Odd Lots - An IMF Economist On The Challenge Of Finding The Neutral Rate Of Interest

Episode Date: October 5, 2020

One of the guiding lights of Fed policy over the years has been the so-called Neutral Rate of Interest or R*. It's at this rate, theoretically, where the economy comes into balance, with full employment and stable prices. Yet, not only has discovering that level become challenging, but the premise itself has been called into question. On this episode, we speak with Peter Williams, an analyst and economist at the IMF, on what it takes to find the right level, and how the concept itself can be salvaged.See omnystudio.com/listener for privacy information.

The news doesn't stop on the weekends. Context changes constantly. And now Bloomberg is the place to stay on top of it all. Hi, I'm David Gurra. Join us every Saturday and Sunday for the new Bloomberg this weekend. I'm Christina Rafini. We'll bring you the latest headlines, in-depth analysis, and big interviews. All the stories that hit home on your days off.

And I'm Lisa Mateo. Watch and listen to Bloomberg this weekend for thoughtful, enlightening conversations about business, lifestyle, people, and culture. On Saturday mornings, we put the past week's, events into context, examining what happened in the markets and the world. That on Sundays we speak with journalists, columnists, and key political figures to prepare you for the week ahead. Join us as soon as you wake up and bring us with you wherever your weekend plans take you. Watch us on Bloomberg Television.

Listen on Bloomberg Radio, stream the show live on the Bloomberg business app, or listen to the podcast. That's Bloomberg this weekend. Saturdays and Sundays starting at 7 a.m. Eastern. Make us part of your weekend routine on Bloomberg Television, and wherever you get your podcasts. And welcome to another episode of the Odd Lots podcast. I'm Tracy Allaway. And I'm Joe Wisenthal.

Joe, you know what question we keep asking lately, especially? Probably something about, you know, monetary policy and fiscal policy of the zero lower bound or whether it all works anymore and whether we have to rethink literally everything everyone has thought for decades. Yeah, I think that's right. But I was thinking, more specifically, we keep asking policymakers if they actually understand inflation. And I think that's something we asked Neil Kishkari very recently, and we asked Vitor Constancio before that. And I'm not entirely convinced that we do understand inflation. Like, clearly the fact that we have been missing the target for many, many years suggests that there is something that we are missing.

I mean, it really is one of the sort of most profound question. And we chatted several weeks ago with the ECB vice president, our former ECB vice president, Vitor Constacio. And, you know, it's like, here's someone high up in a central bank. The central bank's mandate is to either target or control inflation. And yet, there is a lot of doubt whether anyone actually has a sort of robust theory of what causes it. And that's like a pretty profound thing, given how much, given how much is. weighing on this measure or this concept.

Yeah, absolutely. And I got to say, the way policymakers think about inflation, there are a couple things or factors that come up a lot in their intellectual framework for thinking about this. And that is, or R, that is R. That is R. It's R start. Sorry, I'm not making any sense. But those things are known as the stars.

Do you know what I mean, Joe? I do. But what do you like describe them? Because I bet you'll like describe the stars better. This idea that central bankers, policymakers have this idea of stars. I'm kind of, I get it. But what do you like sort of tell people what they are? Let me just say first of all that probably the best known star is our star or the neutral rate of interest, which is the place where monetary policy isn't.

easy or tight, it's just neutral. So the clue is in the name. But the stars in general are kind of, they kind of describe the longer term steady state of the economy. So there are things like the natural rate of unemployment, the neutral rate of interest, which I mentioned, the level of potential output, the growth rate of potential output, as well as the central bank's inflation target. And they're called stars because all of the variables are shown as a lowercase letter and a little star next to it. So R star, Y star, G star, things like that. And they've sort of come into prominence in recent years. So Jerome Powell at his first speech at Jackson Hole when he became Fed chairman, gave a whole talk that was basically just about the R star. And I think the best way to think about

them is probably that the stars are pieces of the missing inflation puzzle that we've been talking about. So if inflation isn't showing up like central banks think that it should, then it's possible that there's more slack in the economy than appreciated, or it's possible that monetary policy isn't actually as accommodative as central banks think it is. And that's where the stars come in. Like, are our assumptions of the stars of the natural state of the economy, the long-term natural state of the economy, are those correct? Right. I mean, by the way, people should really go read that Jackson Hole speech from Powell from 2018, because I think it was very great. It was really good. It sort of anticipated a lot of the discussions that have ensued since then

this idea that tightness or heat in the markets doesn't necessarily show up in inflation like at once did, but showed up more in financial markets, financial instability. And then, but I think this whole concept of stars, you know, there's sort of two big questions in my mind when we, when people talk about them is one is how do we know what they are with any sort of useful precision? Like, no one knows exactly what the perfect rate of interest is at any given moment, but can we get close enough at any given moment for the to be a useful guide? And two, are these useful concepts? at all. I mean, they're inherently sort of unknowable. They're inherently sort of abstract. But are the stars a useful, you know, guide for policy period? Or should we even be thinking in

terms of stars? Both of those questions, I think, have a lot of relevancy and urgency these days. Absolutely. I agree with you on both those points. So today, I'm really excited to say we are going to dig into the stars in detail. And we have someone who sort of adopted Jerome Powell's Jackson Hull speech as the jumping off point for a really in-depth examination and estimation of the stars. We're going to be talking to Peter Williams. He's a research analyst over at the International Monetary Fund. And he wrote a paper this summer called Reading the Stars. So, you know, the good title for decoding all these different stars and natural rates and things like that. Peter, thank you so much for coming on all thoughts.

