Expectations, Communications, and Monetary Policy
My remarks tonight will center on expectations, communications, and monetary policy. Specifically, I would like to share with you my thinking on how the Federal Open Market Committee has "learned to talk" - in other words, how the FOMC has evolved from being regarded as secretive to now being viewed as more transparent. My discussion will emphasize three points.
First, consistent behavior matters.
Second, the FOMC's communications - how we talk about policy goals and actions - have played an important role in enhancing the effectiveness of monetary policy.
Third, we should consider taking additional steps toward greater transparency when it seems likely that they can further enhance the effectiveness of monetary policy.
Please note that the views I express today are mine alone. I do not presume to speak for any of my colleagues in the Federal Reserve System.
I. Consistent Behavior Matters
Let me begin, then, with why I think consistent behavior matters. One of the most important insights of modern macroeconomics is rational expectations. Economists have always recognized that expectations guide the behavior of individuals, businesses, and policymakers. But central bankers owe a great deal to the academics who led the rational expectations revolution of the 1970s and 1980s. One of their key insights is that although people are bound to make mistakes in assessing policies, policymakers themselves cannot systematically exploit these errors.
Specifically, in the case of monetary policy, people will recognize when central bankers have incentives to try to trade a little inflation for extra short-run output. For example, if people believe that the central bank is willing to boost economic activity in the short run by creating greater-than-expected inflation, then they will come to expect the highest inflation rate that the central bank will tolerate. This is clearly a worse outcome for everyone compared with a situation in which people expect - and the central bank delivers - price stability. Establishing a credible commitment to price stability, and conducting monetary policy consistent with that commitment, solves this problem.
Many of you will recognize this line of thinking as one that began with Finn Kydland and Edward Prescott, and was later enriched by other economists. Kydland and Prescott were awarded the 2004 Nobel Prize in economics partly because of this research. I was particularly pleased to learn of this award, because Finn has collaborated with our research staff in Cleveland for a number of years.
As a result of this rational expectations work, and the research it prompted, academic economists began to encourage policymakers to follow predictable rules of behavior. The new theories emphasized the importance of consistent behavior and created a whole new context for thinking about central bank disclosure.
The burden of proof for maintaining secrecy began to shift. Economists began to challenge policymakers to provide more information, and for good reason. People spend resources trying to figure out what policymakers are doing, and then decide how much to save, what kinds of assets to acquire, which skills to learn, and so on. These decisions, in turn, affect the economy's future growth path.
I think that the new emphasis on consistent behavior puts an even higher premium on disclosure. In the simple worlds of economic models, it is relatively easy to specify an appropriate policy rule for setting the federal funds rate. But in the real world, this is a much harder task. It is nearly impossible to specify in advance all of the situations that might arise and how the central bank should react to them. However, I do think we get closer to the ideal environment by being clear about our objectives, by responding consistently and predictably to economic conditions, and by communicating promptly and clearly about our actions, especially when unusual situations present themselves.
II. The FOMC's Communications
At this point, I would like to talk more specifically about FOMC communications. Although the FOMC has political independence to implement monetary policy, ultimately it is a public institution. Accordingly, the FOMC provides information to the public through the minutes of its meetings, the semi-annual monetary policy report to Congress, regular appearances by the Chairman and other Board members before congressional committees and, with a lag, the release of full meeting transcripts. All of these communications help the FOMC fulfill its obligation to be responsible to the Congress and the public.
As important as these public communications are, tonight I want to focus on the role of communications in the implementation of monetary policy. The 1990s, in my estimation, will be remembered quite favorably as the decade when the FOMC first learned how to talk. Although the process is ongoing, I believe that the Committee's communications have made for better monetary policy. Through its communications, the Committee has provided the public with a more complete explanation of its policy decisions.
I think about this progress in terms of two distinct types of information. One category is historical - the rationale that the FOMC provides for decisions it has already taken. Another category is more forward-looking - the opinions that the Committee holds about the future state of the economy and its intentions regarding likely policy actions. These intentions are steeped in probability and subject to revision.
