Transcripts of the monetary policymaking body of the Federal Reserve from 2002–2008.

Why don’t we reconvene? We had some hurried consultation during the break about how best to proceed with this discussion. Breaking with our usual practice, we decided that it might be best for me to begin and to provide you with my sense of a broad schematic of how I see us going forward with our communications, in particular with the projections, which I view as being central to our plan. I note that people can justifiably complain that they haven’t had enough preliminary information for this discussion, which I’m about to give you. But let me also say, first, that there will be, of course, opportunity to react in the round to follow and, second, that we will continue to poll you and to consult with you and that we’re not going to be finalizing this for some time yet. So in the interest of trying to give you something to react to, let me just present an overview of how I see this going.

The extended projections that we have been experimenting with can be a central critical element of a new and expanded communication strategy that will, in fact, address many of the concerns that were raised today. I’ll be very explicit about that as I go through this. In particular, what are the elements of these expanded projections? First, I recommend that we extend the projection horizon. We have used a third year in our experiment so far. I think that might be the right solution, but let me just leave open for discussion the possibility of either using a fourth year or replacing the third year with a third through fifth year average or something of that sort—sort of a long-term average. Second, I recommend that we release this projection quarterly. That would mean approximately quarterly because the calendar is not conducive to an exact quarterly release schedule. That would mean that two of the projections would appear in the context of the Monetary Policy Report. This is not a central concern, but my thought at this point would be that we should revisit the Monetary Policy Report and try to make it more informative. There is a lot that we could do to make it better—to include more information, include boxes, and essentially make it a more-effective publication. That would be, as now, twice a year. Perhaps on the off quarters we could have a small release of some sort, or we could include the projections in the minutes. That’s one of the issues that we need to discuss as we go forward.

Third, I think that in our projections we ought to project a total inflation measure. As many people have noted, there is confusion—and even some resentment, I would say— about what appears to be our excessive attention to core inflation. Projecting a total inflation measure throughout the projection would, first of all, clarify that our definition of price stability is in terms of total inflation and would give us opportunities to explain in this document, in speeches, and elsewhere not only why we do look at core inflation, how we use core inflation, and its role as a forecasting mechanism but also—as President Fisher and others have mentioned—that it’s not a necessarily sufficient statistic and that we should look at other things as we try to forecast overall inflation.

I would go even further here—just, again, to be concrete. I’ve wavered myself on thinking about which particular measure. I’ve had discussions with the staff, and I am not wedded to anything in particular at this point, but currently I am leaning toward suggesting that we use the PCE deflator as our measure of inflation. It’s a technically better measure. It uses chain weighting and has a lower weight on shelter costs, which have their obvious problems. It has disadvantages: There is a significant nonmarket component, for example. An argument can be made for the CPI on the grounds that it is better known although, as has been pointed out to me, if we begin to really focus on the PCE, it may endogenously become better known. Another point to make here is—and I don’t want to get too distracted with this—that since there is a fairly stable wedge between the PCE and the CPI, we might be able to use the CPI in some of our communication as long as we’re clear that we are not picking and choosing as far as our objective and our definition of price stability are concerned.

Another component of the projections that we have been doing, which I think is very valuable, is adding a considerable amount of both qualitative and quantitative information to our projection. That includes, in particular, our explanation of the qualitative material that we have been submitting—describing the forecast, the risk to the forecast, the sense of uncertainty that we have, and so on—and I think that will be very valuable. As I said, we can combine this with other supporting information. There are various ways to do this. There could be a separate document four times a year. It could be in the minutes. I suggested one possibility, which is to include it in the Monetary Policy Report twice a year and keep it in the minutes or as a separate document on the off quarters.

Let me talk a bit about what this would accomplish for us, and then I will summarize that at the end. First of all, the public is very hungry for information about the Federal Reserve’s outlook and our sense of the risks to the economy. We have, as I have noted before, the best forecasting group in the world. We have useful information to add to the debate. By providing that information, I think we can help people make better decisions and understand policy and the economy better. In particular, the more we are forced to explain our predictions and our forecasts, the more credible we’ll be and we’ll be inviting discussion, reaction, and debate that will, I think, make our projections better. One of the advantages of transparency is that we begin to interact more with the outside world.

