What are they talking about? Understanding the Bernanke Fed
Remarks by Greg Ip to the Money Marketeers of New York University, Jan. 16, 2008
Ben Bernanke delivered a forceful speech in Washington last week suggesting the Fed was going to be quite a bit more aggressive about easing monetary policy. It made quite an impression on a lot of people. J.P. Morgan’s report was called: “The Chairman takes charge.” Lyle Gramley’s report was titled: “Bernanke stands tall.”
These broker notes reflected the suspicion that before that speech, Bernanke had been something of a shrinking violet, letting his lesser ranked officials overpower his message and sway his policy. He was too academic, dwelling on the ins and outs of liquidity policy and inflation expectations when what the U.S. wanted was lower interest rates, yesterday.
Interestingly many of these qualities were being praised when he was nominated in 2005. Bush had been perceived as putting loyalty above qualifications and installing polarizing outsiders to run institutions like the CIA and the World Bank. He drew sighs of relief when the man named to run the Fed was someone with a reputation for being collegial, nonpartisan, of great technical ability, and with no record of partisanship. Far from sharing the administration’s penchant for secrecy, Bernanke was a longstanding advocate of more openness.
What’s happened? Have Bernanke’s assets turned out to be liabilities? Or is the story more complicated? Much of what we’ve experienced has little to do with Bernanke per se or even communications, but the stage of the economy. I also think the communications problems that have occurred reflect long-standing tensions within the Fed.
But it’s probably also true that some of Bernanke’s style has muddied the Fed’s message in the last five months. The Fed’s leadership, as we reported in Monday’s Wall Street Journal, is moving to address that. I wouldn’t expect a reversal, though. What Bernanke is trying to accomplish reflects his core beliefs.
Communications problems at the Fed are not new. They occurred even when the Fed didn’t communicate much. There’s the legendary “Thanksgiving day Turkey” of 1989 when a routine open market operation was widely misinterpreted as a policy easing. That incident was one reason the Fed decided to start announcing rate changes. Yet traders remember that the first announcement and accompanying rate increase in February 1994 as a monumental sandbagging.
Starting with that move in 1994, the Fed began to communicate more, and its statements took on added importance. As a result the opportunity to mislead also grew. In November, 2000, as the markets were screaming for the Fed to ease, the Fed issued a statement saying the risks were tilted to inflation. A read of the transcript of that meeting released five years later suggests the Fed didn’t really think inflation was the greater risk; they only said so to keep market expectations in check.
Finally, there was the summer of 2003 when the federal funds rate was approaching the zero nominal bound. Greenspan had called for a “firebreak” against deflation and numerous Fed officials, including Greenspan and Bernanke, had raised the prospect of buying Treasury bonds to influence long-term rates. There was rampant speculation the Fed would ease by 50 basis points and start to buy bonds.
When the Fed did neither, Wall Steet’s outrage spilled over. The Fed has “undermined the popular myth” that it is a “paragon of virtue and effectiveness when it comes to getting its message across,” Melvyn Krauss wrote in the Financial Times. “The Fed’s [statement is] – a badly worded and confusing text … The Fed’s loss of credibility will undermine its ability to guide the economy through the storms that are currently gathering.”
These incidents all came around perceived turning points in the economy when the outlook and the appropriate policy response were muddier than usual. That is certainly true of the present.
Arguably, Bernanke’s principal problem in the last five months hasn’t been communicating, it’s forecasting. Since last August, the Fed has systematically underestimated the gravity of the housing collapse and credit crunch, and has thus systematically communicated a policy stance that has rapidly been overtaken by events. Bernanke initially thought targeted liquidity measures could do a lot of the work of monetary policy. He thought the damage would be contained to the housing market. He was wrong about both those things. He also thought inflation pressure were a serious obstacle to more aggressive easing. He might yet be right that inflation is serious, but he seems to have changed his mind about letting it get in the way of aggressive easing.
