Remarks by Governor Donald L. Kohn
At the Annual Meeting of the American Economic Association, Philadelphia, Pennsylvania
January 9, 2005
Central Bank Communication
A basic tenet of economics is that markets work better--reflect underlying economic forces, efficiently allocate resources--with more information. Because central banks are key players in financial markets, a better public understanding of central bank behavior should improve pricing in those markets.
Over my career, I have witnessed notable efforts by central banks to improve the public's understanding of their behavior by increasing the amount of "talk" they do about policy and the economy. The trend toward more talk reflects several interrelated developments. The demand for information has expanded along with the size and sophistication of financial markets; financial assets have increased far faster than gross domestic product, and a much greater variety of instruments are priced. At the same time, rising wealth has meant that increasing numbers of people have a stake--and an interest--in the forces determining the prices in those markets. Technological change--including the ubiquitous Internet--has made it easier to cater to and feed this interest by making more information available more quickly and cheaply. And, in part as the result of academic progress on the topic, central banks better appreciate that more-accurate expectations of market participants can help improve the performance of the economy and the achievement of economic objectives.
Changes in the political environment surrounding central banks have reinforced the increased demand for transparency. Greater support for independent central banks with goals set by the political process was one of the results of the lessons of the great inflation of the 1970s. But an arm's-length distance from immediate political pressures requires accountability and reporting to elected representatives. The quid pro quo is most evident in many of those countries that have shifted to inflation targeting, but the United States followed the same pattern earlier--in the late 1970s. In that period, laws for the first time established goals for monetary policy, albeit not as specific as inflation targets, and imposed reporting requirements in the form of reports and testimony to the Congress.
Given the trends I have mentioned and the benefits of greater transparency, people sometimes wonder why central bank transparency has evolved as slowly as it has and why some central banks do not take additional steps to talk more--especially about what they see coming in the future. I will give you my own answer to this question, which I hasten to add, does not necessarily reflect the views of my colleagues on the Federal Open Market Committee or its staff.1
The answer, I believe, is that more is not necessarily always better, and at each step of the way central banks have needed to take account of the potential costs as well as the benefits of greater transparency.
One consideration involves the nature of information and its relationship to market pricing. In fact, economists do not fully understand how markets incorporate information. Herding behavior, information cascades, multiple equilibria, and the amount of investment in financial research all pose puzzles about markets and information. The situation is complicated still more when an important participant is seen as having superior information owing to its investment in research or its understanding of its own behavior.
In such circumstances, certain types of central bank talk might actually impinge on welfare-enhancing market pricing by being misunderstood and receiving too much weight relative to private judgments. We need to be particularly careful that people understand how limited our knowledge actually is--the uncertainty and conditionality around any statement we make about future developments. Moreover, what we are doing is complicated and involves weighing many factors; we should not misrepresent it for the sake of clarity in communication. The publication of useful information is complicated further by the fact that, in most countries, policy is made by a committee; representing the thinking of a diverse group is difficult and limits what can be said.
A second type of consideration that has constrained the pace of increase in transparency is its potential interaction with monetary policy decisions. What we say is important, but what we do over time will ultimately determine economic outcomes. We should not allow a desire for clarity of expression to deflect our decisions from those that would contribute best to overall economic performance and which may be difficult to explain easily. And we must take care that policy expectations engendered by communication do not unduly constrain policy action. Furthermore, we cannot allow transparency to limit discussion in the Committee out of concern about how its publication will affect markets and the economy.
Every decision about additional transparency is a balancing of possible costs and benefits. The preexisting structure of central bank talk reached its state for particular reasons; changing it entails adjusting the balance that previous Committees found. And reversing movements toward greater communication in the event of unforeseen consequences could be difficult. But circumstances change; markets change; we gain experience through our own actions or those of other central banks. More information, generally, is better, and markets and central bank governance have greatly benefited from the steps taken around the world to increase transparency over recent decades.
Talk about "Policy Inclination" and "Economic Outlook"
The market reaction to our words could well be heightened by the behavior of market analysts, who tend not to place probabilities on their predictions of our actions over the next few meetings but instead tend to make "zero/one" calls. In any case, the risks of herding, of overreaction, of too little scope for private assessments of economic developments to show through, would seem to be high for central bank talk about policy interest rates.
Because of concern about market interpretation, policymakers often see talk about policy inclination also as having the potential for constraining future decisions in ways that might interfere with the most-effective achievement of policy objectives. The stronger the market expectations about near-term policy actions, the greater the risk of roiling markets and creating confusion in the event the decision differs from those expectations. The possibility that discussions of future policy, even nonspecific, could create presumptions about a string of policy actions makes finding a consensus among policymakers on what to say about future interest rates quite difficult--more so than agreeing on the policy today. It is no accident that the Reserve Bank of New Zealand stands out as about the only central bank to publish such a path and as one of the few in which decisions are the responsibility of only one individual.
Sack and I found that actions, not words, dominated the policy inclination factor. However, the study was completed before the Federal Reserve became more explicit about future policy rates in the summer of 2003.
The unusual situation at that time shifted our assessment of the balance of costs and benefits in favor of a public statement about our expectations for the near-term path of policy. Markets appeared to be anticipating that inflation would pick up soon after the expansion gained traction, and therefore that interest rates would rise fairly steeply. This expectation was contrary to our own outlook. We saw economic slack and rapid productivity growth keeping inflation down for some time. Our expectations about policy also took account of the fact that the level of inflation was already low--lower than it had been for several decades. We thought that our reaction to a strengthening economy would be somewhat different this time than it had been in many past economic expansions and unlike what the markets seemed to anticipate.