Second, I think we should assume optimal monetary policy. I had some other ideas before, but in the end, I think that’s the right thing to do. An important implication of assuming optimal monetary policy is that the projections therefore become essentially, as everyone understands, a plan for how we propose to steer the economy, if you will— subject, of course, to all the qualifications of uncertainty, forecast problems, and so on. It gives an explicit road map with reference to both sides of our mandate about how we expect our policies to move the economy toward our objectives over the next three years. I think that’s very important for a number of reasons, such as accountability and transparency. But let me give you an example of where I think it’s particularly useful.

One issue we have been discussing is the appropriate period for achieving price stability, and two suggestions are out there. One is sort of the standard Bank of England approach, which says that we have a two-year horizon. We have certain concerns about that. In particular, it doesn’t necessarily take into account, at least not explicitly, the state of the real economy, the initial conditions, how far we are from price stability, and so on. Another possibility is to say, well, it’s just an aspirational number. It’s a long-term number. We don’t have any particular schedule for getting there. People have raised the obvious objection: Where’s the discipline? Where’s the credibility associated with that? So, as we understand this, we can explain to the public that the projections go a long way toward solving this problem because they show how far out we think we have to go to get to what we and the public view as being reasonable levels of price stability. So it does in a very important way solve the problem of the appropriate horizon.

Now, I should add a point that will come up, which is that one could object that the projections are not the same as a Committee forecast. We are not going to come together and make a single forecast that the entire Committee buys into, except to the extent that we do have consensus building, which we will have over time in our meetings. That aspect of it could be viewed as a lack of clarity. However, the aggregation process does reflect unique features of the Federal Reserve, including its institutional structure, the large size of Committee, the geographical dispersion of its membership, and our longstanding willingness to accept and encourage diverse views within the Committee. So we won’t be forcing some kind of artificial consensus. There will be opportunity for disparity. In particular, I would recommend that we provide information to the public about the cross- sectional distribution, as we already do. But my inclination—and people can react—would be to provide the entire cross-sectional distribution to convey the sense of uncertainty or the sense of dispersion of views, and that will be informative in the same way that the votes in the BOE’s Monetary Policy Committee are informative. However, as in the case of the Bank of England and other banks, even though we won’t be having a common forecast, nevertheless—as I think we have already seen— the preparation of our individual forecasts does create a certain amount of discipline and has been useful for us in thinking about our forecast.

A very important question is what is conveyed by the third-year projections, and I think that they are at the heart of the innovation created by this step. Assuming that we’re not too far from the steady state initially, which I think characterizes our current situation, it is evident that the third-year projections—or, alternatively, the third through fifth or however we decide to do it—reveal a lot of information about our views on sustainable long-run growth; our views on sustainable unemployment; and, of course, our views of what price stability is. I simply take note of the fact that the latest projections show the central tendency of the Committee’s inflation objectives to be 1.5 to 2 percent on the core PCE deflator. I actually—and I’m speaking entirely for myself—would be not at all displeased if that became known as the Federal Reserve’s comfort zone or informal definition of price stability. First, it’s a compromise among different views. Second, I realize that I’m complicit in the 1 to 2 percent comfort zone, but I do note that the lowest twelve-month core PCE, in 2003, was 1.27. I don’t think we’d be comfortable with inflation rates below 1 percent even though we’re obviously willing to tolerate inflation rates slightly above 2. It’s not symmetrical, and I think that the comfort zone revealed by our third-year forecast would be reasonable and would provide useful information.

Some would be concerned that we’re also providing information about the Committee’s views on sustainable growth and sustainable unemployment. I am not that concerned about it. I think that a transparent Committee should do that. However, for those who are concerned about possible risks, I’ll point out that Committee projections will have a lot of dispersion that probably will essentially encompass most reasonable estimates of these variables. Moreover, there will be forecast errors around those projections, and as we get information about productivity and other factors, we will be able to update those estimates. I don’t think that they will be a major constraint. I think that they will, in fact, just provide some information to the public.