Of course, the Fed wasn’t alone. Most people in this room were similarly wrong and I lay no claim to clairvoyance. But a lot more hinges on Bernanke’s forecast than mine.
So, it goes without saying that, as in any other policy or business field, being the best communicator in the world cannot make a central banker that keeps getting his forecast wrong look smart.
Still, while good communications can’t take the place of good policy, it enhance it or muffle it. My colleague David Wessel likes to say there are two things expected of a central banker: get interest rates right and look like you know what you are doing.
When Bernanke came to the Fed with he had no quarrel with how Greenspan set interest rates. Greenspan has a lot of critics for being too easy from 2002 to 2004, but Bernanke isn’t one of them; he gave the policy full throated support.
What Bernanke wanted to change, rather, was how the FOMC ran itself. He wanted it to be more democratic and more transparent.
This was not some Road to Damascus epiphany. These are longstanding features of his character. Presidential nominees are usually advised not to say anything publicly until their nomination has been announced. Someone forgot to tell Bernanke that and when a Reuters reporter asked him about in 2002, he confirmed it, on the record. In 2003, when Bernanke was still a governor and occasionally giving on the record interviews, I asked him whether the Fed could create too much volatility and hurt itself if it talked more. He told me that if the Fed talked less, someone else would fill the vacuum. “From individuals speaking or from pundits speculating, it’s just going to be noisier and more cacophonous than ever,” he says. “Any kind of guidance we give from the FOMC has got to be better than that.”
As to being non ideological when Bernanke was on the Montgomery Township School board in New Jersey, he negotiated two contracts with the teachers union and to pay for more schools, oversaw an increase in the average school tax bill of 70%. Gee, do you think Bush knew that when he nominated Bernanke?
At the Fed, he also learned how to bridge differences, even in areas like consumer protection. For many years federal law said banks can’t ask borrowers about their race or gender when they apply for a small business loan. This is to discourage discrimination. Congress and the Clinton Administration wanted banks to get the right to collect the data to see if in fact a pattern of discrimination existed. This required the Fed to change its “regulation B. ” Ned Gramlich, a liberal Democrat who was in charge of the board’s community affairs committee, supported the change. But some governors, including Greenspan, opposed the change. They thought community groups would get their hands on the data and use it to launch frivolous lawsuits against banks. This division kept the board deadlocked for years. After Bernanke joined the board in 2002 he proposed a compromise, according to my sources. In effect he said, why don’t we let lenders collect the data, but give them a safe harbor that says the data can’t be used to sue them if it’s collected in good faith? This broke the deadlock and in 2003 the rules were amended.
But bringing the same approach to monetary policy is trickier. There is a longstanding tension in how the FOMC operates. The Fed is a central bank that tries to decide policy by consensus. It is also a collection of independent individuals who speak for themselves. In its first role, it likes to speak in one, clear voice. In its second, it speaks in many, often inconsistent voices.
Under Greenspan, that one clear voice was unambiguously his. The very first statement, in 1994, was actually called the chairman’s statement. It later became the FOMC statement but it was still largely the creation of Greenspan and his closest aides. Sometimes, the rest of the FOMC didn’t get to read it until the meeting was virtually over and people were eying the lunch buffet. As the statement became more and more important in shaping the path for future monetary policy, the rest of the FOMC clamored for more input, which Greenspan gave them, bit by bit.
Bernanke came in having long championed the idea that monetary policy should be as systematic and scientific as possible. In the early 1980s, Bernanke taught economics at MIT with future Bank of England governor Mervyn King, who sometimes attended Bernanke’s lectures. Maybe some bonding went on because Bernanke became an admirer of King’s monetary policy regime. He likes the fact the bank has an inflation target and quarterly reports with forecasts which are the principal market-moving event in the UK. He likes the fact King gives few explicit clues about interest rates in his speeches, seldom moves the market, and is even occasionally outvoted, while appreciating the setup is different from the Fed’s.