Thank you so much for having me. Really appreciate the opportunity to be on the podcast. And I guess I just want to sort of start off by saying that all the views I articulate today are solely my own, not the IMF's executive board, its management, or the institution as a whole. They're just my personal views. Okay. you've done the disclaimer. So excellent. We can dive into the paper. So I guess my first question is, Joe sort of hinted at this in the intro, but how do you even go about estimating things like the stars?

Because to me, they just seem like such big and vast variables that can change quite easily. I don't even know how you begin to try to guess what they actually are. So can you give us a description of how people traditionally go about it? Yeah, so a lot of them were kind of traditional methods, especially sort of before the crisis, you know, they usually had what we're kind of, I would say fairly simple models. Oftentimes they just kind of looked like these, you call them like just moving averages of the data. So, you know, your estimate of the long run rate of unemployment might just kind of look like a slowly moving trend that kind of fluctuates up a little bit in bad times and goes down a little little bit in good times. But to a large extent, you know, the problem of economics, especially in the

70s and 80s was one of inflation. So auditing sort of really nitty gritty kind of like details about, you know, what's the perfect rate of unemployment or the perfect rate of growth, we're kind of just subsumed by trying to get inflation down to a much more lower and stable level. And like when you have interest rates at 10% or 12% inflation's at eight or nine, you know, some of the little tiny variations don't matter as much on the real side because you have a, you have sort of a bigger problem from a monetary policy perspective in inflation. There's some interesting debates that we've been discussing them a lot on the podcast about whether these are useful concepts, period, at all.

But before we get into any of these questions or really dive into some of the more technical questions about estimating them in real time, what's the history of the stars? Like, where did this idea that there is some natural rate of interest out there, or natural rate of inflation or anything like that out there that is. worth pursuing is worth understanding. Where do that come from? So, I mean, the sort of like first real usage of like, especially our star is kind of the one of the longest history. And that goes back to like Nut Wicksell, who's really a name. I've only ever read in textbooks referencing him as kind of the first creator of the idea of the neutral rate of interest. So that, you know,

very long ago and sort of economic history. It's sort of, you know, then the original idea of the stars, especially our star, the most prominent one, which is, this is the rate of interest, which if over a long horizon kind of equilibrates, you know, first it was like a quillibrate savings and investment, but, you know, companies and people invest and save at different interest rates in the policy rate. So, you know, over time, it kind of became more and more policy focused and sort of less this kind of generic whole economy interest rate. And that, you know, that evolution has been happening. We could really stay starting in the 50s with as economics became a bit more quantitative. And then it's kind of continued ever since where like the models

have become more in depth and rigorous. They've added more variables, more data, but also at the same time, you know, the sort of like underlying issues about like, do these things really exist? How precisely measured are they? Are we dealing with like data that has lots of regime shifts? Those problems don't go away in economics. That does make sort of, you know, any policymaker, I understand sometimes the frustration you might feel as a policymaker, like, where should we be going? And the answer is we think right now the answer is here, but the world may change. We may have different problems in five years. You know, it's never a very simple solution as to where we should go. So speaking of that evolution, I mean, you mentioned in your paper that part of the

inspiration for it was Powell's Jackson Hole speech, mostly about R-Star back in 2018. Why do you think our star has sort of like experienced this new popularity or why do you think it's become of interest to policymakers, especially at the Fed? Well, I wouldn't say it wasn't of interest kind of before the crisis, but... Right. I mean, more interest relatively. Right. Well, it wasn't of interest because the concept wasn't particularly interesting, because if you look at, you know, estimates from 1990 to call it the beginning of the financial crisis, the data doesn't really move a whole lot. The estimate, you know, of R-star might go from like two and a half to three and then goes back down or goes back up a little bit. But it's, you know, in a lot of ways, like the great moderation really is a true

kind of feature of the data. Like, everything's pretty nice and stable at sort of like this kind of top of economy, a very macro view from, you know, kind of when inflation falls, early 1990s, until the financial crisis. Obviously, there's lots of interesting things happening inside the economy and there are these sectoral changes, but we just look at the top down, it's like, okay, you know, R-Stars kind of just juggling along. So there wasn't a whole lot to talk about in this sort of classic assumption that R-star is just two, which John Taylor put in his kind of Taylor rule in 1993, very famous paper that sort of guides a large chunk of monetary policy research sense. He just kind of says two, you know, he like has an intuition about it.

But it's basically he just kind of says 2% because it feels right. And that sort of is the anchor for most everybody's estimates until really for most forecasters and policymakers estimates until about 2012 or, you know, 2012, 2015 is when we see this big evolution of, wow, we really need to rethink where we are. This isn't just like a simple shock from the financial crisis. Like all of the data seems to have changed. All of these stars seem to have changed. You know, the world is really, this wasn't just a simple business.

cycle that hit us, but something a bit more kind of long-lasting and structural. I'm Francine Lacqua, an award-winning journalist, and I've got a new podcast, leaders with Francine Lacqua from Bloomberg Podcasts. I've interviewed everyone from heads of state to fashion icons about the news of the moment. But I've always been curious, who are these people as leaders? I don't think there's one right way to be a leader. Make decisions. A poor decision is always better than no decision. Listen to new episodes. other Monday. Follow leaders with Francine Lacroix wherever you get your podcasts. So basically, we had these relationships in the economy that seemed fairly stable,

relationships between unemployment and inflation. We had monetary policy that seemed to do the job as expected, where the Fed might lower rates and then generate some activity and maybe some inflation, then hike rates, and then, uh, you know, slow the economy and, maybe slow inflation. And essentially, it's the post-crisis period in which these relationships seem to break down, in which unemployment plunging didn't seem to induce a rise in inflation. And extremely accommodative monetary policy by historical standards didn't seem to do much easier. That is essentially reignited this debate about where and what the where and what are stars.

