Feb 27, 2017

Tim Duy on the Art of Fed Watching and the Future of U.S. Monetary Policy

David Beckworth Senior Research Fellow , Tim Duy

Hosted by David Beckworth of the Mercatus Center, Macro Musings is a new podcast which pulls back the curtain on the important macroeconomic issues of the past, present, and future.

Tim Duy is a professor of economics at the University of Oregon and a widely read “Fed Watcher.” Today, he joins the show to discuss writing on the Federal Reserve and how to interpret statements from Fed officials. In particular, Tim stresses the importance of the Fed Chair’s post-FOMC meeting press conferences. He also warns against taking too much stock in FOMC minutes. David and Tim also discuss the proper role of the Fed’s balance sheet in stabilizing monetary policy.

Read the full episode transcript

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to macromusings@mercatus.gmu.edu.

David Beckworth: Tim, thank you for coming on the show.

Tim Duy: Thank you very much for having me here, David.

Beckworth: Oh, I'm glad to have you on. Been a long-time reader of your Fed watching and your notes, both Bloomberg, as well as An Economist's View. I've had your colleague on here earlier, Mark Thoma, and I think that's how I first got to know you, as well, is through Mark, and I think we met at a blogging conference once, as well.

Duy: Correct.

Beckworth: Let's get into our conversation today. And you have been an economist in different positions, at the University of Oregon, at the U.S. Department of Treasure. You do a lot of writing now. How did you get into that field? How did you become an economist?

Duy: Well, I think as many of us did, I became interested in economics while I was an undergraduate, and ended up pursuing a Ph.D. after my undergraduate experience, and I actually did that here at the University of Oregon. And so, sometime in that process, I decided that economics was really something that I wanted to do. Subsequently, I did go to Washington, D.C. and work for three years there, first as an international economist at the Department of Treasury; and then second, working with a private consulting firm where we did, among other things, Fed watching, and I started to learn about tracking monetary policy in that position.

Duy: And then, one day, my wife and I decided that we were going to move back to Oregon... and so, we did. And I really didn't know what I was going to be doing when I got back here to Oregon; ended up doing a little bit of teaching here back at the University, then I added in a couple of other jobs at the University... basically cobbled together a full-time position here. And then, subsequently, my colleague, Mark Thoma, had started blogging and had asked faculty members if they were willing to join him, and I believe I was the only faculty member that said, "Sure, I'll do that," and then I went back to sort of what I knew, and that was writing about monetary policy.

Duy: So, that was the genesis of the Fed watch blog, and it has continued to be something I've been doing to this day.

Beckworth: That's interesting. So, purely on a voluntary basis, you joined Mark. Of course, it helped that Mark had one of the most prominent economic blogs, but that volunteering on your part kind of set you up in that position has a prominent Fed watcher.

Beckworth: Also interesting that you kind of cut your teeth on this back in D.C. You used to write for the... what was it called, the G7 Newsletter?

Duy: G7 Group, right. Correct.

Beckworth: G7 Group, okay. And what's interesting to me is, I was also at International Affairs at Treasury; I think I got there after you. And I believe we got your newsletters, among others, and so we read them. And little did I know I'd interact with you several years later in the blogosphere, Twitter... on podcasts. Who knew? So, small world in the world of economics.

Duy: It is, especially in this particular corner of it.

Beckworth: This is true. Not that many people enjoy the things that we do.

Beckworth: Okay, so, let's talk about being a Fed watcher, and for all the aspiring Fed watchers out there, this is your chance to listen into Tim Duy's perspective. Let's talk about, what do you do to collect news, stay abreast, be informed about Fed policy? Do you do... obviously newspapers, but Twitter? Blogs? And what is a typical day like?

Staying tuned in as a Fed Watcher

Duy: What is a typical day like? That's a good question. Excuse me.

Duy: Usually, I cull my information from a wide variety of sources. As you pointed out, you follow the financial journalists. You also follow the speeches that the Federal Reserve members give; that's always an important point of Fed watching. And also, understanding the data and following the data, and more importantly, understanding and following the data as you think the Federal Reserve is most likely to follow the data. It usually doesn't work very well to follow the data from a different perspective if you want to understand what the Federal Reserve is going to be doing.

Duy: And then certainly beyond that, to the extent that.... there's just basically, almost a limited number of sources, it seems like, to gain additional information, whether it be other Fed watchers on Twitter or on their blogs, or certainly academic research, as well, and certainly culling through, whenever possible, the research produced by the individual Federal Reserve banks and the Board... how this basically a never-ending flow of data to pick through.

Beckworth: It's more of a question of managing your time and-

Duy: Yeah, that's exactly what it is, is.... because you can never have enough time to read it all, is what it amounts to. It's just how quickly can you absorb the most important pieces of information is this... really usually a combination of understanding exactly what Federal Reserve participants are thinking about it and how that interacts with the data flow.

Beckworth: Now, you mentioned something interesting, that you want to do data analysis when you're a Fed watcher from the perspective of the Fed officials themselves. What are they thinking? How do they look at the data? You don't want to bring in your theory of the world into it. I think we all have our strong priors in how things should be.

Beckworth: Is that hard to do, though? I mean... so, you're looking at the data, you're like, "Aw, come on, Fed officials. Why haven't you eased more? Why haven't you tightened more?" Instead, you've got to put that to the side and be objective. I mean, how do you wrestle with that?

