But we did have a problem before that, and that is exactly my point. In fact, not only might you not get the right person in the future—and that could have happened this time around, but we were lucky in that regard—but even when you have the right person, the transition to a new Chairman has been very complicated. In particular, people might not remember this, but when Greenspan came in, there was an inflation scare in the bond markets. Long bond rates went up about 200 basis points, and the bond market tanked. I think that there was also some confusion in the markets when our current Chairman took this office about whether he was a hawk or a dove on controlling inflation. Luckily, I think that confusion has dissipated, but I think it actually made your life, Ben, a little more complicated than you would have chosen it to be. So the way I think about this is that I would like the transition to a new Chairman to be very boring, and having an explicit numerical inflation goal for the long run would help in that regard.
Clearly, there are also the standard advantages that having an explicit long-run numerical inflation goal would clarify communication with the public, the markets, and the politicians. It increases transparency and accountability. What is not discussed as much—but I think Bill Poole hit on it—is that it could clarify and make more coherent discussions inside the FOMC. I would really like to know, when somebody is advocating a certain setting of the federal funds rate, what his or her inflation goal is. Jeff Lacker has dissented in this particular meeting. I have an idea why you did because I think you have been clear on what your goal is. But if I were confused about that, I would have a harder time understanding why you’re taking the position that you’re taking. We have seen evidence in looking at how central banks operate that having an explicit goal has actually improved the nature of the deliberations in these committees.
Another thing that is very important is that having an explicit numerical goal would allow us to think about the way of doing monetary policy, which is becoming better understood in the academic literature on optimal policy. The way we think about doing optimal policy is looking at what is an appropriate inflation path. In particular, having a long-run goal that you know you’re heading to and then talking about the path to that goal is exactly the right way to think about policy.
Let me tell you where I would like to head ultimately and then where I think the Committee should head in the short run. There’s a slight difference because of some considerations I’m going to bring up. I’d like to see in the long run exactly what Bill Poole and Jeff Lacker mentioned, which is the FOMC’s agreeing on an explicit long-run numerical inflation goal and doing it in terms of saying that we are defining what we mean by price stability. So I am in full agreement with you that we should ultimately head there. Then the discussion of monetary policy, both inside and outside, should focus on what the appropriate path is to get there and how long it should take. This approach is consistent with what we have learned in terms of modern monetary policy analysis. It is the way we have learned to think about this and, by the way, is consistent with the kind of policies for the most part that have been pursued by the Federal Reserve over the past ten years. A key part of the success of the Federal Reserve’s operations is that, in practice, we have actually been operating in that way.
When you have such an approach, there are four basic principles regarding the way you think about changing the path that you think is appropriate in terms of achieving the long-run goal. First of all, we ask how far away we are currently from the long-run goal—so where we are really matters. Second, the nature of the shocks, whether they’re transitory or permanent, is clearly important. Third, it is very important to be aware of how big the projected output gaps are because we should care about output fluctuations, something I will turn to in a second. Finally, we also want to be very concerned about financial stability issues because, if something is occurring on the financial stability front, we have to be able to deal with it. This kind of approach would allow us to have a concern about business-cycle fluctuations, which I think is key for a central bank. It would also allow us to change policy in response to financial instability concerns. It should allow us to deal with the LTCM type of situation, in which there was a potential small probability of very disastrous things happening. In fact, it should be consistent with what Greenspan called the risk-management approach. So I think it is very important to understand that we’re not talking about going to a rigid form of inflation targeting if we pursue this goal.
Now let me turn to the second issue on the decision tree, which is how we express an explicit nominal definition of price stability. I want to discuss some subtleties that really do concern me right now, which means that I’m going to shift slightly from where I want to get in the long run to where I think we should head in the intermediate term. Four critical considerations worry me.
The first is the political process. It is very important that we do not get too far ahead of the Congress on this. It is extremely important that we express a further definition of what we mean by price stability in a way that is absolutely consistent with the Federal Reserve Act and with the law and with the dual mandate. This consideration became very clear to me in looking at what happened during the testimony that the Chairman gave when he took his position. It was clearly a central issue in the testimony that I gave. It’s very clear that this process is part of the American system and that it differs from many other countries. So that’s consideration one.
Consideration two is that it’s extremely important that we not be considered to be what Mervyn King artfully has called “inflation nutters.” The British have better training in English than we have—they always have better phrases for things. But we need to make very clear that, in fact, we are not inflation fanatics. A key part of the success of the Federal Reserve has been based on the fact that the Federal Reserve is clearly perceived by the public, the politicians, and the markets as having a weight on output fluctuations in our decisions. What is also an important part of our success is that, in contrast to many other central banks, we have been willing to express our concerns about output fluctuations and the business cycle. So in my past writings, I have talked about the dirty little secret of central banking, which is that most central banks are not willing to admit that they care about output fluctuations. I think this is actually a serious problem for them. Luckily, we don’t have that problem, but if we move toward an explicit nominal long-run inflation goal, we need to be clear that we differ in that regard from other central banks. The issue here is that this is a primary reason that we’ve had the support of the public and the politicians. We certainly don’t want to lose that. That’s the second consideration.
The third consideration is that we differ from other countries in a major way, and when I look at the Riksbank versus the Federal Reserve, I see how different we are. We have nineteen participants making decisions about monetary policy, and that’s extremely unusual. The only other comparable situation is the European system of central banks. Our situation is different not only because there are nineteen people but also because the participants reside in different locations. Now there are six but there will be seven of us existing in house, and there are twelve of you who factor in different locations and have a very different role. You are CEOs of an organization, and you have to communicate differently with the public than the Governors do. I really understand this because I’ve been on both sides, having been at a Reserve Bank and also here. That situation actually makes things very, very different.
The fourth consideration is that the Federal Reserve has been extremely successful. That gives people confidence in us and tells us that the more evolutionary we can be the better off we are. This consideration is important in two senses. One is that we want to say that we have done a good job in the past, and we are going to continue to do a good job in the future. Thus we build on the credibility that we’ve established over the past fifteen years or so. The second sense is political: We really want to promote political support, and looking as though we’re doing something that is departing from where we were is not a good way to move forward.
Now that I have given you basic principles, let me put forth a proposal—not one to be voted on today but where I would like to see us head. Although in the long run I would like the FOMC to be able to sit down and agree on what our long-run inflation goals should be in terms of a specific number, I would argue that the four considerations I brought up suggest that we need to go a bit more slowly to that path. So here is what I would suggest. As part of our reporting procedures that we do for our Monetary Policy Report twice a year, we produce forecasts. In a similar vein, I think it would be very effective for all nineteen participants, when they are giving forecasts, to say what their definitions of price stability are and, in fact, provide that number. But I would recommend that this be done only once a year, not twice a year.
What are the advantages of doing this? One advantage is that I think we are fairly close or will become fairly close to having agreement on it. I would argue that the numbers will not be that different among us, and providing our numbers will make it easier for us to come to consensus in the future. So it is an intermediate step to get us there in the long run. Another advantage has to do with the fact that we have nineteen participants. I think it’s very important that all participants are heard from and that we don’t encroach on their particular views until we can actually build a consensus. So it really fits in with the issue that there are nineteen people in different locations with different jobs to do. Also, it’s evolutionary; it really is not a radical departure. It’s very much in the spirit of saying that we have a dual mandate. We’ve actually had individuals talking about comfort zones and so forth. We’re just formalizing the process, and we’re doing it in a way that is a step forward in transparency, so it really moves us very much in the direction we have been going. The bottom line here is that it gets us close to having an agreement. I think that is not too far away.
