Welcome to Economics Applied. I'm Stephen Davis, host of the show and senior fellow at the Hoover Institution. With me is my friend, longtime colleague and CEO and president of the Federal Reserve Bank of Chicago, Austan Goolsbee. Austan, welcome.
Steve, thanks for having me back.
Glad to have you back. You know, I guess the last time we chatted in this format was 13 months ago. Five years ago, I could just walk 30 feet across the hallway and chat with you whenever I want. I have to schedule now.
You got to treat me with more respect, Steve.
I will do my best. Seriously, it's good to see you. Last time, for those who are interested, we talked about the dismal state of productivity in the US construction sector. Today I want to talk about a more conventional topic for a monetary policy maker and it pertains to a speech which I believe you just gave at the OMFIF in London, which is where you are sitting now I believe. And the OMFIF bills itself as an independent forum for central banking, economic policy, and public investment.
I'll stop razing you. So, get the serious stuff. So, this is a speech about how should central banks respond to supply shocks, right? Which has become a central issue for central banks around the world and certainly in the United States. And you start your speech by stating the conventional wisdom and I'm just going to read a line from your speech and then you give an argument to question that wisdom and I'll ask you to set that forth and we'll go from there. So here's your characterization of what I think has been the conventional wisdom in many quarters and I'm reading directly from your speech here.
I was going to say I can't deny it. So I'm just going to quote you here. The conventional wisdom as you describe it: Central banks should look through supply shocks and only respond, and here I think you mean by tightening monetary policy, if they start to have secondary effects on other industries or begin unanchoring inflation expectations. And I think that's a fair description of the view that many have had. So you take issue with that view at least in part. Give us the argument.
Yeah. Okay. So what's behind that argument, and well before we get into my thesis, there's one aspect that's a little bit of a puzzle if you just kind of sat down and said hey isn't what matters for the central bank the gap between supply and demand and if output is growing faster than potential then you're contributing to inflation and that's the thing that the central bank is supposed to slow down. Now that could happen because demand went up or that could happen because supply went down. Why do they treat demand overheating differently from supply induced overheating? I think it is because up until recently supply shocks have historically been more transitory. They don't last very long. They either unwind if you look at weather or oil shocks or supply disruptions. They're short-lived. Either they are undone or even if they're not fully undone, they're a one-time increase in the price level. And so, as an inflation shock, they're not persistent. In a world where this thing is going to go away on its own and monetary policy takes a year or more to affect the economy, it seems like a bit of a fool's errand to go out and try to fight something that's going to go away already.
So then comes in my thesis. Number one, the last six years have taught us that it feels like these supply shocks are coming more frequently. They're bigger and they're more persistent than that kind of conventional wisdom suggests. So, I think that there are some supply shocks that the central bank can't look through. Namely, they're the ones that are persistent on inflation. And in a way, it goes back to the law, as you know, gives the Fed the responsibility and virtually all central banks the responsibility to restore price stability, which we have said is a 2% inflation target. And think of the extreme case. There's a persistent supply shock that's going to raise inflation to 4% for the next 50 years, but you know it's going to go away. Should you look through it? Well, of course you can't look through it because if inflation's going to be 4% for the next 50 years, you're failing on the inflation target. So, okay, the central bank must respond. It doesn't have to, part two, it shouldn't respond as strongly as you would respond to a demand shock, but you can't just look through it.
Okay, there's a, I know what's coming in the part two. Well, we'll get to that speech.
What if I change it on you?
Well, okay, fair enough. I'll adapt on the fly as best I can. But I want to just comment on two parts of what you just said. First, I think there's an implicit element in your argument that you did not state explicitly and that you don't state in your speech either, which I think is worth articulating. And you can correct me if I'm wrong, but I think implicitly you also have in mind that supply shocks tend to come to the upside in terms of their inflationary pressure.
Would you say that? I'm thinking of negative supply shocks.
Not a productivity growth increase.
That's important. That's important. So, you can't just expect the negative ones and the positive ones to kind of average out over time.
Yeah. At least I agree. I agree with that. And if we want, when we get into the second part where we think about why would supply and demand, why would the reaction to supply versus demand be different, it's going to have to do with wage stickiness and that might introduce an asymmetry, let's call it, between that supply type of asymmetry. So we can come to that. But I wanted to also just press you on another aspect of what you just said.
And you use the term go away on its own.
And there's an awful lot packed into there.
And let's take the case of the oil price shocks in the 1970s.
They went away on their own in terms of their inflationary impulse, but they didn't reverse in terms of their price level effect really until '86.
Okay. So, we're kind of mixing level shifts and growth rate shifts here.
I don't want to, it's important that we not, I would characterize the 70s, I take a slight bit of dispute with the characterization of what went wrong in the 70s. It's not that we had an inflationary impulse and the inflationary impulse went away, but the level remained high for too long. In that case, it was the secondary effects blowing into inflation expectations. And that's the thing that turned what could have been a temporary inflation shock into a persistent, maybe even permanent inflation shock. Not just the level. So
I'm not going to put words in your mouth, but let's turn that to a tariff. So let's think of a tariff.
Where I think your argument would apply most clearly to a one-time tariff would be a one-time increase in the price level. If you put in a tariff of, whatever, 10% impact, you'd have 10% inflation for one year and then even if the tariff itself never went away, the inflationary impulse of it should go away after that first year.
What I'm arguing is that the bigger the supply shock, what we kind of learned in the COVID time, the COVID inflation was supposed, the supply chain disruption was supposed to go away in a matter of months. And that was the logic of hey, this is transitory and team transitory says we don't need to raise rates because this thing is going to go away and the price level effects themselves would reverse naturally. So the level, even if you thought the level was never going to go back, that still is supposed to go away on its own as an inflation shock.
But these big ones, it's not going away. That's why I think for big negative supply shocks, they're far more, you'd be better off just assuming from the beginning this thing's going to be pretty persistent on inflation. Okay, that's certainly what happened in COVID. That's what happened with the tariffs. That's maybe what's happening with oil. So, we better be careful, right?
And I just want to remind our listeners here who aren't all steeped in monetary policy and inflation history.
That's a reminder. There's some important context to put on the table about the 70s and you've implicitly alluded to it in your remarks. The 70s were a period of rising inflation ever since the late 60s continuing throughout much of the 70s. So when the oil price shock hit, the Fed was not operating from a position of credibility.
And that very much contributed to the follow-on effect of the oil price shock pushing up inflationary expectations.
I think that's right. I think that's a perfect description.
That was a very different setting than say when the COVID supply chain shock hit.
Okay. So, just so the audience understands that in some historical settings, the concerns about what in principle is only a price level effect contributing to some kind of longer term inflationary response through expectations responses is much more pronounced. And then I can't resist noting that another prominent aspect of the 1970s was pressure from the executive branch on the Fed to loosen monetary policy. Not that that would ever happen these days. So you don't have to respond to that.
No, look, I'm not going to get myself in trouble. And as much as you try to get, I went through a lot of years with you as my colleague getting me in trouble, I'm not going to let you get me in trouble on this. But Paul Volcker was a friend and a mentor to me and really one of my personal heroes. And the essence of the Volcker episode where he becomes the Fed chair in 1979 and they literally raised the Fed funds rate to over 20%. The essence of that is how extremely painful it is to reestablish credibility once it is lost. And it's definitely worth remembering that. I thought your description was spot on.
Yeah. So yeah, that is definitely worth remembering. I think that's on the minds of central bankers and of anybody who is familiar with monetary history. Okay so I want to just ask you about one other thing. It's not something you address in your speech directly, but it's somewhat raised by these issues, these discussions about level versus inflation effects and so on, which kind of begs the question of, well, why doesn't the Fed target a price level path that has the price level, you know, because then you'd have to get back to that path if you overshoot for a period of time because of a supply shock of the sort we've been talking about. So why not do that?
Yeah. Why not do that? I think, and I say this flippantly and maybe I shouldn't, but I think this is a little bit the, my grandfather when I was a kid and to go visit him. He lived in Abilene, Texas, but he refused to take me or my cousins to the movies because his thing was like, 'When I was a kid, going to the movies cost 10 cents. I'm not going to take you to the movies if you go at that time whatever it costs $8. I'm not going to take you to an $8 movie.' And I remember even I was a kid, but I remember thinking first of all, was that a silent movie? Like that was in black and white. Like maybe it's worth more money. But also I remember thinking, well, how much did people make then? And so the question of radical disinflation or deflation. So if you saw the price level, price levels up 20%. Why can't we just go back down and have 20% deflation and get back to the price level that existed?
Well, a more practical version would be why can't we have 1% inflation for several years?
No, no, I understand that's why so I want to start with the extreme case which is on the public's mind. Why can't we go back to the price level? And the answer partly is because prices are up 20%. And wages are up 22%. So if you wanted the price level to go back down, the wage level would have to go back down. And we don't really know how to get nominal wages down without a huge recession and grinding them down. So now say would it be feasible that we would say okay well we're 20% above the target of where it would have been if it had just grown 2% this whole time. All we need to do is have a half a percent inflation for the next three or four years. But I think it's still the same problem. How do we get to half a percent inflation? The answer is big recession grind down and you start running into the problem which is the more sophisticated version of the very thing that I'm talking about in my talk which is why is the response to demand different? If you decide that you're not going to look through supply but you have to respond because you think it's persistent. Why is the response different than to demand?
Yeah, because you got this trade-off. You got a dual mandate and you got to harm employment to get, so, can I just pause you here for a second, Austan, because you're making a different argument, I think, than what has been widely expressed arguments about why you don't want inflation rate to be too low because you get close to the zero lower bound and all that stuff. You're making an argument that turns on downward stickiness of nominal wages.
Which is a little different.
It is very different. It's my response to your, I thought you were expressing a thing there. There's a problem with setting the inflation target at half a percent or 0%. Which goes to, hey, if inflation is that low on an extended basis, you're going to very frequently be hitting the zero lower bound. I do think that's valid. Your argument wasn't though, let's change the inflation target to be super low. It was why can't we just have lower than target inflation for a short period to get the price level back to where it was?
Yeah. Maybe the zero lower bound is a component, but I think more the component is it's not that easy to get inflation down really low even for a short time without really messing up the employment side.
Okay. And by the, just to be clear, I wasn't like advocating that. I'm just, it's a natural question and I think you've given an answer as to why that might not be as, this price level path target at 2% per year on average over time might not be as appealing as it would look otherwise.
But you have, we have started to get into your argument, the second part of your argument in your speech. About why there's an asymmetry between demand overheating and the supply shocks that tend to raise inflation on average as we talked about earlier. And that has to do with wage stickiness. So you want to describe that part of the argument.
Look, it's a little subtle, but this is a podcast of subtle. This is a serious podcast. Let's go through this. This is a serious podcast. The idea is the following. You for the same size gap in output. The tool that the central bank has is a demand tool. By raising the rates, the interest rate sensitive sectors of the economy like construction, business investment, consumer durables, those things slow down and that is the sense in which you reduce aggregate demand by raising the rates. That's the demand side. So if there is a demand shock you can in principle just reverse it, just reverse the demand shock to, especially if the demand shock is concentrated on those interest rate sensitive sectors. Yeah, to the extent the cycle, but in a way you don't, it doesn't even have to be in the same sectors as long as you're reducing aggregate demand by as much as aggregate demand is going up. You don't have to get any wage adjustment when you do that. Now think of the supply side goes down. So you've created the same overheating wedge, but now the Fed can't pump oil like they say. So you can't expand supply as the central bank. All you can do is try to bring the economy back into balance by reducing demand and getting everybody used to disappointment. In a way, it's like we've got to reduce wages. We've got to reduce output to match the lower potential that has happened to us.
Okay. But the problem with if you do that instantly is you will overshoot on the negative side on employment and output because wages are sticky. They don't adjust instantly. But here's, there's a part of the argument that seems somewhat strained. Which is so let's distinguish between real wages and nominal wages.