As a reporter, I like this stuff, too. The more information the central banker gives, the more there is for me to report on. And a committee where many people have different views is more complex and interesting to follow than an individual. There is also some academic evidence that committees make better decisions than individuals.
But to my surprise I have found that many on Wall Street don’t like these things. They want the Fed to speak with a single voice, not too often, and as unambiguously as possible about the next change in interest rates, which is of course the one thing the Bernanke doesn’t want to say.
This demand for simplicity is why Wall Street has been frustrated by the multiplicity of messages from different FOMC members since August. This, again, is not necessarily new. Shortly after Larry Meyer joined the Fed, Greenspan advised him not to talk about monetary policy, but if he insisted on doing so anyway, not to move the market. In spite of that, Larry regularly did.
Back on April 10, 2001, St. Louis Fed president Bill Poole memorably told one news organization, “There are compelling times when quick action is necessary, but this is not one of them.” Eight days later the Fed cut interest rates. The next month, Poole was at a panel discussion where he said, “I have to offer the standard Federal Reserve disclaimer: Perhaps, after April 18, more people will believe it than perhaps did before: I am speaking only for myself.”
In some ways, there ought to be less cognitive dissonance emanating from FOMC chatter these days, because the membership of the FOMC has become more homogenous. More of the FOMC are economists, and more of them share the basic framework for the economy even if they differ on the specific parameters such as the flatness of the Phillips Curve, the sacrifice ratio, potential growth and the influence of the external sector.
This homogeneity may actually increase in coming years because of the increased emphasis on the FOMC forecasts. Greenspan despised the FOMC forecast, didn’t contribute, and pretty much ignored them in testimony. He explored getting rid of them altogether but his staff told him Congress would object. He thought they were internally inconsistent and therefore misleading, but I suspect he just didn’t like putting numbers out there. Since Greenspan relegated them to the sidelines, many FOMC members put little effort into them. Some presidents had junior staff prepare them and didn’t personally endorse them.
But if Wall Street, the media and FOMC members begin to put more emphasis on the forecasts, FOMC members will put more effort into them and they may find themselves increasingly drawing on the same toolkit.
Yet the forecasts could also aggravate the communications problems the Fed occasionally has. While in theory, each of the 12 voting FOMC members is equal, the chairman is obviously far more important. Yet the FOMC forecasts treat the views not just of the 12 voters but the seven non-voting presidents as equal because they are all included in the ranges and central tendencies. For example, the October forecasts tell us FOMC members’ preference for inflation ranges from 1.5% to 2%. But what if Bernanke is at 2% and a non-voting president is at 1.5%? Bernanke presumably formulates his policy recommendations around his personal view, but that may not correspond to our understanding of the Fed’s objective.
The forecasts are also of limited help at times like the present when risk management is the primary driver of where the Fed sets interest rates. Risk management, you’ll recall, is the notion that the Fed should set the funds rate not based on the modal, or baseline, forecast, but on the tail risks around that forecast. To cut off those tails, the Fed will raise or, more often, lower the funds rate more than the forecast would call for.
It’s almost impossible for outsiders to model the Fed’s risk management reaction function. How much weight do they assign to the risk of a mild recession? Of a deep recession? Of inflation rising half a percentage point? Those are judgment which force outsiders to rely heavily on Fedspeak.
Of course, even listening to Fedspeak has been misleading lately. In September, the Fed said both inflation and weak growth were a problem. Since August, it has implied it was equally worried about inflation as growth or at any rate hasn’t explicitly said it was more worried about growth, yet cut rates each time. On the street, and even in some corners of the Fed, there’s a suspicion Bernanke has played up inflation to placate FOMC hawks. That has probably played a part, but I wouldn’t minimize Bernanke’s own concerns about inflation.
To try and make sense of Fed chatter, the usual process is to “weight” the speakers. Assign a lot of importance to Bernanke, somewhat less to Kohn, and a lot less to the other FOMC members. Larry Meyer and Brian Sack find evidence that the market does in fact do this. Also, calibrate the speaker according to his previous predisposition. If a hawk like Charles Plosser goes from ultra hawkish to just mildly hawkish, there’s signal value there.