Yeah, I'd say that seems right. I think, you know, initially right after the crisis, if you look at sort of like private sector forecasts or sort of the FOMC's like Tealbook forecasts from the staff, you know, you sort of see that everybody kind of thinks, you know, after three to four years, we'll kind of be back to normal. Obviously, we don't get back to normal sort of in our papers view until maybe 2018 or 19. You know, it's a very long and drawn out recovery and not just on sort of the growth side, which I think is easier for people to kind of think about, But also, you know, the unemployment rate, you know, was quite high after the crisis, but it also falls, you know, quite slowly. And so we have, you know, this situation says, okay, that, you know, why is the unemployment rate falling so slowly? Why is growth not rebounding in the way we thought it might? And, you know, there are other confounding factors as, you know, I think every economist is sort of continuously reminded when trying to actually do work with data. You know, it's easy to make models that sort of describe the mechanics pretty simply, but the frustrating reality is that, you know, we, you know, in a model like this, we sort of have. You know, you know, in a model like this, we sort of have. the neutral rate and sort of the policy rate are kind of these two variables that say,

okay, whatever the gap between these is, we call like the stance of monetary policy, that will sort of accelerate growth or it will decelerate growth. But, you know, in reality, you've got fiscal policy, you've got overseas developments, you have financial conditions, shocks, you have all these other confounding variables that make it pretty hard to get a read-through into exactly what the stants of policy really is once we kind of control for all these other things that, I mean, frankly, you know, if we had a perfect world, then we can model everything just the way we wanted it to, we could take that into account. But, you know, sadly, economic data is not usually particularly informative in the way we would like it to be. And it's

everything moves in the same, you know, everything's always shifting. All the data is moving around usually in very correlated ways. So it's, you know, it's a challenging set of statistical problems in addition to being a challenging kind of definitional and conceptual issue as well. Could you talk a little bit more about your model then and how you sort of accounted for those difficulties and what your model does differently to traditional estimates of the stars. I read Chair Powell's 2018 Jackson Hole speech a whole bunch of times right when it came out and was just kind of stewing over it. It's like, well, we've been having these exact same debates he outlines. You know, growth has been low. The unemployment rate has now by that point gotten to be

very low levels, but inflation doesn't really seem to be there a whole lot. And we're just trying think, okay, what is, you know, we conceptual, like, as institutions, we kind of want to have these guides, these sort of guiding variables. They help us, they help us in forecasting, they help us understand the economy right now. You know, they help us think about sort of longer run structural and framework issues. So you kind of, you kind of need them as sort of organizing thoughts and principles as an economist, at least in sort of like classical, like, classic sort of policymaking institution style econ. If you're heterodox, you can kind of dispense with them and maybe things become a bit simpler in some ways. But, you know, inside the big

institutions you like them. And so what we decided to do, sort of taking Chair Powell's prompt was, first of all, we know that sort of of the stars, we know that one of them is kind of fixed. The Fed has told us we want inflation to over the long run be 2%. And they've tweaked the definition recently, you know, going from sort of symmetric to 2% with instead now they're at sort of an average 2%, but, you know, I don't think that makes a particularly large difference in terms of the model perspective. You know, it may lead to changes in policy going forward, but, you know, from a modeling perspective, you can handle them pretty similarly. So you know, Pi star, the long-run inflation target is 2%. And so you can sort of treat that as an organizing principle around the rest

of the economy. And you say, when inflation is at 2%, we would expect everything else to be at equilibrium. So we can sort of fix all these other kind of equilibrium conditions, we call them in economics, around that kind of primary identity. And then you sort of add in some ways like, okay, what are the other things that drive inflation? Because if we describe inflation, because if we describe inflation well, now having sort of fixed this relationship between sort of 2% equilibrium, if we describe inflation well, maybe then we can describe the rest of the economy a bit better and more realistically as well. So you start at, you know, we added in, um, we, it's just, you know, Vitor talks about this in his podcast as well. You know, it's originally

the Phillips curve, the relationship between, it's called the output gap or the unemployment rate, and inflation was like, you know, was very obvious. Sadly, it's not so obvious anymore, making, you know, economist life much more difficult, but also allowing us to write papers about it. So I suppose there's a certain personal perk. And so we sort of add in, you know, oil prices. We add in some other, we call them supply shocks in economics. We can think of it as like these things that affect inflation in kind of the short run, but we don't necessarily expect they'll have very long run effects on the rest of the economy. So like short term changes in productivity growth also tend to affect inflation a little bit, oil prices, import prices, these sorts of things.

So we say we've described inflation pretty well. Now let's look about the rest of the economy. And so to sort of pin down sort of the real side of the economy, you can start adding in other variables, sort of anchor everything else. So, you know, we have GDP, very common. And then you also sort of use gross domestic income, which is conceptually very similar to GDP. It just measures on the income side, not the production side. And you can add in labor market variables. Most important one is obviously kind of the headline unemployment rate, but you can add in others because the relationships tend to be pretty similar. but sometimes you'll get just a bit more confidence in your estimates by adding in more data.

