in the U.K. a few years later under Prime Minister Margaret Thatcher and elsewhere, cemented the idea that the principal qualifications for a good central banker were a loathing of inflation and a willingness to court unpopularity to keep it suppressed.
The notion that central banks should be made independent of finance ministries followed more or less directly: If doing tough, unpopular things was a job requirement, then central banks would need to be shielded from the importuning of weaker-willed elected politicians.
Surprisingly, most governments saw this as a convenient division of labor -- and a democratically permissible one as well. Suppressing inflation, they told themselves, didn’t involve trade-offs of the kind that required political mediation. Things would be different if tolerating higher inflation might produce permanently higher employment. In that case, choosing the right balance between a good thing (more jobs) and a bad one (higher inflation) would be a task for elected politicians. But economists assured them this trade-off was strictly short-term, and not one they could responsibly exploit.
This brushed an important issue under the carpet -- how to manage the short-term trade-off. Few economists denied that the central bank could choose whether to restore employment after a downturn quickly or slowly, or that this choice had implications for price stability. And it’s hard to argue that this narrower question should be hoisted above politics. Nonetheless, the larger benefits of shielding central banks from popular pressure were deemed so great that governments settled for the new dispensation. They told central banks to concentrate on keeping inflation low, and surrendered power to the technocrats.
Independent central banks and an unswerving commitment to keep inflation low: These were the elements of the post-Volcker consensus. Today both ideas need a second look.
The first big challenge to the consensus was Japan, which fell into a deflationary spiral in the 1990s. Deflation -- persistently falling prices and wages -- isn’t quite the opposite of inflation. It introduces complications all its own. Central banks can curb inflation by raising interest rates as high as necessary, but they can’t stop deflation by forever lowering them. Nominal interest rates can’t fall to less than zero. Worse, when prices are dropping, short-term interest rates stay positive in real terms even if nominal rates are zero. The “zero lower bound” disables the central bank’s main policy tool.
Meanwhile, falling prices increase the inflation-adjusted burden of debt, which keeps the economy in a prolonged stall. In this so-called liquidity trap, with no way to cut interest rates further, the central bank can’t use orthodox methods to revive the economy. Its goal is unorthodox too: It has to encourage inflation rather than suppress it.
Drawing on their knowledge of the Great Depression, experts were beginning to understand the economic analytics of Japan’s plight by the mid-1990s. In a situation like this, they concluded, it wouldn’t be enough to restore price stability -- meaning a low rate of inflation. To cut real short-term interest rates as much as necessary, central banks had to promise a controlled blip of higher inflation. In a classic scholarly paper in 1998, Paul Krugman put the point provocatively: The central bank had to promise to be irresponsible.
Ben S. Bernanke, a scholar of the Depression before he was chairman of the Fed, also recommended that Japan commit itself to a higher rate of inflation. So did others. Only now, more than a decade later, has the Bank of Japan raised its inflation target -- all the way from 1 percent to 2 percent. Under the leadership of its new governor, Haruhiko Kuroda, it may even try to hit it.
What’s strange is that for more than a decade, most policy makers have behaved as though Japan’s experience has no great relevance for other countries -- even after 2008, when deflation threatened the U.S. and Europe. This bears witness to the agonies of the early 1980s and the subsequent tenacity of the post-Volcker consensus. It’s as though the correct prescription for Japan was knowledge too dangerous to recognize. It couldn’t ever make sense for a central bank to be irresponsible.
In theory, however, it sometimes does make sense. In later refinements of the economic models, this feature persists: Not just during outright deflation, but also when short-term interest rates reach the zero lower bound, a spell of inflation that exceeds the central bank’s normal target is usually warranted.
Theorists reach this conclusion in a variety of ways. For instance, economists weigh the merits of setting a target for the level of prices rather than inflation, or of promising to target a projected path for the money value of gross domestic product (also known as nominal GDP, or NGDP). In both cases, though, a spell of higher-than-desired inflation is involved, and this is the very property that each approach relies on for success.
Consider what might happen if the Fed had a target for the path of NGDP in place right now. Suppose that path was calculated by taking the growth in nominal GDP experienced between 1990 and 2008 and extrapolating it forward, through the years of the recession and beyond. By 2013, because of persistent undershooting, the gap between that path and the economy’s actual NGDP would be huge -- some 14 percentage points.