But weighting still leaves a lot of room for being led astray. The distribution of FOMC speakers does not match the distribution of FOMC views. Fed presidents speak far more often than governors because it’s one way of getting their views across to the FOMC between meetings. Presidents also give on the record interviews more often. None of the current Fed Governors give on the record interviews. For what it’s worth, they should reconsider that. It would balance out the commentary we get. And they are public servants, and taking questions from the press is one way to be accountable. To be sure, they do appear before Congress, but only rarely, and sometimes face few, if any, questions.
Fed officials also have an abiding fear of creating volatility. This is surprising, when you think about it. “Causing volatility” is another way of saying “moving the market.” And in theory a central banker should want to be able to move the market with his words, since that can leverage the power of his actions.
To avoid causing volatility, an FOMC member may fuzz up anything relevant to the current outlook or the next FOMC meeting. He or she may fall back on citing the FOMC statement or speech of the chairman. This can render his comments less useful, or even stale. If many members do this, it can create the impression of a coordinated message even though no actual coordination has occurred.
In late 2002, Greenspan said if the Fed funds rate fell to zero, the Fed could buy bonds. Thereafter, whenever an FOMC member was asked what the Fed should do if the funds rate fell to zero, they would say, “buy bonds.” The repetition of this answer created the impression that everyone was thinking the same thing. When the bond market we realized no such plans were in the works, it sold off sharply. Because we are told that every member speaks for himself, hearing a bunch of members say the same thing leads to the obvious conclusion that a particular viewpoint must be widespread.
Wording matters. A member who is explicit about his preference on interest rates will move the market more than a member who feels the same way but couches his language more carefully.
Timing matters. A member speaking the same day as Bernanke will have little impact. A member who is the only one on the tape that week will have more impact.
The Fed’s blackout, which lasts the week before and the week of a meeting, can make matters worse. Suppose there’s an important piece of data or market development during the blackout. FOMC members have handcuffed themselves from saying what it means for monetary policy – including saying it doesn’t mean very much.
Hawks also speak more often than doves. Former Dallas Fed president Robert McTeer, who was a dove, was fond of saying “Only hawks go to central banker heaven.” A central banker admitting to dovish tendencies is like a wine critic admitting to drinking Merlot out of a foil bag. Nothing wrong with it, but best done behind closed doors.
All of these factors make it hard to identify the core position of the FOMC just by listening to a lot of FOMC speeches.
It is unlikely that Bernanke is going to tell his colleagues to speak less or get their speeches vetted. Bernanke believes in more information, not less. And self-censorship could deprive the FOMC of important dissenting views and breed groupthink.
Finally, the Fed has lost its Miranda rights. The markets do not give it the right to remain silent, because silence is taken as approval of whatever the market is pricing. In October, as equity markets deteriorated, the market began pricing in a cut at the end of the month. Since FOMC officials did not explicitly counter this pricing, their silence was taken as affirmation. But in fact officials were not at all convinced another cut was needed, and may in fact have stood pat absent overwhelming probability of a move. Similarly, in the runup to the December meeting Wall Street began to expect a narrowing in the discount rate spread. Since no one from the FOMC pushed back, it became the consensus and the Fed’s failure to act fueled the market rout.
The Fed is addressing some of these issues. As we reported on Monday, Bernanke and company have decided one way to clarify the message without reining in their colleagues is for him, and Don Kohn to speak more often. This won’t guarantee the message is clear or that the forecast is right, but it will help the market identify the FOMC’s center of gravity.
Fed watchers, including me, could learn some lessons, too: FOMC silence is not affirmation of the market pricing; be aware of how timing, wording and self selection can overamplify a speaker’s words. In the meantime, it looks like Bernanke will give me a fair bit to write about for some time to come.