The tradeoff is that you make the model more complicated. So it's a difficult balancing act trying to create a paper like this or a model like this that sort of goes through. You have enough data to describe everything precisely, but not such a big model that you're suffering from sort of classic overfitting problems. It's, you know, if we could get another 20 years of data without any structural changes in the economy, that would be great. And my model would be estimated a bit more precisely, but I don't think that's going to happen. just knowing the way the last 20 years have gone, it seems unlikely that you'll get this nice, kind of perfect economist's dream scenario.

You know, you described this process of unemployment, maybe some productivity. Heterodox eotypes would say, look, this whole thing is like nonsense. There is no such thing as R-Star. There's not going to be some rate of interest that sort of magically brings the economy into balance. And, you know, the pursuit of it is a waste of,

time and of course you can have a lot of fantasy equations with Greek letters that somehow make it look like you've mathematically derived this thing but you know it's all as you say and you as you acknowledge you know if you do too much you risk overfitting it and not actually getting any data just step back and sort of in your view defend the premise of this search for a sort of mythical number that if we were to find it and if we were to find it and if we were to know it, and if we were to have the formula, would roughly bring the economy into balance. Actually, Joe, that's quite a tough one. Sorry.

You know, I understand sort of the heterodox perspective, because it does seem like sort of classical econ has struggled a lot in the last, especially, you know, since the crisis. But I think a lot of it is that to some extent, both as a policy makers, you know, you want to have a sense of how much you're actually affecting the economy at any given point, right? So this notion of like where the neutral rate is. So a good analogy might be sort of driving if you can't see the speedometer in your car. You have some other indicators, but you don't really know how fast you're going and if you're actually appropriate for the space around you.

So sort of we can think about the neutral rate and these stars is sort of helping give us a bit more context for where we are relative to where we, you know, kind of should be based on the rest of the economy or our make-believe spodometer-less car. And so it helps us make sure we're going at the appropriate level and we're not over-stimulating the economy because the problem is that if you over-stimulate the economy and you give it, you know, way too much support. It's not been a problem since we hit the zero lower bound. But historically, you know, eventually you would get in a situation where you've overheated the economy. Things are sort of booming too much and you've kind of created this investment boom or whatever kind of boom.

And then, you know, booms pop and recessions are, you know, to be avoided. And, you know, I think everyone in economics would say it's easier to sort of be a bit more forward-looking and kind of think about where you are and you can go from there and to create more of an optimal path, whereas if you try and just always be reactive and backward looking, you know, it's much harder to kind of really get a sense and sort of avoid some of these predictable issues. And I think it's an interesting example, actually, is that, you know, not exactly the neutral rate, but the Fed has sort of applied these same sorts of principles in other ways. Obviously, they're still guided by the neutral rate for the sort of setting the Fed funds rate.

But on regulatory policy, you see sort of similar things. You know, they, the stress testing exercises are actually, I think they're not applying the neutral rate concept, but it's sort of the same way we want to have safety guideposts sort of around the economy. We're not just sort of, you know, we sort of moved on and changed a bit sort of the underlying paradigms on the regulatory side. I think that actually helps compensate for some of perhaps the weaknesses of the neutral rate, just because the neutral rate is so low now that it's hard for, you know, regulatory policy to maybe support as much, but then you can say if we keep the financial system pretty safe, one positive of that is that it makes recession probably less likely and less likely to be particularly bad. So you can sort of use other policy

tools to kind of help make up to this kind of lack of policy support through the classic interest rate channel. They might phrase it a bit differently, but I think that you can see that in some other policy reform agendas that have kind of come along for it, especially since COVID hit. So I wanted to ask you a little bit more about that. So one of the things from your paper is that your estimate of our star, the neutral rate of interest is basically lower than I think almost everyone else's in academia. Were you surprised at that finding? And I guess secondly, to the financial stability point, what did your paper actually tell us about the previous financial crisis and what our star looked like, you know, building up to 2008? Okay. So I'll tackle the first

one of those. So, yeah, when we kind of like first run the model and we see the outputs come out, I was a little surprised by the R-Star estimate. I'll agree with you. I think it's the lowest I've seen, at least for the U.S., you know, from, I think anywhere. But I think it makes sense because sort of what ends up determining R-star is, you know, the output gap or like, we call it like Slack, the business cycle, the output gap. After the financial crisis, the output gap was very, very negative, but also recovered very, very slowly. And so because it was recovering a lot more slowly than sort of the model kind of naively would expect otherwise,

it suggests the policy was a lot less supportive than it normally would have been, which, you know, when you sort of work out the math, if policy is less supportive, that means your neutral rate is actually lower than you might have expected. And, you know, we make some other mechanical tweaks in our model that may affect our views of the neutral rate a little bit,

like we include what's called the shadow rate, which sort of tries to incorporate the stance of monetary policy via QE when you're at the zero lower back. So the shadow rate could affect our estimates a little bit from all these sort of QE programs and forward guidance. But in the end, you know, it was a very long, very difficult recovery after a very large shock. And I think it seems kind of natural to imagine that policy was less supportive than we thought

