Richard H. Clarida, vice chair of the Board of Governors of the Federal Reserve System, made a presentation on the US economic outlook and monetary policy followed by a discussion with PIIE president Adam Posen.
Adam Posen (AP): Thank you very much, Rich. Obviously, you're making a lot of strong, clear statements in your speech. I want to start our discussion, which is on the record, obviously, in a different place. For me and to many people, the biggest strangeness of recent months has been the continuing decline of US long bond rates. The 10-year going down to below 1.2. The FOMC's own estimates of R-star and neutral rate are well below 2 and going down. The 5-year, 5-year real forward is negative. Notwithstanding whatever market reaction to your remarks today, this is a very striking trend when there seem to be many things pushing up inflation and when there is a prospect of the Fed raising rates. What is going on? Is it secular stagnation is worse than we thought? Is it something technical? What does it matter that it's global and it doesn't seem to be Fed specific? How do you think about the long bond and R-star estimates at this point?
Richard Clarida (RC): Well, Adam, not surprisingly, you teed that up very well, because I think each of those is an element. Now, let me say at the outset, I have been surprised certainly by the magnitude of the decline in bond yields. Interestingly enough, it doesn't appear to be driven by a big shift downward in expected inflation. That's noisy. But those measures, whether or not you look at TIPS or surveys or other indicators, have not changed all that much. As you indicated correctly, it has been a move in real yields and it's been a global move in real yields. And the move is quite pronounced. For example, in March, the 5-year, 5-year real TIPS yield was about 0.6 percent positive and now it's about -0.4 percent negative. That's a big move in real yields. I think it is global. I think, in part, it probably does reflect some concerns about the virus that are maybe more prevalent now than they would have been in the spring. I'm watching it closely. Forecasting is hard. Usually 20/20 hindsight is pretty accurate. But even with this case, I'm still analyzing it. I think the decline in real rates, obviously, secular stagnation was a very popular and persuasive theory for a number of years. I think it's too soon for me to embrace that as the sole explanation. But there are a lot of things going on and I'm looking at them closely.
AP: Thank you. Bridging that discussion back to some of the things you addressed in your speech, let's talk for a moment about flexible average inflation targeting, the new framework that Chair Powell and you and others advanced a year ago with the support of the committee. One issue with flexible average inflation targeting is the credibility of the commitment to make up for past undershoots. In your speech, I haven't done the arithmetic carefully, but given your remarks about what on core PCE in the next two years would constitute enough or moderate overshooting in your definition, it does seem to be asymmetric that you're not fully making up the past undershoots. Now, is this because I'm misunderstanding the timing and the measurement? Is this because just this is your opinion? How should we think about the credibility, the commitment to the sort of symmetric average aspect of flexibility?
RC: Yeah. Let me be clear. I've given a number of remarks, including at Peterson, and what I've emphasized since we rolled out the framework is that flexible average inflation targeting is a regime that has an ex ante aspiration for inflation to average 2 percent over time. We had extensive staff briefings and we were concluded that for a variety of reasons, trying to make up for any particular backward calculation window of inflation is not really a robust strategy. We put much more weight on forward looking measures of inflation expectations. Really, all of this, Adam, is in service to keeping inflation expectations well anchored. And the concrete implementation of flexible average inflation targeting is really the rate guidance that we offered beginning in September of 2020. We gave conditions for liftoff. We have essentially said that in order to help us achieve inflation that averages 2 percent over time, we want to delay liftoff until we actually reach 2 percent inflation. Not in anticipation that we would. And also that when we're at least in the zip code of maximum employment, obviously, we don't measure it precisely, but we want to have confidence that we're close to that. And we believe, based on analysis that the staff has done, including some excellent work by John Roberts and other staff have done, that a regime that delays liftoff until those conditions are met is a regime that at least ex ante can deliver inflation that averages 2 percent over time. That's how I'm thinking about the framework and that's how I'm thinking about the current set of circumstances.
AP: Great. But isn't there a risk that by not seeming to be sufficiently symmetric, you might be reinforcing some of these long term low rate expectations? I mean, again, I'm not trying to belabor this. But we all know how it was for Japan to try and fail to get out. Why are you confident this is enough given where rates are?