Under most circumstances, this sort of disconnect between the central bank and the markets would not be a big problem; we could compensate for the market's tendency to raise intermediate and long-term rates unduly by keeping policy easier for longer. But with the federal funds rate already at 1 percent, our scope for that was limited; and if the economy suffered any downward shocks, policy could reach the constraint of a zero federal funds rate, with uncertain consequences. Under these circumstances, giving markets more information about our policy inclination, and thereby holding down longer-term interest rates, seemed to be the less-risky way to stabilize inflation at a reasonable level and encourage a vigorous expansion.
I would judge the outcome to have been successful. We did influence rates to better reflect the actual path of policy; economic outcomes have been good; and, to date, our discussion of the path for rates has not constrained our actions. 3 That is partly because we have been able to pull back gradually on the degree of commitment on our near-term actions in a way consistent with incoming data and without roiling markets. As the FOMC's minutes make clear, however, our experience has also illustrated how difficult it can be for a diverse committee to talk about the future course of rates.
I take from this experience the lesson that, despite their drawbacks much of the time, conditional statements from the central bank about the near-term course of policy can be useful in certain circumstances. These circumstances might include a situation in which the policymakers and the markets seemed to have substantially conflicting forecasts about the economy and the path of policy; such differences persist despite the central bank's efforts to explain its outlook; and the effect of those divergent views on financial conditions threaten to detract from economic performance.
In my view, the most useful service the central bank can provide in this arena is its analysis of the forces bearing on the outlook--the determinants of aggregate demand, potential supply, and inflation. This type of discussion can help the public interpret developments and allow markets to respond constructively to surprises in the data. Forecasts can be used as a framework for such a discussion, but the public should appreciate the limits of a numerical forecast. The relationship of the forecast to the policy decision is loose. Inevitably, point forecasts will be incorrect; they should be seen as the centers of wide distributions of possible outcomes; and low-probability outcomes can be very important in policy decisions in certain circumstances. Moreover, quite often, the FOMC can reach a consensus on policy without reaching one on a specific forecast--only a general agreement on the general degree of strength in economic activity and the likely tilt to inflation if alternative policy paths are chosen.
Sack and I note the power of even nonspecific central bank talk about the economic outlook to greatly influence expectations embodied in intermediate- and long-term interest rates. Is it too much, as some argue?4 Is central bank talk about the economy crowding out other legitimate perspectives?
This is an empirical question, and I do not think we yet have the studies to provide an answer. One way to frame the question would be to ask whether market forecasts over the intermediate and longer run have improved or deteriorated as central banks have ramped up their talk. I suspect that the greater provision of informed analysis from central banks has led to an improvement.
In the United States, we have some indirect evidence that crowding out of private views has not increased even as the Federal Reserve has become more talkative. Market interest rates have continued to respond substantially to surprises in economic data. The degree of the reaction to individual indicators has risen or fallen depending on the economic situation and the focus of policymakers; for example, the reaction to price data decreased in recent years as low, steady inflation became more firmly established, and the reaction to labor market data increased as the "jobless recovery" became more of a focus. But we do not see a general falloff in the strength of response, as one might expect if markets were relying more on central bank views and less on their own interpretation of data.
That markets continue to react strongly to incoming data is not surprising. Predicting interest rates far enough into the future is not just about what others--including the central bank--think; over time those rates should be tied to objective factors--for example, the forces of productivity and thrift. Differing views about these factors give scope for opportunities to profit from independent research and betting against the crowd.
Earlier Release of the Minutes
In addition, the minutes are another chance for the Committee, as a whole, to talk. One advantage is that they should give the public a broader context in which to interpret the many statements by individual members between meetings. In that regard, the fact that the minutes convey the range of Committee members' views should be helpful. The minutes are not an attempt to articulate a single consensus explanation of our actions or outlook, but rather, reflecting a strength of the FOMC, they summarize the give and take in Committee discussion arising from differing perspectives on difficult issues.
Early release of the minutes could have costs if Committee members became more guarded in their discussion out of concern about the effects of their remarks when reported or if, over time, the minutes themselves became less comprehensive. In my view, neither of these developments is an inevitable consequence of the new schedule, and I am sure the Committee will resist any temptation to allow them to occur.
Early release had been considered before, but recent experience convinced members that the balance had swung in its favor. Successful dry runs alleviated concerns about the practical problems of getting timely agreement on minutes from nineteen geographically dispersed members. On one or two occasions in recent years, longer delays in release of the minutes had resulted in market confusion because the minutes were interpreted as pertaining to the most recent decision, not the one at the preceding meeting for which the minutes were prepared. Finally, the difficulty of structuring talk about the future in the brief announcement released immediately after the meeting has enhanced the perceived value of the minutes for this purpose.
Expedited release is an incremental step. The market will not have information it did not eventually have before, but it will have it sooner.
2. Donald L. Kohn and Brian P. Sack (2003), "Central Bank Talk: Does It Matter and Why?" Finance and Economics Discussion Series 2003-55 (Washington: Board of Governors of the Federal Reserve System, November). Return to text
3. A discussion of the effect of these statements on interest rates is in Ben S. Bernanke, Vincent R. Reinhart, and Brian P. Sack (2004), "Monetary Policy Alternatives at the Zero Bound: An Empirical Assessment," Finance and Economics Discussion Series 2004-48 (Washington: Board of Governors of the Federal Reserve System, September). Return to text
5. The Committee unanimously decided on December 14 to expedite the release of the minutes of each of its regularly scheduled meetings by issuing them three weeks after the date of the policy decision. The new schedule began with the release, on January 4, 2005, at 2 p.m. EST, of the minutes of the December 14, 2004, meeting. The previous practice had been to release the minutes of a regularly scheduled meeting on the Thursday following the subsequent regularly scheduled meeting. Return to text