Duy: Yeah, that's certainly one of the more difficult challenges, I think, for Fed watchers, is... here is what I think is the right path for the Federal Reserve to pursue, but that's not necessarily the path that they're going to pursue... and making sure that you focus on differentiating between those two things. And certainly, when you write, you differentiate between those two things. This is the way a Fed's going to do it; I don't know if that's going to always be right, given my view or outlook on the economy.

Duy: But I think sometimes you'll see.. I've seen in the past maybe some people who have left the Federal Reserve system, and then will comment subsequently on Federal Reserve policy, and oftentimes it seems to me that they're commenting less about what the Fed's actually doing versus what they think the Fed should be doing, and haven't kind of learned to make that adjustment yet.

Beckworth: Yeah. It's tough. I mean, we all want to see the world run a certain way, and particularly, monetary policy, if that's our thing.

Beckworth: Now, one question I have had with all those information you just alluded to, that there's just this smorgasbord of information sources. And one of the ones that's become, I think, really important over the past few years... really, since the crisis... is the speeches, and there's been a lot of speeches by regional Fed presidents, as well as by governors, and sometimes they move in opposite directions, or they give conflicting signals.

Beckworth: And I've asked this to other guests before; I've asked Andy Levin, I've talked to Ylan Mui... different people who follow the Fed or have been involved with the Fed... and I've asked them, "Do all these speeches add value, or what does it do to the signals and noise ration?" So, I'm asking you as a Fed watcher, you just have sometimes cacophony of noise out there from different Fed officials, sometimes going against each other... do you still find it useful to have them talking as much as they do?

Duy: That's a yes and a no question. No, I... actually, I'll back up. I think that, in fact, there is a lot of use to them talking about policy and about how policy is constructed, the policy process itself. I'm always cautions that there's much use in them, particularly regional presidents, forecasting individual meetings as it seems that they often want to do. And when I say that, I think there's less use for them to give some insight or try to give some insight on what they think is going to happen with the March meeting or the June meeting, for example. I think that's where all the confusion starts to occur.

Duy: I'm not so much... I don't find the speaking by the Fed presidents themselves, the regional bank presidents themselves a problem; what I tend to find a problem is that there's insufficient speeches from the board itself, from the governors, and the chair, in particular... and I think that's where a lot of the confusion lies, because there is an immense amount of policy that's really constructed by the board and directed by the board, and we don't often hear enough of that, and that's, I think, a source of confusion about where monetary policy is headed.

Beckworth: Yeah, we had Narayana Kocherlakota on the show, and he made that same point. He'd like to see more speeches by the chair, Janet Yellen... like, maybe after every meeting, have her talk about it and speak to what they've decided and have more press conferences?

Duy: I agree, and I would say that, along those lines, yeah, I think they need to be shifting their focus more toward explaining how their decisions relate to the medium-term forecast, and to push away this idea that they are only focused on the short-term or the most recent data point. I think somewhere within the communication structure, they have to spend much more time saying, "Look, this is what we think is happening over the next five years, and our policies are really set to deal with that situation."

Beckworth: It's interesting. I had, as I mentioned, Andy Levin on the show, and I've talked to Scott Sumner about this, and both of them have suggested that maybe they have a quarterly report the Fed puts out that would allow it to say, this is where we're going, where we'd like to be... more of a medium-term perspective, not play by play in real time.

Duy: Right. Right. And in some cases, you do see that information within the SEP. I think they could highlight some of that information more.

Duy: So, for example, if you look at Stan Fisher's speech this weekend, where he specifically draws to some of the forecasts done within the Federal Reserve in preparation of a meeting in 2011... it makes it very clear that they're looking at the impact of their policy over a five year horizon. And so, getting I think more of that sense of the appropriate time frame for policy would be helpful for them communicating their strategy.

Beckworth: Yeah. Anything to improve communication, I think, is very useful because I think there's definitely a mixed track record over the past eight years since the crisis where they could have done a better job communicating.

Beckworth: Let's move on, though, in this discussion of being a Fed watcher. Let's talk about kind of the FOMC meeting itself; you mentioned that's an important part of your job. And let's talk about the two parts; first, let's talk about the lead up to FOMC meetings. So, you're getting ready for it. There's... eight meetings a year, is that right? Is that the right number?

Preparing for FOMC meetings

Duy: Every six weeks.

Beckworth: Yes, okay. And what do you do as a Fed watcher to prepare for those meetings?

Duy: So, it's almost as if it's a continuous process, right? I'm not always preparing for one meeting; I'm almost always preparing for a meeting, is one way to think about. Mostly as I go through that process, I'm trying to track a combination of what's happening with the data flow and how does that inform me about the likely direction the Fed's forecast is going to change, and that will also presumably inform me about which way interest rates are likely to move, either at that meeting or over the course of the foreseeable future... so that's always the critical issue.

Duy: The next issue is, I think, following the speeches that I think are most critical in maybe redefining the Fed's objective function. So, for example, tracking... over the last year, for example, trying to track information on to what extent the Fed was aware what the natural rate of interest was, and then that would obviously feed into what they thought their short-run and medium-term interest rate targets were going to be, and that's information you really need to get from the speeches over time... and so, generally tracking those two things together to try and understand what's likely to change at this meeting relative to the next, and how is it likely to influence policy going forward.

Beckworth: Now, I also recall the Bernanke instances where he has speeches that changed markets and probably changed the tone leading into the meeting. So, leading up to QE2, he made an announcement at a speech... I believe that was back at the Jackson Hole meeting, I can't remember for sure.

Duy: Right. Right.

Beckworth: And then probably the most famous one is the taper tantrum that he created back in early part of-

Duy: 2013.