I do want to mention the once-a-year issue, which would come up if the FOMC actually votes on this. I think people worry about that very much because—and I agree strongly with President Poole and President Lacker—we do not want to have the number change very much over time. (We could also talk about some tricky issues that are created by bad inflation expectations dynamics, but not now because they are more complicated.) In practice, we have actually seen something that gives us a lot of information about this issue because one of the concerns that people had when inflation targeting started was that it would be decided by governments. The problem with governments is that we know what politicians are like—they have to get elected, and they think short term; this is one reason that we like to insulate the central bank from the political process. There was then a concern that, if the government provides a number, it will do so opportunistically. We have not found that to happen, and the reason it has not happened is that providing a number is so transparent. When you actually have to give a number, you have to have a darned good reason for changing it. In fact, I think it would be very unlikely that we would find this Committee, even if we did this on an annual basis, wanting to change it. Transparency really has huge benefits. It’s remarkable that, when there have been such changes, they have clearly been for technical reasons and have been in the right direction. In one case, the Bank of England changed the number because of a change in the calculation of their price index. It was done in the case of the Reserve Bank of New Zealand, with some resistance from the Reserve Bank, but they agreed that the range they went to—eventually to 1 to 3 from zero to 2—was a better number for them. Thus, I think that concern is really not going to be a problem even if we do this once a year.
I have an aside that I wasn’t going to mention, but I think it’s important to mention. It has to do with the issue of potential GDP and potential employment or a discussion of the NAIRU, and it relates a bit to President Moskow’s suggestion, which I do not think is the right way to go. If we provide information about our explicit numerical inflation goal, it’s very important to explain why doing so should not drive us to provide a similar definition of potential employment, potential output, or the NAIRU for several reasons. The first is that it’s extremely hard to know what the correct values are for those numbers, and they’re extremely hard to estimate in real time, as we know. I don’t even know what they are conceptually. This is a big issue in the academic literature. We have a good example of this issue right now, which is that the Greenbook has a very different view on this from what we find in the private markets—you see much slower growth of potential GDP than the markets do. Any suggestion that we have something like an employment goal would be very, very damaging to us. We also know that, when the Federal Reserve has acted this way, which they did in the ’70s, it led to disastrous outcomes; Burns focused too much on output gaps, and it got the Fed into a lot of trouble. This finding comes from work done by your staff—in particular, Athanasios Orphanides.
In contrast, it is important to clarify that the reason for the big difference between defining price stability and defining employment goals is that we know from a welfare viewpoint it doesn’t matter much whether we are off by 50 or even 100 basis points. I think this is the point that Bill made. He likes zero, taking into account measurement error, but if we said it was 100 basis points more than that, I don’t think it would be that big a deal. We know that what is really important is to have a number to pin down inflation expectations, and not the specific number. That is absolutely not true of an employment goal. So it is very important to get that across, and it is the reason I reacted to President Plosser’s comments about this issue of talking about potential, although I think that he agrees with me.
It is also one reason that I have a concern with the proposal that President Moskow has made. First of all, I don’t think it is explicit enough. But I also have the problem that, if you talk about a five-year inflation forecast as a goal, people start to interpret that as a goal, and if you talk about a five-year GDP forecast, they’ll also interpret that as a goal. I think it leads exactly to a lack of clarity that will get us into trouble. So we need to be much clearer on this. If you ask me my five-year inflation forecast, that might not be the same thing as my view of the appropriate level of price stability because I have to think about what everybody else’s decisions are going to be, who may be the next Chairman, or whatever.
Another issue, which is not the main subject today, is important to discuss because it may affect what the staff has to do. I have another proposal, which is completely independent of our decision to go toward an explicit inflation goal; it is that markets would tremendously benefit from more information about our forecasts, which are now twice a year, and to have that information four times a year. By the way, this proposal came out very strongly in the Meyer survey, which I’m sure you’re all familiar with. They were quite in favor of it. Again, this could be done in an evolutionary fashion. We have a reporting procedure on monetary policy, which we are mandated to carry out twice a year. By saying that we’re willing to do it four times a year, we are actually saying that, in the spirit of cooperating with the Congress, we are trying to be even more helpful. So I think it would fit in very naturally. What it would require, of course, is a little more work on the part of all of us because we’d have to do it four times a year rather than twice a year. But I’m willing to do that, and I think other people would be willing to do it as well.
Another key issue is that we need to greatly improve the quality of the written documents that go with this process. The current Monetary Policy Report is really terrible. It’s dull; it’s sex made boring. I don’t want to criticize too much, but it is. [Laughter]