If the Fed announced its intention to close that gap as quickly as it could, it would be telling the financial markets to expect both a huge monetary stimulus and a rise in inflation well in excess of the Fed’s “price stability” benchmark of 2 percent. The jump in inflation would cut real short-term interest rates and stimulate demand. It wouldn’t be an unwanted side-effect. It would be part of the remedy.
Opponents might see this cure as worse than the disease. Even some advocates would concede that the Fed would have to be cautious in closing so great a distance between actual NGDP and the target-path. But the advocates would also point out that the gap wouldn’t have grown so large if the target had been operational from the start -- that is, if the Fed’s monetary stimulus had been adequate in 2008 and after. Nonetheless, the idea of trying to correct past mistakes rather than ignoring them -- by essentially resetting the target every year -- is a vital part of the NGDP-level approach. Tolerating, even seeking, a temporary rise in inflation is an essential part of this method.
The idea remains controversial -- and in fact it’s an old controversy. As Charles Goodhart, one of the U.K.’s leading monetary economists, recently wrote, “Adopting a nominal income (NGDP) target is viewed as innovative only by those unfamiliar with the debate on the design of monetary policy of the past few decades.” He continued: “A NGDP target would be perceived as a thinly disguised way of aiming for higher inflation. As such, it would unloose the anchor to inflation expectations, which could raise, not lower, interest rates by elevating uncertainty about the central bank’s reaction function.”
Supposing for a moment that skeptics like Goodhart are wrong and a prescription like this makes sense, could it actually be dispensed? In other words, if a central bank decided it wanted to raise demand and drive up the rate of inflation, would it be able to do so? The answer is yes -- but the details only add to the fears of those who hold to the post-Volcker consensus.
At the zero lower bound, nominal short-term interest rates can’t be cut. How then is the central bank to engineer the necessary rise in demand?
By itself, a target for inflation won’t suffice. Simply promising to raise inflation would be enough only if investors believed the promise. But financial markets aren’t as credulous as central banks might wish. To squeeze inflation out during the 1980s, the Fed and its counterparts had to do more than make promises. They had to cause a recession by raising interest rates.
What levers do central banks have in deflationary conditions? In the past few years, all have relied to a greater or lesser extent on quantitative easing: buying longer-dated government bonds and other securities to drive up their prices and drive down their yields.
The evidence is that it has worked. It has reduced long-term interest rates, which has encouraged investment, boosted share prices and helped revive depressed markets for housing. Even so, in practice the mechanism has been less powerful than it could have been.
The Bank of Japan has set the standard for self-neutering QE. First, it bought mostly short-dated government debt. Those assets are close substitutes for money, and little is gained by swapping one kind of money for another. Second, its announcements on projected inflation practically promised that the policy would have no effect. So much for guiding expectations.
The Fed, in contrast, has combined larger-scale purchases of long-term government bonds and mortgage-backed securities with evolving “forward guidance” on interest rates. In the most recent iteration, it promised to keep interest rates close to zero as long as U.S. unemployment was more than 6.5 percent and inflation less than 2.5 percent.
This announcement contained two subtle yet important innovations. First was the explicit reference to labor markets. This tilted the balance of the Fed’s “dual mandate” -- price stability and full employment -- away from the inflation component, hitherto first among equals. Second was the inflation ceiling of 2.5 percent, higher than the Fed’s usual benchmark of 2 percent.
Strictly speaking, the move didn’t constitute a new inflation target: It merely clarified the permissible margin of error around the old one. Still, it was meant -- and perceived -- as a signal that the Fed was easing further. It wasn’t an outright commitment to raise inflation, but it sure wasn’t a Volcker-style declaration of anti-inflation zeal.
The Bank of England has found a different middle way by accident: affirming its inflation target, for which it’s accountable under the law, even as it repeatedly exceeds it. The central bank has been doing what it can through QE and other means to provide monetary stimulus; a tighter monetary policy -- tight enough to keep inflation on target -- would have slowed the limping British economy even more.
And then there’s the European Central Bank, which shifted course under the leadership of Mario Draghi last year, leading markets to expect a more flexible approach to monetary policy. The ECB’s new toolbox contains “outright monetary transactions,” another name for QE. But the ECB hasn’t tampered with its inflation goal. The treaty that created the bank instructs its board to achieve “price stability,” and from the start the bank has defined that term as meaning inflation of “below 2 percent.”