Beckworth: 2013, yeah. And of course, what was... he brought this term up, they might have to "taper" some of the Fed purchases, and it wasn't for many months later until they actually started doing it because of all the turmoil it created.

Beckworth: But I'm curious... a moment like that when Bernanke really kind of roiled the markets with his talk, are you scratching your head, or are you like, "Thanks, Bernanke, more insight," or what?

Duy: It's usually a combination of both. But there are situations where a change can happen relatively quickly in how the Federal Reserve is viewing the economy or viewing the use of their tools, and certainly, those are exciting periods of time. There's also periods where they do that, and I think that was one of them, where maybe the message is a little bit ahead of where the economy is. And I think that's why you had, in fact, the taper tantrum, that sort of excessive reaction on the part of markets relative to what actually occurred, is... it wasn't clear at that point in time what in the data was driving this decision to taper asset purchases.

Duy: And those are, in some cases, the most interesting periods, and those are periods, in a sense, where I think the Fed's communication strategy falls short.

Beckworth: Yeah, so I've been a vocal critic of Fed policy, as you know, during this time. There's things they've done right. There's things I think they did not so well. And I've been making the argument for a while that they effectively started tightening in mid-2014 by talking up interest rate hikes, and the reason I say this is because you look to Fed Fund futures, interest rates... maybe in the year horizon, they start going up mid-2014 all the way to about the end of 2015... you get to the first rate hike, so the market was anticipating it, and I argue that was a form of implicit tightening. The market was tightening in anticipation of the Fed's actions. But Narayana Kocherlakota actually says, "No, no, no, Beckworth, it goes back to 2013 when Bernanke did his taper talk. That's when the Fed started tightening early." So, it's an interesting different perspective that he has.

Beckworth: But the economy... my recollection, in 2014, the first part of the year, the economy was doing relatively well. We had some rapid growth. I remember one columnist for The Washington Post saying we're going to have a boom again. Of course, that didn't happen; things began to kind of weaken after that. And again, my view is that to some extent, what you mentioned earlier, the Fed may have gotten ahead of the recovery.

Duy: Yeah, no, 2014 was a... the first part of that was a pretty hot year, and was really shaping up to be the first real solid year after this expansion, but then it petered out for a number of reasons, and the Fed adjusted their policy position accordingly. But I would agree with Narayana Kocherlakota that 2013 was really a point where the Fed did change the tone and began tightening monetary policy, and I think a little bit prematurely.

Beckworth: All right, let's talk about the meeting itself. Now, the FOMCs... now, we're still on this track of being a Fed watcher. And for those listeners who don't know much about FOMC meetings... and maybe I should define that, that's the Federal Open Market Committee meting, as Tim's been talking about, and that's where Fed officials come together and they set monetary policy for our country, for the U.S. economy.

Beckworth: Tell us about what comes out of that. So, you mentioned the SEP. What is the SEP? What is the dot plot? What is the statement and the press coms? Tell us in some detail what those things are about and what we learn from them.

What Comes Out of FOMC meetings?

Duy: Right, so there's multiple outcomes of the policy meeting, but I guess you start with the FOMC meeting is the eight times a year where the Federal Reserve Open Market Committee sits down and actually sets the course of policy. So, these are the eight times of year, really, where you have the likely chance that something is going to change, that they're going to change interest rates. Now, certainly they can have emergency meetings, but we'll sort of put that aside.

Duy: What comes out of those meetings? A couple different things. Every meeting has a statement associated with it, and the statement is usually a summary of what went on at the meeting and what the conclusions of the committee were, usually some statement about what's happened for current economic activity recently; what's likely to happen in the forecast; some basic statement about policy; and a list of who voted for/voted against the policy option. So, the statement gives you the first piece of information about what has happened or transpired during that meeting.

Duy: Then the other pieces don't come as often, and it's... you have every other meeting, once a quarter... you also have the summary of economic projections, where FOMC participants will submit their forecasts, and those forecasts will be presented in the economic projections. And you also get their individual projections on the appropriate path of monetary policy over the near and medium terms; that's often referred to as the dot plot, where policy makers think interest rates will be at the end of the year. That's not so much a forecast, and we shouldn't think of that as an official forecast of the Federal Reserve, but guidelines as to where policy makers think that policy is likely to be headed.

Duy: And then, finally, we have what's arguably the most important event of the quarter, and that's the chair's press conference, and the press conference is the chair's opportunity to explain really what the committee did and what actions... and why did they take those actions, and how they expect that to influence policy going forward. And that's an opportunity for a Q&A with the press right after that quarterly meeting.

Beckworth: Yeah, so you mentioned this summary of economic projections, or SEP. I wanted to talk a little bit more about that. So, if you go to the Board of Governors' website where they release this, they'll say this is the projected... it's not a forecast of what's going to happen, it's more of a projection of what should happen given each individual's assessment... and they use a term, appropriate monetary policy.

Duy: Right.

Beckworth: So, given what they think... this is what we should be doing, and they're going to have different views... given where we think monetary policy should be going, these are the outcomes for inflation, GDP, and then, of course, the dot plots, the actual forecast of the Federal Funds rate would be where they think appropriate monetary policy is going to be.

Duy: Correct.

Beckworth: Now this is where I want you to take off your Fed watching hat and be Tim Duy, the economic professor... do you ever find it surprising, some of the... their ideal stance of monetary policy, the appropriate path that they put down?