at the time, you know, just because it took so much longer to get back to normal than we thought it should have taken. And so going back to the kind of the story of the crisis that flows out of that sort of, you know, the way we defined everything in our paper, you know, it's all about the inflation target, long-run inflation expectations. Because we sort of divorced the inflation targeting mandate from kind of the financial stability one, you know, the paper does give, I think, a look at the period right before the global financial crisis that's a little different from most others. So we see, you know, right before the crisis, we actually don't think there's

a positive kind of output gap. We think the economy was, you know, pretty close to normalish conditions, But what was happening underneath was we'd had, you know, okay growth in the mid-2000s, not spectacular by the standards of the 90s, but it was okay. But what it actually happened is that sort of the stars were all kind of deteriorating about, like potential growth was falling. And so it looks like what happened prior to that, you know, in the lead up to the crisis, was basically we had a lot of credit fuel growth, which is sort of supporting overall growth. But it wasn't really leading to any sort of positive benefits on the supply side of the economy. So you have what is effectively is like credit. it boom, which kind of deteriorates sort of long run growth, it affects the other sort of potential estimates. And then we sort of the credit cycle blows up and we kind of just gotten back to an

okay state of the economy. And it was interesting when we saw those estimates, in one sense, it was surprising because you think the economy was like overheating. We just had a big bubble that popped. Obviously the real side of the economy should have been overheated too, but really it's the financial system was full of excess. But I think, you know, if you go back and sort of look at people's kind of lived experiences in the mid-2000s, you know, the economy wasn't really that great for, you know, many or most people, you know, so I think it's actually one of the interesting parts to me about this paper is that we were the anchor on the inflation target, which I think for a lot of people that's very hawkish, that means you must not really care about people,

you know, it's very kind of austere way of looking at the world. But the results we get by doing that are actually, you know, suggestive of, I think, you know, we actually made a lot of mistakes in the last 20 years. And so as a result, you know, we were, we're kind of thinking about and seeing the world do kind of this overly hawkish lens because they actually weren't thinking about inflation hard enough. Like we weren't really meeting our inflation mandate. It's lessons to learn. You can get the news whenever you want it with Bloomberg News Now. I'm Amy Morris. And I'm Karen Moscow here to tell you about our new on-demand news report delivered right to your podcast feed. Bloomberg News Now is a short five-minute audio report on the

day's top stories. Episodes are published throughout the day with the latest information and data to keep you informed. Yes, there are other products like this from a variety of news organizations, but they usually rerun their radio newscasts throughout the day. That's not what we do. We create customized episodes that can only be heard on Bloomberg News Now. And we don't wait an hour to publish breaking news. When news breaks, we'll have an episode up in your podcast feed within minutes. So you're always getting the latest story. and developments. Get the reporting and the context from Bloomberg's 3,000 journalists and analysts

we're all over the world. Listen to the latest from Bloomberg News Now on Apple, Spotify, or anywhere you listen. This seems to be, you know, we recently had an episode with Minneapolis Fed President Neil Cash-Cari, and I guess it's sort of implicit in the new Fed framework. But this seems to be the sort of counterintuitive conclusion. that a lot of people are arriving at, which is that whereas at one point a sort of inflation first perspective may have led people down a sort of more hawkish policy path that over the last several years actually taking the inflation mandate seriously and taking seriously the idea

of hitting the goal is the path toward would in theory have led to more doveish policy, more accommodative either or more accommodative approach. I think that's right. I think, you know, another part of it, too, the paper kind of suggests is that right after the crisis, you know, we had this big shock and nobody, there was a lot of debate about where we really were in the business cycle and if the stars had changed. And if you'd kind of been a little bit, actually, perversely, if you'd been a bit more model dependent, which is another sort of perhaps anathoma to a lot of heterodox people, but if you'd been more model dependent, the models actually are all very uniformly revised down their estimates of our star and potential growth, you know, basically 2008 Q1. all of these things fall pretty dramatically. Kind of the irony is that a lot of intuitions we'd all developed, or not we, I was in college at the time, but a lot of intuitions that had been developed over the previous, you know, 10 or 20 years,

ended up actually betraying forecasters and policymakers precisely because they were sort of looking back at a period that it was structurally different from the one they were now in, but it was hard to recognize it unless you just said, we've got two quarters of data, I'm going to trust the model, which, you know, it's hard to do, but I think it's actually one of these interesting insights as well, is not just that the inflation mandate became a dovish thing after so long of being hawkish,

but then also a lot of models that people perhaps have expected sort of similar results from, all of a sudden became sort of these model outputs suggested we needed really dovish policy as well. And then sort of policymakers and forecasts, forecasters catch up to that by, I call it 2015, give or take. So it's interesting that it happens in sort of both sort of the model side but also on the inflation side. So I had a sort of geeky question, but one of the things that's come up this summer, especially as we sort of watch the coronavirus crisis unfold and its impact on the economy is everyone is laser focused on data. And people are watching all sorts of different types of data to kind of gauge the health of the consumer. I was wondering, in your model, you're trying essentially to estimate Slack in real. real time. Could you do that with alternative data? And, you know, what would that actually look like

and how much do you think it might change the outcome? You know, it's the interesting part about the kind of the COVID shock as it hits. It's, you know, obviously it's like historically, unprecedentedly bad GDP print in Q2. But if you actually kind of look about sort of the normal relationships between GDP and unemployment, the unemployment hit is probably much worse as sort of an outlier sense than as the GDP one. So it's an interesting point because it looks like a lot of these kind of normal relationships, I want to say break down during COVID, but they certainly become a lot noisier. So alternative data has been very helpful for us trying to do forecast and trying to understand where the economy is at a given moment. You know, formally incorporating that sort of alternative data into what is already a pretty big and complicated model becomes more challenging because a lot of the problem with these alternative data sets. And I think it's true in the financial industry.