The transcripts of the FOMC meetings in the mid-1990s reveal that the Committee's early discussions about enhanced communications centered on whether, and how, to reveal information about the decisions it had already made. Late in the decade, the discussions broadened to include communications about the Committee's intentions. Let's take a closer look at these two types of information, beginning with the historical information.
A general argument for secrecy during the 1980s and early 1990s was that markets might not react appropriately to the disclosure of policy information. Just think about this. Before 1994, the FOMC did not reveal its policy decisions immediately. The prevailing mindset was that markets were better off not knowing for certain what everybody seemed to know - the FOMC's operational objective.
It is useful to recall the precise situation in February 1994. The federal funds rate hike at that FOMC meeting followed a long period - 15 months, to be exact - of no change in the funds rate target. The transcripts of that meeting contain comments suggesting that disclosure of the policy decision would be desirable because it could enhance the public's understanding of the change. The transcript also shows that there were mixed opinions as to whether it would be a good idea to adopt such announcements as standard practice.
When the Committee announced its policy decision that February, it also chose to convey that announcing the decision should be regarded as a temporary departure from its customary practice. Clearly, many FOMC members were still concerned about introducing volatility in financial markets if announcements were made as a matter of course. The initial announcement did not even refer to the funds rate directly. The language was still couched in terms of increasing "the degree of pressure on reserve positions."
I think of the period since 1994 as one in which the FOMC started learning how to talk. To me, the FOMC learning how to talk is in some sense like a person learning how to walk through a dark room without knowing where the furniture is. You move very slowly, feeling your way. Sometimes you discover you're a bit off course. You may stumble, but you learn to adapt and move to your destination.
From my perspective, changes in Committee communications since February 1994 look like steps in the process of refining and becoming more comfortable with greater transparency. Many of you remember these steps: In February 1995, the Committee agreed to announce every change in the stance of monetary policy on the day of the decision. A few months later, in July 1995, the Committee agreed to replace the phrase describing reserve pressures with an explicit number for the funds rate target.
Let me now turn to the treatment of forward-looking information. More recently, steps toward greater transparency have been made to condition expectations about what might happen at future meetings. The "policy tilt" language first appeared in May 1999. The basic idea was to convey a major shift in the Committee's sentiments about potential changes in its federal funds rate target at some point in the future, even if no actual change in policy occurred at the meeting that produced the statement. Unfortunately, markets seemed confused by the language and by subsequent comments from FOMC participants. This reaction caused the Committee to work out the kinks in the language, which resulted in the "balance of risks" statement that the Committee adopted in February 2000. The construction of the balance of risks statement became more flexible in March 2003, and the statement remains in use today. So, in terms of my analogy, I might say that the stumble in the dark led to some careful course corrections.
You are certainly familiar with the "considerable period" language first adopted after the August 2003 meeting. That language was designed to address a particular problem at a particular point in time. Specifically, as we all know, the language was aimed at being perfectly transparent about the Committee's desire to avoid unwelcome disinflation. As that problem passed, the language was gradually modified, evolving into the "measured pace" phrase that the FOMC has used since May of last year.
I think that, on balance, innovations in communications have improved the effectiveness of monetary policy and thus have enhanced economic welfare. Studies show that since the 1980s, U.S. financial markets and private-sector forecasters have been able to better forecast the federal funds rate out several months and have been less surprised by FOMC announcements. Research at the Federal Reserve Bank of Cleveland shows that market participants expect the Committee to respond predictably to information that might have a bearing on inflation and employment.
Do markets understand our behavior because we are more predictable or because we communicate better? I suspect that both forces are at work. As they say, talk is cheap, and I am not suggesting that better communications are of any value without actions that back up the words.
A key implication of recent research is that the public understands what information is likely to guide the FOMC's policy actions, and that the FOMC and financial markets react to that information in a consistent way. Our communications have evolved over time in ways that I believe contribute to this understanding.