Duy: Well, sure. You always wonder about the high and the low dots, in particular. What are the models that some of these participants are using that give them either a very, what we call, a hawkish outlook, that you need to raise interest rates dramatically to contain inflationary pressures... or the opposite, a very doveish outlook. And, then, of course, one of the... not mysteries, but one of the standouts right now is St. Louis Federal Reserve President Bullard's dot, which is at the bottom because his model is a little bit different than everyone else's and he doesn't think interest rates are going to shift appreciably over the near term at all and can't forecast a longer term one.

Duy: So, certainly, those outlier dots always jump out at you, when oftentimes it's a question... those are often ones that you sort of throw out, too, because they're probably so far out of the norm for what policy makers are thinking that they're probably not terribly useful pieces of information, which is why the financial market participants are always drawn to that median dot as a representation of a forecast, but I I prefer to think if it as a baseline.

Beckworth: Yeah, as a baseline given their views of what appropriate monetary policy should be.

Beckworth: So, look, I guess where I want to go this is if you look at those summary of economic projections and look at the path of inflation given the appropriate monetary policy, and I've been tracking this for some time, and I actually went back in... as far back as I could collect the data... but since the crisis has begun, you can get these inflation, these median core PC inflation forecasts up to two years out, in many cases, and other times, three years out, depending on the timing of the meeting.

Beckworth: The FOMC participants all have it as most, two percent, the projected path given the appropriate... and often times, below it; so, we're talking two years out, they see as appropriate inflation... let me rephrase that. They see inflation hitting two percent at most given appropriate monetary policy, and oftentimes below that. And the thing is, this is true... 2008, 2009, all the way to the present... and that really... I don't want to say frustrates me, but it blows my mind, particularly in 2008, you've just had core... we didn't have core deflation, we had headline deflation, but core inflation had dropped, the economy was weak, and if you're going to have a symmetric inflation target where you hit 2 percent on average, you would expect to see at some point a little bit of, at least, a temporary overshoot.

Beckworth: And so, I guess, correct me if I'm wrong here, but what I'm reading from that is that these participants are saying appropriate monetary policy is a very conservative, very take-it-slow, we're not going to be too aggressive. We're not going to try the heat the economy up and make up for missed opportunities and lost time.

Duy: Right, and certainly, this has been a criticism of policy for a number of years now, is that, why aren't we overshooting now to compensate for some past undershooting. And I think... I think what this really comes down to is... the Fed, as you know, and maybe our listeners don't know, but as you know, the Fed takes seriously this idea of inflation expectations, and does not want to take policy action, I think, that causes people's inflation expectations to shift, and they certainly don't want policy that causes people to think that the Federal Reserve does not have a consistent expectation or consistent target for inflation.

Duy: And so, I think what that amounts to in practice is, they never think they can forecast above target inflation. They're going to forecast target inflation over the longer term because that's their forecast and they think they're held to that, which they are. And if we're below target right now, then the path to target is directly to target. Do not Pass Go, do not collect $200, right?

Beckworth: Right.

Duy: And that's what I think feeds through in those forecasts. Now, I think in practice, should we ever get inflation above two percent again, which... with good policy, I think we should expect; I think we might see something very similar, except from the other side, is that policy makers will continuous forecast inflation to decrease to two percent, and not undershoot. And if they forecasted to undershoot, they would think that there was an error in what they considered to be appropriate monetary policy, just as I think now if they forecast it to overshoot, they would think there would have been an error in what would be appropriate monetary policy.

Beckworth: Well-

Duy: And I know that's not always the answer everybody wants to hear, but-

Beckworth: Well, I guess what it tells me, then, is... what is communicating is a bit puzzling, a little bit... I have to get Tim Duy on here to decipher this for me. It's not very clear what they're trying to project.

Duy: Well, if they are perfectly clear, then none of us would have jobs anymore, right?

Beckworth: Nice.

Duy: It would be certainly an issue. But you're right, some of these things are fairly nuanced, I think, is one way to think about it. And at different levels, people need different information. So, I was at a Brookings conference where one of the analogies was that, so the Fed needs to give information for people that are on Level 1 of the video game, and they need to give information for those of us that might be on level 10 or 11 of the video game. And they're not the same kind of information; they're often much more complex, and that's, I think, what we're working with here, is that Level 10 information.

Beckworth: Okay. Well, I guess they're not catering to me. I'll leave it at that.

Beckworth: So, let's talk about, then, some more things from the FOMC meeting. We have this actual statement you mentioned; it's released when the meeting's over, so we get the news; four times a year, along with that, we'll get a press conference and we'll get this data, this summary of economic projections that Fed watchers pay close attention to.

Beckworth: And then, three weeks after the meeting, we get actual minutes of the meeting, and my question to you... do you find those minutes... and it's been three weeks, a lot of the stuff has been digested already... do you find those minutes useful as a Fed watcher?

Duy: I find them not as useful as the other sources of information. The minutes are often more balanced than I think the actual committee decisions are. So, for example, I think you'll often interpret them, which has been the case, I think, in recent years, much more hawkishly than the actual committee decisions have turned out to be and the path has turned out to be, and I think because all voices come out in the minutes and it's harder to separate what's most versus a few participants saying this or that... so I think it's something you have to read very cautiously.

Duy: I've said before that whatever your prior is about policy, you can find it in the minutes.

Beckworth: Interesting.

Duy: And it's also very important, I think, to distinguish between the portion that's the general discussion within the minutes, and then the portion that's the committee discussion. So, try to focus on what is the central view of the committee and you'll find that more in the committee discussion. That will give you maybe a little bit more insight.

Duy: I also find the staff review interesting, especially where maybe the staff's forecast, the balance of risk, seems different than the participants' forecast and the balance of risk.