I note that some participants have talked about an opportunistic approach to disinflation, which still seems to have some adherents around the table—that is, people who may say, “Well, I’m sort of for 2 percent now, but I can see over time, if the opportunity arises, very gradually moving down to 1½ and so forth.” Obviously, the revealed preference shown by the third-year projection doesn’t distinguish between those things. One thing that might happen—and I don’t think it’s necessarily a bad thing—is that, if opportunistic disinflation happens and those individuals therefore lower their projections, the Committee’s projection might drift down a bit. I don’t think it would drift up. I think there would be a strong resistance to that. But it would be responding appropriately to changing conditions, and I don’t think it would change very much; so the fact that it would not be literally rigid is not necessarily a problem to me.

Some thought will have to be given to vocabulary, how we describe these things. I think that I will myself want to talk about the Committee’s projections. It’s not quite right to call them a target because we will not have agreed universally and chosen a number once and for all. That being said, I think that the normative implications of the number will not be missed by the public, and that it will do a lot to clarify people’s views on what the Committee is trying to do and will also be useful for internal discussions. In particular, in testimony, speeches, and the like, I would use the projections as my reference point as I talk about what the Committee is trying to do, where we are heading, and what we think is the best way to get there.

So let me summarize what I think this component of our communication strategy could do for us, noting that some other things could come in the package—perhaps faster minutes, changes in the statements, and so on. But let me just talk about what I think is best for us. First, it is going to allow us to provide considerably more information to the public and in a more timely way because it will be quarterly. Second, it will give us an opportunity to clarify how quickly we intend to move toward our objectives. It will give us a way to deal with the problem of the horizon. Those of you who were at the St. Louis Fed’s conference on inflation targeting in 2003 might remember a paper by Board staff members Jon Faust and Dale Henderson in which they talked about inflation targeting as being focused on the mean of the objective but needing also to focus on the variance. What they mean by that is the speed with which misses are fixed and that some regimes are not adequately constructed to deal with that. Again, this would allow us to be very explicit and very accountable about how we return to our objectives.

Third, it would allow us, again, to move to total inflation as our objective and to clarify how we use core inflation as an input into forecasting total inflation. Fourth, it would clear up the current confusion about comfort zones and individual views. We’re not going to take a vote on the particular number, but I think a reasonable way to proceed is to consider the range of third-year projections as being a kind of consensus view of what most of us think is an appropriate measure of price stability. Again, I think the normative implications will be clear. Fifth, it will improve our internal discussion and decisionmaking. We’re already seeing that. In fact, it has highlighted some problems that we have in our communications and in our coherence. This will help, I think. Sixth, it respects the Fed’s unique institutional structure, the nature of its Committee, its governance procedures, and its attention to diversity of views. Moreover, it builds explicitly on a communication device that we’ve been using for thirty years. In that respect, it will look in some sense as incremental even though I think it’s very substantive. But I think with its incremental nature, the transition risks—be they market risks, political risks, and so on—will be more moderate than they would otherwise possibly be.

So this is an outline. I know that some of you may think that this is a bridge too far. Some of you will think that I’ve only started on the road to Damascus here. [Laughter] As I said, there will be an opportunity now to go around as you talk to give any reaction you might have to this. Also, I have not been very specific about the questions raised in exhibit 1 that Vince put out. These things bear on the details, but I’m pretty flexible about them. I think I’d like to see the minutes moved up if possible, consistent with doing all of what I’ve described. Otherwise I’m pretty flexible about this.

Let me just say a few other things about going forward. This, together with the other elements, is part of an ongoing process. We’re not going to lock this down. I will ask the subcommittee to continue in existence. Certainly we’re going to have to move forward, if we all agree that this is the direction that we want to go, to implement the details of the minutes, the Monetary Policy Report, the collection of this information, and so forth.

I would like to talk about the whole package, whatever we come to, in the fall—that is, in the next three or four months. I will surprise Don by saying this. It would be very nice if I could have something to point to in October, but it’s certainly possible and I think it’s very important for us to move deliberately. If that’s not possible to do in a safe and clean way, then we could wait until January to actually deliver some product. So that is a question. I think that what will happen is that we will see the reaction we get. We will see how the public and the markets respond. We may have to take further steps. When we see how this goes, the markets and the public will tell us what they need to know that we’re not yet telling them, and then we can move it still further if we need to. But my sense is that, with some of the details that we need to work out—and your comments are more than welcome, as they are with the entire vision—as a central part of a package, this would move us in a good direction. So I’m going to stop there, and we’ll have a go-round. Did you have an intervention?

Keyboard shortcuts

j previous speech k next speech