It's true in lots of parts of economics too. is, you know, they don't really have very long time histories. And so you can sort of try and constrain them to look like the series you know, but you also just may end up getting to a point where, like, you're trying to add in these new series, but you only have like six months of data or, you know, like a year of data. It's always publicly available. And it becomes quite challenge. So we've been trying to, you know, think about that in terms of, you know, like sort of monthly GDP estimates and other things that we can sort of use to help your sort of nowcasts right now. But it isn't necessarily obvious.

that adding in all these alternative data indicators is necessarily good from a formal sense, so much as you have a model, you have all this new, interesting novel data, and you should gut-check the model's outputs on all this other stuff you can look at, but maybe formally incorporating it because of statistical histories and everything else, is a bit more of a challenge. I think that's probably a place where most sort of economists and policymakers have been right now. This data is a very useful novel, really interesting to play around with, very helpful in understanding who in the economy has been affected by COVID,

but it doesn't necessarily have a whole lot of bearing on a formal modeling sense just because the data is new and novel. And in general, you kind of like longer data sets. It's a bit of a, it's been a very tricky, you know, challenge, trying to map between the two sort of novel data and more like classic macro variables. I want to go back to, you know, you're talking about the sort of generational divide. And this is this is the recurring theme. It was in the Powell speech,

is this idea of, you know, we see pay more attention to financial conditions, the sort of way if we focus on inflation, that would have, at one point, that might have inclined policymakers to take a more hawkish path, now more doveish one. But I still don't think I have a sense of like,

what change? Like, why, what is the theory by which some of these sort of, old workhorse approaches suddenly didn't seem to work anymore. It's something fundamental about the economy. Is there something different? So I think the answer is yes. I don't necessarily have this developed an answer for what the things that, I mean, I can tell you the stars that change, but not like the real sort of like fundamental underlying driving things that changed. You know, it's an interesting problem in economics because if you sort of take off like the classic

sort of equations and you go back and estimate them like the 60s and 70s, a lot of times you get very good relationships. Like the parameters are pretty different from zero. They're very statistically significant. And they're like, if you just roll those estimates forward, they just become progressively less significant over time. This, you know, affects our ability to actually make models, but it also affects your ability to sort of know that what the model is saying is useful.

So that's one reason like in our paper, we just add a lot more data than most other modeling approaches because we figure if we have more data, things are easier to pin down. generally speaking. I think part of it is that maybe, you know, the economy is more forward-looking. There's perhaps some of it is that there's more, you know, higher debt load so everyone kind of pays more interest to interest rates. You know, it's not necessarily obvious to me what these sort of underlying changes really have been that it sort of make, you know, it's like all the coefficients in all these models. It kind of looks like everything just been squashed towards having

no effect. Like they don't have no effect anymore, but they have much smaller effects than they used to. You know, it's, I don't have a good answer, Joe. I mean, it's, I think if somebody had a particularly good answer for what's sort of driven this change over time, you know, that person deserves a much higher sort of like pay grade and probably a Nobel Prize. But at the moment, it's kind of the big open area in the econ. I don't really see particularly good answers. You know, you can do all sorts of like fancy statistical techniques. You're like, well, actually the coefficients that we do like these seven other things kind of recover and are about the same, but it's not quite as satisfying, it's like saying the effect is the same. And so maybe it's just that, you know, the

past theories have changed and it's, you know, we don't really have particularly good theories of expectation formation or sort of how people relate, you know, how people's expectations relate to their behavior at any given moment and sort of, you know, maybe people are more forward looking now and that means the models kind of have more squashed effects because all of the sort of channels we used to think of as like direct behavioral effects are now expectations issues. It's hard to say. So I have a related question, but, you know, if our star is really low or lower than a lot of people have anticipated in recent years, what are the economic conditions that would actually force R star up? So classically, you would have said an increase in the rate of potential growth was sort of one of like the main drivers of the neutral rate of interest.

I'm a little less sure about that read on the data. You know, since the crisis, even since 2000, you know, potential growth has fallen some. Our star has fallen by a lot more, which sort of means that, like, the other factors that have driven R star the last 20 years kind of dominated. You know, part of it may be just that demographic, you know, you can't really get around demographic features in the short, or on any sort of like relevant policymaking horizon. And so part of it just maybe with an aging economy and sort of a higher sort of, like higher net financial assets that leads to lower. interest rates because people are like, there's more saving, effectively. Inequality could also have an effect there as well. It's sort of like a classic, kind of more, I know, not a classic response. It's more than a novel response that has become very popular, I think, that sort of like,

in more sort of unequal states of the world, you know, rich people spend less of their money, therefore they save more. So interest rates fall because there's more saving and relatively less investment. That's maybe part of it. So, you know, perhaps a decrease in inequality would raise our star. There's other more interesting theory. sort of just that both like R-star and like the equity risk premium or sort of other measures of risk in the economy are kind of connected. So like because we've had falling rates of potential growth, but it seems like the economy has become more risky at the same time that sort of decreased R-star, but it's kind of raised the equity risk premium. Maybe that equity risk

channel might be challenged a bit in post-COVID shock right now. But I think generally that those sorts of theories as well that sort of risk behavior and people's risk preferences have both effects on R-star and the equity risk premium kind of nice because they say, sort of tie in, you know, asset markets policy and GDP growth and sort of like the classic real economy. But it's, you know, I'm not sure that you could give me a policy that would tell me with 100% uncertainty. I would say, yes, that will raise our star. It's kind of the honest answer. So I think a lot of it has to just be, you know, hoping for sort of structural reform. Even the U.S. economies have been doing relatively welcome or doing a lot of its other peers.

But, you know, we can always think, we can all think of sort of plenty of structural issues in the U.S. that could use a bit more support or policy changes that might help growth. And he sort of helps that everything that helps growth will also sort of bring up interest rates as well. But it's not guaranteed. I want to ask another question that's sort of what I would characterize is the sort of heterodox critique of all this. And so I sort of already, you know, I already talked about are these stars even worth finding? But there's another thing that I think maybe even deeper sort of precedes that logic, which is that a lot of heterodox economists would say that within this sort of more traditional mainstream approach,

that there are certain embedded ideologies that, you know, you look at a paper, and I looked at your paper, but it's all sort of a lot of papers that's like, okay, you have your abstract in the beginning in which I could read things in plain English, and then a bunch of equations that I gloss over because I don't understand that, And then I, like, read the conclusion. And their argument might be or would be that within those equations that what appears to be science or what appears to be math is a certain ideology about A, being anti-too-much government spending and anti-distorting the economy, ideology that there is such thing as too much employment and that there is such thing as too much, this implicit idea that maybe it's dangerous, if there's, if we get, too much full employment and so forth. And I'm curious, like, what you make of that because I never have done academic econ. I just learn about it from talking to folks like you. But in your view, what do you make of that critique, that there is sort of a sort of anti-government spending,

anti-labor ideology embedded in this, which then translates into actual policies of raising the interest rate too soon, even though we're not at full employment or raising interest rates in response to stimulus, to sort of blunt the effect of fiscal policy, and that basically what appears to be science and math is really sort of politics. Yeah, I'm a little skeptical of the it's all actually, you know, economists have this sort of deep-seated kind of conservative bias kind of view of the world. I mean, I am one and I don't necessarily think of myself particularly conservative, so I think there's that. But also, I think more generally, when I was sort of approaching this paper, and I think we're all like, I think a lot of heterodox,

people think about maybe sort of like the classic theories of the economy from like the 60s and 70s with like perfect markets. Everything's in equilibrium all the time. And I sort of still assume that everything that's been written since by like people in the mainstream still sort of flows from that same sort of set of assumptions. But I think, you know, when you look around it, you know, kind of cutting edge econ and sort of any field, it's almost, you know, whether it's, you know, macro, like subset of micro, it doesn't really make those sorts of assumptions anymore. A lot of times it's like asking very, very detailed and specific questions, you know, how do we the best policy interventions, you know, given that there are constraints to real resources or,

you know, perhaps, yes, perhaps even assuming that there are constraints to the government's ability to spend. Heterodox people might do that as like, you know, a very dramatic take. But I, you know, in one sense, I kind of understand the mechanical logic and the MMT argument, but more broadly, you know, I don't think there's a really conservative bias. I'm like, when I look at this paper, I see statistical relationships between data and like a point of 2% inflation that the pet has kind of told us it once. hit over the long run and then everything else kind of revolves around that. So, you know,

if there's a deep, you know, intellectual bias inside of us, you know, it's, uh, you know, my co-opendist and I weren't particularly aware of it. Maybe it's still there, but I doesn't really strike me as right. I think, you know, the errors that were made, you know, before the crisis on regulatory policy or after the crisis, you know, in terms of understanding how monetary policy evolved. Like, you know, a lot of it's just inertial behavior. We, you know, we read something, we think it's very persuasive. We sort of anchor to it. Our behavior sort of adapts pretty slowly under normal times. And maybe after a big shock, it adapts a little bit faster. But it's still, you know, our own sort of built-up wisdom is there. And I guess, you know, maybe that is an aspect that you're critiquing. But the problem is that in the end, you know, policymaking decisions are made by relatively small groups of people. And, you know, if they were just continuously bombarded by 700 different papers, trying to explain everything in the world with an infinite number of approaches, You can never make a decision.

And so I think what policymakers kind of need is reasonably built and developed papers that are, you know, modeling approaches that sort of take into account conditions in the real world, make reasonable sets of assumptions about the real world, and then hopefully lead to good policy outcomes. But I think, you know, the notion that there's kind of like a dark force driving everything. Maybe it's self-interested to sort of think that, but, you know, I've always been a little skeptical. Just on that note, I mean, we've been talking a lot on all thoughts about this idea of a handoff from monetary policy to fiscal policy is the suggestion from the stars discussion and the idea that, you know, if we get the stars right, we could actually calibrate monetary policy correctly is the implication that monetary policy, you know, isn't completely useless and can actually generate growth and inflation? Well, I remain fully confident that monetary policy will continuously generate complaints

by almost everyone across the ID. But no, I think that's, you know, going forward, I think the Fed's framework review sort of suggests an approach, which is consistent with, you know, these sorts of model-based estimates. And I think, you know, a lot of people see action as being useful. And, you know, the Fed, I'm sure, is sort of having some internal debates about, you know, what more policy actions they could do, what are the rest of. policy action, the best policy action is kind of just sitting there and letting their new approach kind of take time. But like lower for longer, the interest rate policies, which of our

paper thinks we have the natural conclusion of a lot of this, like that is still a policy stance. It's not maybe not necessarily one that people love, given the nature of it is slow, inertial, and non-dramatic. But that's sort of what comes out of the model, is that if you're at the zero lower bound, we think, you know, negative interest rates are like feasible in the U.S., then the best policy stance is to be as supportive as you can be. for kind of as long as, you know, until inflation is sustainably at 2% or, you know, slightly over it, so you can sort of maintain the 2% average. You know, that's not like a policy regime that I think people get excited about because it

means the policy is going to be quite low, quite inertial and stable. But I think that's still a policy regime, which is helpful and, you know, using the interest rate policy instrument as the way to try and achieve 2% inflation and maximum employment. You know, that seems pretty reasonable. It's interesting. It kind of ties back into your question of a minute. a minute ago on the heterodox side, Joe, if we think there's like some amount of like transitional unemployment

or did you even point like a 3.9% or 4% of long run unemployment target is only slightly above like frictional unemployment. Like that's pretty close to maximum employment. And so I think it's interesting that some of those heterodox petiques that have sort of been made about, you know, economists, you know, always trying to sort of want higher employment to keep inflation stable.

Like, you know, 4% unemployment is very few people being unemployed of anything more. more than sort of a very transitional basis. Peter, that was really, really amazing. Thank you so much for simultaneously going into great academic depth, but also making everything really understandable. Thanks so much. And thank you for defending the entire economics profession.

Someone had to come on odd lots and do it, Joe. All right, that was great, Peter. Thank you so much. Thank you all. So, Joe, I really enjoyed that discussion. I'm trying to think of a way to sum it all up because it was such a wide-ranging discussion. But I think the sort of last note that we left it on, this idea that maybe monetary policy hasn't been completely neutered. It's just we've gotten some basic assumptions incorrect.

Part of me finds that notion quite optimistic and quite hopeful. Maybe because I'm pessimistic about the prospect of governments, you know, really getting their act together. and enacting fiscal stimulus. So I want to see monetary policy help out a little bit. You know, Tracy, are you skeptical that governments will get their act together? Or are you, like, skeptical that governments getting their act together is a good thing? Both, I would say. I'm double skeptical. I knew it. No, but it's true, right? Like, even if people could agree to do something, the chances of them doing something that's well designed, I think are pretty low.

But anyway, our star. Peter is great. And it was certainly refreshing because, you know, it's like we have so we talked to a lot of people that bash this stuff pretty much all the time. And we've had like a sort of like stream of sort of heterodox in some way or another thinkers. And so but look, you know, it's like, A, we have the Fed. We have the central banks. They have their mandate. They have their jobs.

There are economics departments around the world. and I thought Peter sort of made a very compelling defense of like how the, you know, the approach that a lot of these economists and central banks are taking and essentially how they can be salvaged is I think what he's going for. Yeah, I think that's exactly right. But also, I think he put it quite well in the sense that you need some sort of guiding principle in order to make monetary policy decisions. And it feels like variables like the stars are kind of, the best you're going to come up with. And so it becomes essential to really study those and write papers like this and make sure that you're fully understanding them. I agree. And I think that like whether we talk about monetary policy, uh, sort of centric stabilization or fiscal policy, uh, centric stabilization, that problem will never go away. The idea of like, okay, whatever you're going to do, what are the variables that you're going to use? So it's like, you know, you could say like,

all right, let's have, you know, fiscal stimulus be our primary engine, but that doesn't really get you away from the problem of the sort of determining, like, oh, is inflation actually becoming a problem now? Is this noise? Is this temporary? Does everyone who wants a job have one? And so forth. So there's almost sort of no getting away from these questions, regardless of the policy regime. Yeah, I think that's exactly right. And full employment kind of seems like the obvious one to focus on here, like if you are designing fiscal policies or some sort of job guarantee program or universal basic income or something like that and you're trying to get to a place of full employment,

you better have a good idea of what that place actually is and what it looks like. So yeah, the questions aren't going away even if we do get that long awaited handoff from monetary policy to fiscal spending. So all right. Should we leave it there? I'll save it there. Okay, this has been another episode of the Odd Lots podcast. I'm Tracy Allaway. You can follow me on Twitter at Tracy Allaway. And I'm Joe Wisenthall. You can follow me on Twitter at The Stallwart. Follow our producer, Laura Carlson. She's at Laura M. Carlson. Follow the Bloomberg head of podcast, Francesca Levy, at Francesca Today. And check out all of our podcasts at Bloomberg under the handle at podcast. Thanks for listening.

This is Tom Keene, inviting you to join us for the Bloomberg Surveillance Podcast. It's about making you smarter every business day. I'm Paul Sweeney. We bring you complete coverage of the U.S. market open. We cover stocks, bonds, commodities, even crypto, all the information you need to excel. And I'm Alexis Christophers. Bloomberg Surveillance also brings you the analysis behind the headlines. We do that through conversations with the smartest names in economics, finance, investment, and international relations.

We do all this live each and every weekday that bring you the best analysis in our daily podcast. Search for Bloomberg Surveillance on Apple, Spotify, YouTube, or anywhere else you listen. On the East Coast, listen at lunch. And on the West Coast, listen as soon as you wake up. That's the Bloomberg Surveillance Podcast with Tom Keene, Paul Sweeney, and me, Alexis Christophorus.

Subscribe today, wherever you get your podcasts. Bloomberg's Surveillance, Essential Listening, each and every business day.