Beckworth: Yeah. You mentioned the minutes... you can find whatever you want to find in there to fit your priors, to fit your view of the world.

Beckworth: So, personal story, 2011, the FOMC was talking about a nominal GDP target late in the year, and it came out in the minutes, and I was like, "A-ha! So you're saying there's a chance."

Duy: That hasn't really played out.

Beckworth: It did not, and I got my hopes up. I blogged about it. I was super happy. I was stoked, but now it's a big tease.

Duy: And that's what happens with the minutes, and it happens, I think, to all of us sooner or later. You see something you've been looking for or fits your pre-conceived notion of what the Fed is going to do, and then you jump on it, but it's never as clear as that seems.

Beckworth: All right, so we will call the minutes the big tease and leave it at that.

Beckworth: So, we have the minutes, and the one last thing... I think I know the answer based on what your answer was for the minutes... we've got the transcripts, the actual verbatim words, five years afterwards; and so, at that point, for a Fed watcher, I'm guessing they're almost pointless.

Duy: Well, yeah. I mean, the challenge is that the faces have changed, the data has changed, even maybe some of the underlying theories have changed. Whether they add value or can't add value is, again, understanding the process... if they accept that there's some institutional process that repeats itself again and again, I think that has some value.

Duy: But you're right... I mean, five years ago, the economy was much more different than it is now, and luckily, the economy is different than it was five years ago.

Beckworth: Yeah. So, it's a great source of information for academics doing research, trying to explain what happened in the past, and there are some treasures, as you mentioned, that maybe help you understand how the institution works.

Beckworth: One example comes to my mind, and I went through looking in the early 2000s, and I found that the Federal Reserve actually estimates its own short-run, natural Federal Funds rate; so, a natural interest rate, a short run version of it.

Duy: Right.

Beckworth: With bands around it for competence intervals and stuff. I was like, "Well, that's pretty useful." And so, I knew the Fed is generating it, and I've always wanted them to release it, and Janet Yellen gave a talk at the end of 2015, I believe, where she actually reproduced that same information. And one of my great quests in life is to get the Federal Reserve to release that on a regular basis, because even today, when Janet Yellen gave her testimony for the Senate, she alluded to the natural interest rate... now, the summary of economic projections kind of gives a forecast of the long run naturally.

Duy: Right.

Beckworth: But she's talking about this short run, where should... our listeners who don't know, the natural interest rate is where rates would be on their own, rates would be neutral... and the Fed, in theory, would put its target industry at the same level or value as the natural interest rate, and that's the best it can do.

Beckworth: The problem is it's not observable. We don't know it, but we can estimate it with models, different types, and if that has its own models... and so, it'd be nice if they would release that. They'd give us a sense of what they're thinking.

Duy: Right. So, it's interesting, and I don't know the answer to this, as to what extent that that information is somewhere embedded in the FRB model that they've released that you can get an EV's version of, I believe?

Beckworth: Yep.

Duy: So I don't know if somewhere that information is actually embedded into that. But I do think one of the missing pieces in the communication strategy is more information about the forecast, but one of the challenges is the committee itself doesn't have a floor cast that they... again, the dot plot isn't a forecast in the committee, it's a range of you as a possible appropriate monitor of policy. It would be nice, maybe, to release just the stats forecast, in which that information would be contained that you just said, and make it clear that this is not the committee's forecast itself, but they have not seen it as something they want to release just yet.

Beckworth: I think it would be useful. I think it would help conversation, communication. I think hearings before Congress would be better. And even if the Fed didn't perfectly line up its Federal Fund rate target with its natural rate, it could say, "Hey, here's where they are, here's why we're doing it. There's something has come up." So I think it would be useful. It would provide a nice benchmark. Otherwise, we're left scratching our heads. Well, what is this natural rate?

Beckworth: Okay, well let's move from Fed watching and now kind of stepping back and evaluating Fed policy kind of in general, and in a few minutes, we'll get maybe more granular into some of these issues... but kind of just stepping back, taking the 30,000 foot perspective... how you do view Fed policies since the crisis in 2008? Overall, has it been a success? Have there been areas where you wish they would have done things differently? What is your big takeaway for the past eight years?

Evaluating Fed Policy since 2008

Duy: My general takeaway is has been fairly successful, that in the absence of aggressive policy, we were spiraling into a situation that could easily become another Great Depression. So, I give the Federal Reserve a lot of kudos for stepping up to the plate and taking very aggressive action that was basically unthinkable five years earlier.

Duy: So, generally, I think it was good. Could it have been better? Well, of course, in retrospect, it could have been better given some of the... I think, certainly, in 2008, there was a bit of hawkishness that probably would have... in retrospect, seems to be unnecessary. Where I think you could have been a little bit more... I think it could have been better, was on the management and some of the balance sheet issues. One of the issues that I've had with the management of the balance sheet, of the expansion of the balance sheet, the quantitative easing... was that I think they put up expectations that that was going to be reversed a little bit too quickly... that basically the Fed was doing this as a temporary action and they would reverse it as soon as possible. I think that, again, set up expectations of tighter future monetary policy that A, wasn't actually realized, and B, was not particularly helpful.

Duy: So, that's one of my main criticisms of policy over this period of time, was... that, and other instances, and you pointed to the taper tantrum of trying to turn policy tighter sooner than we really needed to turn policy tighter.

Beckworth: Yeah, so, we'll come to this in a minute. The Fed is, at some point, going to shrink its balance sheet. In fact, there's increased talk about it. You've written about it in your recent column. But, I guess from the get go, maybe it wasn't as clear that that shrinking of the balance sheet was conditional upon really robust recovery.