III. Should We Take Additional Steps to Enhance FOMC Transparency?
We have truly come a long way. But do we need to go further? That brings me to my third point - I think that we should consider taking additional steps in the direction of greater transparency when it seems likely that they can further enhance the effectiveness of monetary policy.
What we have learned about the central role that transparency can play has given rise today to the notion of a "communication policy" as an integral part of monetary policy itself. By "communication policy," I mean a strategy for reinforcing the traditional policy tools, such as open market operations, with printed and spoken words that can help the public make better decisions.
As you now know from the just-released minutes of the March 22 FOMC meeting, there are some on the Committee who think that the phrase "measured pace" has outlived its usefulness. In my view, this language has provided useful guidance in the limited time we have used it, but there is a risk that at some point the Committee will take an action that the public regards as contrary to what is implied by the language. To me, this risk suggests that the Committee should provide this type of guidance only when it is highly confident in the course of its near-term policy actions and when it perceives the cost of being misunderstood as exceptionally great.
I can tell you from my experience that although the Committee is inclined toward greater transparency, changes in its practices generally have been driven by real circumstances that confront policymakers in real time. As the FOMC pursues price stability and maximum sustainable growth, changes in our communications - what to say, how to say it, and when to say it - should be designed foremost to improve the effectiveness of monetary policy.
So far, I have talked only about greater transparency in our operating procedures. In the final portion of my remarks, I want to talk about what some regard as the "next frontier": Increasing transparency about the FOMC's inflation objective.
As you know, the FOMC discussed the pros and cons of establishing an explicit numerical inflation objective at the February meeting. I think that being more explicit about our inflation objective could help us to be successful in maintaining price stability, but my expectations are modest. I do not regard an explicit numerical price objective as a panacea.
We might gain some additional credibility with the public by simply being clearer than we are today and, at the same time, greater clarity might impose some extra self-discipline when we really need it. Let me make my own contribution to the cause. My view is that the rate of inflation should average about 1½ percent, as measured by the Personal Consumption Expenditure price index, over periods of about three to five years.
Inflation is certain to vary in the short run, even when we achieve the objective over time. So putting a range around that long-run objective makes sense to me. My personal tolerance zone is a 1 percentage point spread above and below my 1½ percent inflation objective. I don't view this necessarily as a policy-triggering boundary, but when inflation falls outside that range, I would feel more obligated to explain why I regarded that situation as acceptable.
These are my personal guideposts. I generally support the idea of a Committee objective and range. I say generally because I think it is not particularly useful to offer a blanket endorsement for a proposal that is not yet on the table. Furthermore, I'm sure many of you have been keeping score and know that some of my colleagues are in favor of more formal numerical objectives, but others are not. This does not trouble me, because I do not think it is necessary to jump to formal targeting in one leap. However, I think it would be useful to take a step in that direction.
The FOMC's semi-annual economic projections provide a mechanism for taking that step. As you know, twice a year the FOMC now provides the public with economic projections for the current year and the year ahead. The step I have in mind would have the FOMC provide an additional three- to five-year projection for inflation. This would be based on the participants' working definitions of price stability and policies that support them.
The ranges and central tendencies of these extended projections would be made public, perhaps in an expanded discussion in the Monetary Policy Report. I would not be surprised to discover that the extended three- to five-year inflation projections of the individual FOMC participants converge to a fairly narrow range. This convergence could provide the foundation for a more formal inflation objective at some point in the future.
Taking this step ought to be regarded as a logical extension of our current practice. In fact, I regard it as entirely consistent with the gradual approach the Committee has taken over the years to improve communications with the public.
I think we can all agree that the FOMC behaves more consistently and communicates more effectively to the public about its decisions today than it did in years past. Although more information does not automatically equal useful information, I believe that greater transparency has helped to improve the effectiveness of monetary policy. Going forward, I expect the Committee to adopt new communication practices as needed. After all, necessity is still the mother of invention.