The combination of QE and an explicitly higher inflation goal -- the step central banks are so reluctant to take -- would be far more powerful than QE by itself. But the most powerful monetary stimulus of all would go even further than this.
With or without a more commodious inflation target, QE is self-limiting in the sense that the central bank promises eventually to unwind it. This is why financial markets are so preoccupied with the Fed’s exit strategy. Once the U.S. economy is back at full employment, investors assume the Fed will start to sell the securities it has bought under the QE program back into the markets, thus shrinking its balance sheet, reducing the supply of money and returning monetary policy to normal operations.
The exit from QE could be done sooner or later, quickly or slowly. Because the volume of securities is so large -- the Fed currently holds $3 trillion worth -- such details will be important for investors.
There’s still another possibility, however, broached only recently by Bernanke: The Fed needn’t sell all its securities. Instead, it could wait for some to reach maturity, and then let the Treasury redeem them.
But what if the Treasury simply rolled over its debt to the Fed, replacing maturing bonds with new perpetual instruments? Then the enlargement of the Fed’s balance sheet and the monetary stimulus it represents would never be unwound. The debt in question would be permanently monetized. The loans the Fed gave the government, in the form of bond purchases, would essentially become grants. It wouldn’t ask for its money back.
There’s another name for this form of QE: “helicopter money.” The metaphor comes courtesy of Milton Friedman, the father of monetarist economics, who liked to point out that deflation could always be defeated, if need be, by printing money and dropping it from helicopters. Directly monetizing government debt amounts to the same thing. To say that the post-Volcker consensus frowns on the idea would be something of an understatement. The treaty that created the ECB simply made it illegal.
Note that directly monetizing debt -- the most powerful form of monetary stimulus -- is actually a hybrid of monetary and fiscal policy. The government is using the central bank’s printing press to create purchasing power out of thin air. It’s consuming and distributing real resources, and paying for them with money it simply manufactures. This isn’t just inflationary, it also makes a mockery of the principle of central-bank independence. In ordinary times, helicopter money is the defining example of fiscal irresponsibility. That’s why the European Union outlawed it.
In 2002, when Bernanke was a Fed governor but before he became chairman, he famously made the same point as Friedman in a speech about deflation. To his critics, he’s been “Helicopter Ben” ever since. For good or ill, he doesn’t deserve the name. As chairman, he has shown no interest in the idea of directly monetizing debt, even though in recent years the possibility of deflation in the U.S. has been more than an academic curiosity.
It’s the consensus again. Helicopter money; promising to be irresponsible; setting out to raise inflation. In deflationary times, these are serious ideas, deserving of serious attention. But in central-banking circles they’re ideas that dare not speak their name.
Last month Adair Turner, chairman of Britain’s Financial Services Authority, gave a speech in which he broke this taboo and called for “overt monetary finance” (his term for helicopter money) to be taken seriously as a policy option. Discussing it openly would inform the wider debate about monetary policy, he said; there might be extreme circumstances in which it’s the right thing to do; and talking about it now would reduce “the danger that we eventually use this option in an undisciplined and dangerously inflationary fashion.”
For the moment, Turner appears to be in a minority of one among senior financial officials in wishing to start this conversation. But there are other signs that the consensus might be fraying at the top. One is the confusion that attends central-banking policy statements these days, despite (or perhaps because of) the new commitment to transparency. Innovation under pressure tends to be an untidy process, and the clarity of the old understandings has gone even if the cultural commitment to them hasn’t.
Changes of leadership also bring fresh ideas. The Bank of Japan has a new boss, not like the old boss. And the next governor of the Bank of England, Mark Carney, seems open to defying orthodoxies. He’s expressed interest in NGDP targets, for instance.
Turner gave an interview to the Financial Times last month, and the subject of German intolerance of inflation came up. Turner mentioned the hyperinflation of 1923 as the “iconic event” that shaped Germany’s morbid fear of inflation -- then pointed out that the history isn’t quite right. The Nazi Party achieved its breakthrough several years later, and during a period of falling, not rising, prices.
It’s interesting, he said, that Germany’s reading of history places so much weight on dangers of inflation rather than deflation. Questions that seem settled are sometimes worth revisiting.
To contact the senior editor responsible for Bloomberg View’s editorials: David Shipley at firstname.lastname@example.org.