Duy: Yeah, and this is... somewhere early on, the Fed should have spent a lot more time thinking about what kind of economic conditions were going to trigger changes in policy, and communicating those changes, and we finally started to see that with the Evans Rule emerge, for example. The Federal Reserve is not going to consider changing interest rates as long as inflation was below target and unemployment was high.

Duy: So, those were real economic conditions that were going to drive policy changes going forward, and I think it was very much important to lay those out. Sometimes the policy seemed to be, we don't like the balance sheet as big as it is, so we're going to change that as soon as we can. Well, what is as soon as we can, and why are you so desperate to change it, and what is it you don't like? And I think those are less helpful ways to design policy.

Beckworth: So what you're saying is it created some added uncertainty about what the policy is happening in the future. And I would extend this, and I'm going to admit before I do it, this is Monday morning quarterbacking here... I'm going to be critical with what I'm about to say... but I think a lot of what they did during this period was very ad hoc. Now, maybe with anyone, this would have been true, but the QE programs... QE1, they stopped in March 2010, thought, "Okay, we've solved the problem." "No, no, we need to do QE2."

Beckworth: QE2 is a fixed amount, they've begun that, and with 600 billion dollars, as opposed to being conditioned on the state of the economy... then they do Operation Twist, and finally they get... I think they're refining their process until they get to QE3, which is much more conditional on the state of the economy. They're going to keep doing QE, which I think is definitely... they should have done that from the beginning; nonetheless, they did that.

Beckworth: And so, I kind of see this kind of... you're pointing out a good point about, it's unclear when they're going to reverse themselves, but I see this issue of, they're kind of making it up as they go along.

Duy: And they were making it up as they go along, right? I mean, this is not a situation that the Federal Reserve expected to find itself into, and certainly you get to plenty of criticism of policy ahead of the crisis, that the Federal Reserve didn't see the danger of the housing market as clearly as they should have. Certainly, they, at that point, had no idea that the short-term rate was going to drop to zero percent, and they'd be forced, even at that point, to find alternative tools

Duy: So, they were sort of groping in the midst of a crisis to find polices that worked. And in retrospect, we would have started off QE and said we're going to do it at a fixed amount for 40 billion a month or whatever until the economy turns around, but that's, again, Monday morning quarterbacking. We certainly would have liked them to do that, but we would have liked them to do what they did rather than doing nothing, as well.

Beckworth: Right, right. One last critique before we move on. What about forward guidance? That was, in my mind, very ad hoc-ish, too.

Beckworth: Just to give an example... so, after December 2008, they lower the Federal Funds rate target to a quarter of a percent, and they say they're going to keep it there for "some time." Then, from March 2009 up through most of August 2011, they say "for an extended period of time." They never defined exactly what that means. Then, in September 2011, they say they're going to keep it low "at least through mid-2013." And then that promise gets pushed back to, let's make it mid-2014. Let's make it mid-2015... and eventually, the Evans Rule comes in and kind of saves, I think, the day for them.

Beckworth: Do you think... putting aside my ad hoc criticism, do you think forward guidance really made that much difference?

Duy: Oh, I do think that that made the difference, and that it did help reduce yields, and did help cement some certainty about, at least, the near-term direction of policy. And again, you're right; I mean, it was certainly ad hoc policy at the time, and in retrospect, I have to imagine the Fed would like to basically say... instead of talking about time periods that they would hold interest rates low, they would like to just say, "We're going to hold them low until they don't need to be low anymore. And when they don't need to be low anymore is when we see a reasonable chance that we're going to hit our forecasts over the medium term, and at that point, we can start to reverse course."

Duy: But this is lessons I think we've learned over the last several years, so I would imagine that, should we go through another crisis, they would jump much more quickly at defining the economic outcomes and policy in terms of the forecast much more quickly than they did this time.

Beckworth: All right.

The Fed’s Balance Sheet

Beckworth: Let's move to the Fed's balance sheet now and spend the rest of our time there. As part of that, I want to ask another general question about quantitative easing, QE, or what some people call large-scale asset purchases. And to be clear to our listeners, the Fed's balance sheet, on the asset side of this balance sheet, those are the assets that's buys up... so, the Fed's balance sheet in 2008, prior to the crash, was about 800 billion, and then it jumps at its peak, a little over 4.4 trillion, so there's a huge increase and it buys up a bunch of Treasury securities, buys up a bunch of mortgage-backed securities from Fannie and Freddie.

Beckworth: And then on the liability side, which is the side we normally think of... where the Federal Reserve, when it creates money, it creates a liability on its balance sheet, the monetary base... most of that becomes bank reserves, in this case. There's not a huge increase in currency. Most of it is bank reserves, often mostly excess bank reserves.

Beckworth: So, this approach, QE... I think someone can make a reasonable argument it put a floor under the economy. QE1 prevented, as you mentioned earlier, the Great Depression from reemerging. But my take, and I just want to hear what you think... it didn't deliver a robust recovery, right? In fact, even now, I think we still have the vestiges of a weak, sluggish recovery. QE did not give us... QE did not live up, I guess, to the hopes and expectations of many people when it was first introduced.

Duy: Right. Well, so I think a couple things are going on when it was introduced, I don't know that anyone really understood that the potential growth rate in the economy had dropped probably significantly in the wake of the financial crisis, possibly and I think likely for reasons that were unrelated to the financial crisis itself; so, for example, falling productivity and a slow of labor force growth.