RC: Well, first of all, this is a regime that is going to be monitored subject to data. And I, myself, have really emphasized -- for me, the reality check is closely monitoring wide range of measures of inflation expectations. The Fed staff has an excellent version of that. There are other ways to do it. I think that there's a pragmatic and practical matter, Adam. You don't want to put all your eggs in either looking at models or surveys or markets. And so the nice thing about the Fed staff index is that it's essentially a common factor principal components model. And so I take comfort that those indicators and others done not just by the Fed staff show that inflation expectations. I think other countries got into problems because they did not begin to address potentially that risk until inflation expectations had already drifted down. And we're not in that situation now. And so I think that's one big difference, perhaps with the approach taken in other jurisdictions.
AP: Thank you. A key word in your speech and many discussions recently in the Chair's recent remarks to Congress is transitory as in transitory inflation or transitory shocks. Now, transitory is different from overshooting. Overshooting can be accidental or intentional. How should we think about the transitory nature of inflation? Let me be a little more specific. How are you and the Fed staff discerning what constitutes transitory and what doesn't? I mean, is it fair to say that given the nature of the supply shock, the unprecedented nature of the recovery you spoke about, that we can't discern what's transitory and what's not? Or is that too defeatist?
RC: I think that's too defeatist. I think doing so with econometric precision is difficult. But I do think that it's useful to put price increases into perhaps two, maybe three different categories. One of these are what I would call the round trip rebound effect. Hotel rooms declined by 20 percent. This year, they go up by 25 percent. Over the entire 16 months since COVID, not much of a change. There are a lot of sectors that have done a lot of round trips. That's why in my remarks, I referred to inflation since February of 2020. The nice thing about doing that calculation is that you don't have to get into a position of which item you're throwing out of the calculation. You look at everything, but you just set the base year before the pandemic. And if you do that, then on the PCE index, then inflation on a PCE basis since February of 2020, I think in my remarks I said it's 2.7. It's somewhat higher on CPI. I think the other category that you want to look at, Adam, is the reality that last year, not only did the US, but the global economy shut down. And this year not only the US, but the global economy is opening up. That opening up is putting pressure, especially in commodity markets and in a lot of tradable intermediate inputs. And as those relative prices are adjusting up, it's showing up as inflation. But there's no reason to think that those relative price changes are going to go on indefinitely. That's another category. What I would put is global reopening pressure on commodities. The third category is probably the most important and the one that right now I have the most uncertainty about, and that's the state of the labor market. If you look at unemployment or you look at the shortfall in participation, then this looks like a labor market in 2012 or 2013. If you look at vacancies, if you look at quits, if you look at reservation wages, this looks much more like a labor market like 2017 or 2018. And that obviously makes a big difference for how you think about the economy. I've said very many times -- I'll say it again, I think we're going to know more about the labor market over the next several months than we do right now. I think there are some factors that plausibly are holding back effective labor supply, whether or not that is child-care responsibilities for children who are doing remote learning, whether or not it's concern about the virus in contact intensive sectors, vaccination and obviously unemployment benefits. Supplemental unemployment benefits are also expiring in the fall. I and many of my colleagues have indicated that in that third category, we're going to learn a lot more about the labor market. My baseline view -- and always baseline and risk -- but my baseline view is that we'll begin to see in the fall some pretty healthy increases in effective labor supply and employment gains. And that's certainly in line with outside forecasters as well. But there is a risk to that outlook and there's a downside risk to that outlook. And we and my colleagues would have to study that very carefully if that plays out. That's how I'm thinking about it.
AP: Great. Let's, if we could stay on the labor market for a bit. I mean, as the new framework, I think, in my view rightly, has put more emphasis on the set of labor market indicators and more emphasis, frankly, on the welfare benefits of full employment. And I, for one, am very big fan of the Fed having done that. But just going back to your speech, I guess, because it's what's in the SEP, you mentioned your own modal forecast or the modal forecast, the median modal of the employment forecasts for 2022 and 2023. And you have a very benign forecast of a very slow decrease in the mid threes over 2022 and 2023. What are you thinking about in terms of labor force participation to go with that unemployment outcome? And how much is labor force participation getting back to or above pre-pandemic levels important versus a narrow unemployment measure?
RC: Yeah. Excellent question. And actually, I think there are two parts. I will answer both. We have said in our statement with unanimous support, I'm very proud that we were able to achieve that a year ago in our new framework, that we recognize that maximum employment is a broad based and inclusive goal. Let me say what that means to me. What it means to me is that the historical record has shown going back decades that when the labor market improves, the gains across society are uneven. Typically, it's only in what macro economists call the mid or late cycles of past recoveries that you begin to get broad based gains in terms of participation, employment, wage gains at the lower end of the wage distribution. We realize, as policymakers, how important it is to get to a low unemployment rate, one consistent with price stability, as fast as we can and keep the economy there. One thing I mentioned in the speech is that historically and again, only historically -- we don't have a crystal ball. Historically, that when the unemployment rate gets down into the range that I talked about in the SEP estimates, other indicators of the labor market also historically also tend to benefit as well. I think it is possible to have a broad based and inclusive goal. But in terms of the outlook, use the unemployment rate as a way to focus on getting in the vicinity of where that would be. Obviously, at the time, the committee would be looking at a number of indicators. That's the first point. Your second point about participation is a good one indeed. I went back at them and reread my first ever speech as vice chair, which was at the Peterson Institute in October of 2018. And I made the point in that speech that at that point in 2018 we were beginning to see gains in prime age participation that quite frankly had been disappointing and very, very underwhelming for basically from 07 till about 2015 or 16 even prime age participation. Correcting for demographics have been declining. And that began to turn around. I made the point, I think maybe in the Q&A, that there could be more to run on that and indeed there was. And indeed, participation in the last cycle went above many estimates of what the underlying demographic trend was. And so my own personal forecast is I believe that we will certainly be able, under my baseline outlook, for participation to move up certainly to and perhaps somewhat above the demographic trend. And so I think we will get a pickup in participation. The challenge, of course -- and it's relevant for some of us -- is that demographics is destiny. There's nothing we can do about it. The US population is getting older. More people get to retirement age. And so for participation to actually go up much beyond the demographic trend is going to require some additional momentum. And I'm certainly hopeful we can see that. But we'll have to wait till we get closer to that to make that judgment.
AP: Rich, you and others on the committee, again, rightly in my opinion, but my opinion doesn't matter, have been speaking about the benefits of an economy run hot and full employment for inequality in our society, for reducing wage differentials and unemployment differentials across racial, gender and other biases. How does that turn into operational concerns as you're making these decisions about the labor market over the next few months? Is it that you still remain focused solely on the aggregate number? And if you get progress on those things, great. Or is it that that's part of your assessing how tight is the labor market?
RC: I think that from my perspective, Adam, that I think of this in the context that we do have a dual mandate. And so I think of maximum employment as the maximum sustainable level of employment that is consistent with our objective, which is to average 2 percent inflation over time. If we have a level of employment that is so high and a level of unemployment that is so low that it begins to move up inflation and inflation expectations, that would be counter to that leg of the dual mandate, then I think that would indicate to me that from my perspective, we've reached maximum employment. I think you have to look at both maximum employment and price stability in tandem.
AP: And we're talking a moment ago. We're getting into the lightning round and to a big question. But you were talking about a moment ago, Rich, about some the long term factors like demographics affecting the US labor force. But one of the things that we historians have pointed out and technological people have pointed out about the current situation is there is a possibility for substitution of capital for labor, depending on wage developments, depending on response to the pandemic. How much do you view this medium term threat of greater digitalization or automation affecting labor supply, or do you see that not happening in the extent that you worry about?
RC: Adam, I would put me in the camp of believing two things very strongly. I do think there will be long term secular implications of COVID, including along the dimensions that you mentioned. But I also right now have no confidence in my ability to identify what they will be and what they will mean for our outlook. Perhaps I'm somewhere in the camp that Bob Solow was 30 years ago, that we see computers everywhere except in the productivity statistics. It took a while and then it happened. And that may well be true. But as a policymaker, I can't have any confidence making policy on a forecast of an inflection point in technology or capital labor substitution. Certainly something I'm following closely. I will say that on a secular basis, looking at 10, 15, 20 years, we are going to have an aging society. I would imagine that would be part of the way that the economy evolves. But in terms of the implications near-term for policy, there are certainly not anything I'm seeing now that I would be factoring into that outlook. But again, in the broad category of wanting to understand better how the post pandemic labor market is going to clear, that would certainly be an element of that as well.
AP: Thank you. Again, switching back now from long term. In your speech, if I understood you correctly, you said in your opinion, if your modal forecasts and roughly speaking, the committee's median modal forecast turns out to be right, you could see tapering at the end of this calendar year, December 2021. I hope I didn't misread that. Just more broadly, rather than your specific call, what is going into the timing decision of when you'll taper? Once you start announcing to taper, how long do you think it takes to do it? How much lag between tapering and a rate hike? Once you taper, are you automatically moving to a rate hike? If it's not time dependent, what is the determination of when you shift from tapering to hiking rates? I hate the term normalization and I won't use it. Just a little more about what's the mechanism of getting to taper and from taper to rate hikes or not?
RC: Well, first, I'll do two things. First, I'll read the text from my speech and then I'll amplify on that. What I said in the speech, is that I and other members of the committee expect the economy will continue to move toward substantial further progress. We, obviously, did not meet that standard in July. And in coming meetings, we will continue to assess that. What I would say is that if my baseline outlook does materialize, then I could certainly see supporting announcing a reduction in the pace of our purchases later this year. Again, if the economy does not evolve as expected, then that would obviously impact that. But I could certainly see that if my baseline outlook materializes, then I could certainly see myself supporting announcing a moderation in the pace of purchases later this year.
AP: But on the mechanism, if you announce that. What does that tell us at all, if anything, about a coming rate hike or are these decisions separate?
RC: They're completely separate decisions. And I think the committee deserves credit because we laid out specific conditions for liftoff and there's a different metric for liftoff than there is a metric for reducing the pace of asset purchases. That's a substantial further progress metric. As I said in my prepared remarks, we're not thinking about hiking rates. It's certainly not on my -- I don't think on the committee's radar screen. Those are two separate two separate decisions.
AP: How much do you view that as learning a lesson from the 2013 taper tantrum? Are there any other things that the Fed or you think as a result of the experience in 2013 that you want to do differently this time?
RC: Well, obviously, I wasn't on the Fed then, although a number of my colleagues were, and I've certainly benefited from their perspective. I think what I would say on that is as we laid out, for example, in December what the metric would be – substantial further progress for the taper. And then at each FOMC meeting and in a lot of communication in between meetings, we've also laid out our views that we have not yet made this progress. I think, certainly that communications approach in that sequencing, I think, certainly drew on the experience of the last episode. And then again, our decision in September of 2020 to lay out threshold metrics for liftoff is also serving us well now because they're quite different than the metrics that we've laid out for reducing the pace of purchases. So I guess both of those might fit into that category.
AP: Thanks. One more round on the more technical side, although not the market structure, but the technical side of taper. We used to think when we were expanding QE, I say we, when I was at Bank of England and was part of the committee, about sort of matching up equivalents of QE to slugs of rate cuts. What do you think is the effect of tapering? How are you calibrating that or is it not calibrating, this is just a signal? And also, how did the new repo arrangements and the potential reverse repos, how does that play into the operation of tapering? What is this going to look like?
RC: Okay. Well, broadly, what it will look like is once we announce the plan, it's going to be very clear what the plan is. And so folks will be able to form views about that. I think broadly, economists talk about with regards to asset purchases, there's both a stock effect and a flow effect. I am the camp that think both are relevant, but that the stock effect is the more substantial effect of QE than the flow effect. Although in different conditions both can be relevant. For example, one could argue that back in December of 2008, when the Fed first embarked on QE and mortgages, that the flows were very important because of the dislocation in the mortgage market. So I think that evolves over time. I guess from my perspective, we've laid out these conditions. We will be updating the public at our meetings and in between meetings on whether or not the conditions have been satisfied. And I would like to think that by the time this is done, it more or less is going to be something that has been absorbed into market pricing. Obviously, we have we have no plan to surprise anybody with this, as Chair Powell has indicated, so I think that's an important element of this.
AP: Just with respect to the monetary policy stance and conditions more broadly, Rich. I mean, is this an indication that you all think you can put it -- in a provocative way -- fine tune your way into a soft landing that you do just enough tapering and then you do just enough rate hikes. Or I mean, generally, when you were teaching us monetary policy, that was not seen as something one could do and lags mattered a lot. How do you see this actually being useful? What's the way in which the tapering does what you're supposed to do if it's not fine tuning?
RC: Let me talk broadly about the policy stance, because that includes both the balance sheet and, obviously, the funds rate. I think that from my perspective, what's relevant. First of all, let me begin. I think there is an output gap right now. The unemployment rate where it is today is above where it will ultimately end up. We have seven million fewer people who have jobs than did 16 months ago. That's sort of a basic macroeconomic fact. What I would point out, and I think this gets lost in some of the discussion is part of the motivation for our new framework, Adam, was that having a low global level of neutral rates, which has been a fact of life now for a dozen years or more, poses real problems for conventional inflation targeting monetary policy. That's certainly true. But it's also the case if you have low neutral interest rates, you don't have a long way to go before you start approaching what they are. And so that's also a different part of the analysis, I think, as well. Soft landings are something, I think, all central bankers aspire to. I personally believe that we had a soft landing in 2019 into 2020 that would have prevailed but for the pandemic, certainly in February of 2020. We had an unemployment rate around 3-½ percent. We had core inflation at 1.9 and we had a funds rate of 1-¾ percent, I should say, in that range. And so certainly we had a soft landing in the 1990s as well. They're certainly feasible with the existing toolkit and, obviously, central bankers aspire to them. I'll just leave it at that.
AP: Thanks, Rich. I realize we're running up against the limits of the time you've generously given us. Let me get two last questions in if I may. First, housing. With all the talk through the years about spillovers of QE and low rates on asset markets, one of the clearest channels seems to be on housing prices. Obviously, housing price inflation does matter for inflation outcomes as well as for inequality. How are you taking this into account and why is the Fed or the FSOC not engaging in some kind of macroprudential response to the current housing price issues?
CLARIDA : Excellent question. One thing I would put on the table is one thing that's quite different now than it was 15 years ago is the rise in housing values is not being driven by loose credit standards or poor lending. Essentially, what's going on now is you've got powerful demographic change. You have a lot of folks who delayed entering into housing who are doing so now, we had income support through the CARES Act and other fiscal measures and obviously monetary policy, low rates. But credit standards are not being are not being relaxed. I think that's an important distinction. I'm not going to now, in part because we're running out of time, venture into macroprudential tools. Other than I can say, obviously, we know that housing is an important part of the economy. But we also recognize that a big part of what's going on with housing now is really demographics and incomes in many households. But obviously, we recognize that low interest rates also boost housing demand as well.
AP: Rich, finally, international has to be asked. You're an expert on that who commissioned my Peterson colleague Maurice Obstfeld's paper for the Fed Listens conference [RC: Great paper, yeah.] on spill backs on the Fed. Could you say a word about what you think the US overshooting on inflation, as its doing, and being more aggressive on fiscal policy and fiscal-monetary mix than the other G-7 economies is doing to the rest of the world and what spill backs there might be on the US?
RC: Yeah, excellent point. First and foremost, I think the direct impact of the current policy mix and impressive recovery in the US is that we're exporting some demand to the rest of the world through our widening current account deficit. As we expand faster than the rest of the world, some of that differential shows up in our imports. And so I think that's a relevant factor. I think the other thing that I'm very cognizant of that I know when I when we do either live or now virtual BIS meetings, other central bankers are aware is that different countries in different regions, in the global economy are much different places in terms of vaccinations. And that's going to lead to some disparity in recovery and the profile of global growth, hopefully not for too much longer. But that's certainly been a fact of life this year. Term premia, which reflect uncertainty about the outlook. Other countries that are tied to the dollar in their financial markets, they have a spread over dollar rates. Obviously now, for reasons we began the Q&A with, dollar rates are lower now. That in and of itself should not be a problem for most for most countries. I think, the interesting thing is it was a global shock and it will be a global recovery. There's obviously some asynchronous aspect to it. But everybody is in recovery mode now. I think that's another consequence that is worth remembering. I think I'll leave it at that.
AP: Thank you very much for your time, your frankness, your substance and for your public service. And we're grateful to have had you back once again at the Peterson Institute virtually this time for your speech today, Outlooks, Outcomes and Prospects for US Monetary Policy, which is available, of course, at the Fed website, Peterson website and all over the world.
Please join me, everyone, virtually and sincerely in thanking Richard H. Clarida, the Vice Chair of the Board of Governors of the Federal Reserve System.