Duy: So, of course, the gap between expectations and reality, by definition, was going to be high, because that reality was really based on a model that really wasn't relevant to us anymore. So, I think that was certainly one issue. So, essentially what that meant is you weren't going to get four or five percent growth to get out of this recovery; you're going to get much more moderate growth as we exited. So, I think that's one place.

Duy: And two, it's always hard to prove the counter-factual, right?

Beckworth: Sure.

Duy: What have happened in the absence of quantitative easing? And I tend to think that it would have taken longer for financial markets to heal, that it would have set up expectations of lower inflation, which would have been reached, hampered the recovery.

Duy: So, I think that putting those two things together would give you a realization that quantitative easing probably helped more than we often realize.

Beckworth: No, I think that's fair. I think that in the sense that it... again, it put a floor under the economy. Things would have been far worse had the Fed done nothing. It arguably could have made this a Great Depression 2.0.

Beckworth: I guess my critique is, putting aside even the real recovery that was weakened... and like you said, it could be due to some exogenous change in growth potential, aging populations, any number of things... even the nominal economy, though, was weak, right? Even the inflation... we mentioned earlier, inflation was weak. Nominal GDP... any kind of measure of that part of the economy where macro-policies should have influenced at least over several years; maybe not immediately... it was anemic by historical standards. And I guess that's where I feel disappointed, or I feel that QE didn't deliver. It didn't hit its inflation target.

Beckworth: And then some folks would argue, and I think Narayana Kocherlakota would be one... but there's others, there've even some Fed officials... there was a paper, I think, by David Reifschneider who made this argument... that potential growth itself may not have fallen as much had there been a more aggressive response: more fiscal policy, more monetary policy.

Beckworth: So, I guess that's kind of the... where I'm... maybe I'm pushing. I would like to know what your thoughts are.

Duy: Yeah, so it does seem that maybe we could have boosted inflation expectations higher, and if we had some failures, I think it goes back to what we were talking about is... when the Federal Reserve made clear they wanted to reverse quantitative easing as soon as possible... basically take an operation that could be largely permanent and change it into a temporary operation... did that prevent inflation expectations from rising as much as they could have? And had those higher inflation expectations been realized, would we indeed have a more aggressive recovery than we have now?

Duy: And I think that certainly is a fair question to ask, and some of this goes back to the decision to set an inflation target, right? That was not there before the crisis, the two percent inflation target. And I think at the time that the target was viewed as a way to entrench expectations to allow the Federal Reserve to do more, right? You could do more QE without affecting inflation expectations.

Duy: But unfortunately, I think one of the conclusions was that maybe we needed to boost inflation expectations and learn more quickly than we really did. So, I think that's a place where we could have done better. We could have done better by, I think, not doing the stop/go, stop/go of quantitative easing, and we talked about that earlier, that maybe had we had a consistent policy... again, we would have set expectations to supercharge the economy a little bit more in those early stages.

Duy: But we all have to take this in context with... there was some serious severe damage in the wake of the crisis... to household balance sheets, for firm balance sheets... that were not going to go away very easily no matter what the Federal Reserve did.

Beckworth: I think that's true and you raise an interesting point that making the two percent official... I believe there's evidence that it as implicitly understood to be two percent before 2012, but making it official, the point you raised that it kind of hardens the already kind of expectations of low inflation may have actually... they saw it as a future; maybe we see it as a bug?

Duy: Right.

Beckworth: But I want to be clear to our listeners. I'm not... and I think you, too... I'm not calling for a repeat of 1970s inflation here.

Duy: No, no.

Beckworth: When we say higher inflation, the least I'm thinking of is a temporary boost to offset the temporary decline so that you have kind of a level path. In fact, this is why I have been calling for some kind of a level target where you make up. I mean, it's very much a predictable, systematic, almost a rules-based approach where you aim for the levels, as opposed for the growth rate. But again, that's... from sitting from our seats behind the mics, that's much easier said than done.

Duy: Right, right.

Beckworth: Well, let's talk about your columns you've written about the... So, we've been talking about the Fed's expanded balance sheet; over 4.5 trillion dollars now. And the Fed has begun talking about actually shrinking it. So, there's been exit strategy principles that's outlined in several meetings to June 2011; FOMC, the September 2014 FOMC; Janet Yellen in her August 2016 talk at Jackson Hole kind of reiterated this. It was in the footnote but she reiterated these plans, and the general thrust has always been... when we normalize policies, so we kind of get away from this unconventional craziness we've been doing, we're going to first do it by raising interest rates, and then we're going to let the balance sheet shrink. And you have kind of pushed back against this view, and you have a column in the Bloomberg View, says, "Fed's Bullard (note: this is James Bullard, the St. Louis Fed. Notice his treasury yield curve)." So, tell us about your critique of that approach and what you would have the Fed do?

Duy: Yeah, so, I have a couple issues with this. One is, it's always kind of surprised me that the Fed didn't sort of go back out the way they came in; so it lowered interest rates to zero first and then... using the quantitative easing, and now we want to instead of sort of exiting quantitative easing first, they want to start with raising interest rates.

Duy: And certainly, the reason is this is that they have a lot more confidence, I think, in interest rate policy over balance sheet policy, and primarily would like the balance sheet, I think, just to go away on its own and sort of solve itself as a problem for them, and solving itself as a problem for them means maybe they could just stop reinvestment of the principal and let the balance sheet roll off on its own, and we could sort of revert back to some position like it was prior to the crisis, although I don't think anyone believes that they can get to a balance sheet that was quite as small as then.

Duy: So, I think that the Fed would like this problem to go away by itself, and I think there's others of us that maybe think that there's a fairly significant tool that they could be using right now. So, Bullard was on one of those, he said, "Look, maybe rather than trying to manipulate the short-term rate," which he doesn't believe needs to go up at this point, "we should start to use some balance sheet policy now and start to end balance,"... reinvestments and allow the long end of the yield curve to rise, and let that rising interest rates on the long end stem or slow economic activity to what we think is an appropriate... I think, certainly, he sees this issue of, if we just keep raising interest rates at the short end, and the long end stays where it's at, then we're going to flatten the yield curve and sort of hamper the credit intermediation process that that steep yield curve provides.

Duy: And I don't think that he's entirely without support there. We've certainly seen it in other speeches where his questions about what happens if the long rate doesn't move or the short rate moves? How are we going to manage financial stability under those circumstances? So, I've been wondering whether we're going about this the right way.

Duy: And I think my second point is that even if we do it… the way to do it is a passive strategy. I mean, if you believe that increasing the balance sheet eased financial conditions... the financial accommodation was much greater than it would be otherwise... then, ending reinvestments and shrinking the balance sheet is going to have the opposite effect. And so, that seems that that has, at a minimum, some trade-off for a interest rate policy.

Duy: So, how is that being worked into the appropriate policy rate projections in the summary of economic projections? And I'm not seeing where those two pieces of information are being squared in those projections.

Beckworth: So, your big concern, if I'm hearing you correctly, is that by going the order they want to go, raising short-term rates first, then unwinding the balance sheet, which holds a lot of long-term treasures in it, therefore affecting long-term rates, you could create financial stability concerns because banks, for example, they borrow. They fund themselves on short-term rates and they lend out at long-term rates, and if you compress the difference between the two, bank profits shrink and it creates financial stability concerns.

Duy: Right, right, exactly. And so... again, this is a point that Dan Truitt made in a speech, is that we look at these financial stability concerns and say, oh, well, one answer is to raise interest rates, and he said, well, it's not clear that raising interest rates will actually ease those concerns, but they might heighten them because we would affect that process.

Duy: And so, I think the natural thing then to say is, well, if that's the case, then we need to use interest rates to tamp bubbles, but we need to do it in such a way that does not increase financial instability when we're thinking to decrease financial stability, and maybe we should be using balance sheet policy to do it.

Duy: So, it seems that balance sheet policy would give us another tool to address not just economic growth, but economic growth in the context of financial stability, but we'd have to take a much more proactive approach to it, and the Federal Reserve appears to want to take a passive approach to it.

Beckworth: So to the extent that the Fed does influence long-term rates through its balance sheet, you want it to use that to allow long-term rates rise, at least in conjunction, or at least as much as short-term rates go up to keep that spread stable.

Duy: Yeah, at least if you want to shift the yield curve rather than flattening it. Again, I think, at a minimum, they should be having this conversation because they kind of let the cat out of the bag with the QE thing, right? It's that now we have the balance sheet; clearly, it's a tool; clearly, it can be used. How much could we be using it? And in which ways could we perhaps improve our economic outcomes by utilizing this tool more aggressively? I think that's a reasonable question to at least have the discussion about, and I'm not that this Fed is interested in having that, and I can understand why not given that there is political pushback on this particular issue.

Beckworth: Well, in the minutes we have left, let me ask you as a Fed watcher, is James Bullard the only one pushing for this conversation, or are there others who are... is the discussion increasingly going toward shrinking the balance sheet?

Duy: Yeah, I think the discussion is more toward, here's what our plan is, and we're going to stick to that plan... that sometime at the end of this year, that assuming they've got to boost interest rates, they'll start to end the process of reinvesting the principal payments, and at that point, that's where I think they have to sort of start to make a decision and have to have some strategy about, okay, what outcomes are we expecting to see from this? Are we going to switch off between the pace at which the balance sheet is declining and raising interest rates? Are we going to try to do them both jointly? What do we expect from this process? And I'm not seeing that part of the discussion just yet.

Beckworth: Okay. Not as much as it should be. So once again, James Bullard has been an outlier, just like in the dot plots, huh?

Duy: Yeah, I think at least he's opening up this discussion, and that the Fed, I think more broadly, should be taking this up in public.

Beckworth: Well, maybe this podcast in help in that direction, huh?

Duy: I hope.

Beckworth: Well, I think also there's the political economy issue here, and that is... as the economy recovers and as the Fed is able to raise interest rates, if the Fed does not shrink the balance sheet, what's going to get really, I think, toxic politically is going to be larger and larger payments to banks, interest and reserve payments... so I think that creates bad optics that will encourage this discussion because I can just see a Republican Congress and Senate, and maybe even the President saying, "Hey, why are we paying these banks we bailed out more and more money interest on payments?"

Beckworth: So, my suspicion is that force, if indeed we do get rates up high enough, will be another wind on the back of the Fed.

Duy: Yeah, and of course, then, that sort of argues though for maybe we should be doing that sooner than later, right?

Beckworth: No, absolutely, yeah.

Duy: So, if that's, in fact, when the issue's in your way, there's certainly the issue that the optics on the balance sheet has been difficult from the get to. I mean, you and I might agree on the basic path of policy, but again, explaining that to politicians has been a little bit more challenging.

Beckworth: Oh, yeah, and the optics are going to get worse, I think, if they don't do something.

Beckworth: Well, our guest today has been Tim Duy. Tim, thanks so much for being on the show.

Duy: Oh, my pleasure. Thank you so much for having me. I had a great time.

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate