It seems … off to write about macroeconomics with the grim news from Ukraine as a backdrop. But even as the bombs fall, the ordinary business of life goes on; getting and spending and Federal Open Market Committee meetings will continue. In short, we still need to talk about inflation.
So where are we in the inflation debate? As I see it, there are two serious points of view, which I think of as the Overheaters and the Skewers. That is, one side of the debate sees inflation as a result of too much overall demand, pushing the economy as a whole above its speed limit. The other side sees inflation mainly as a result of the distortion of spending caused by the pandemic, with consumers shunning services and rushing to buy goods, overstressing supply chains. I’ve been mainly a Skewer, although one who’s willing to give some credence to the Overheater view.
What I want to argue now is that the Overheater and Skewer positions have a lot more in common than many seem to realize. Specifically, given the data, the only way to tell an overheating story about inflation is to accept a view about how the economy works that also makes the Skewer position credible. And conversely, the assumptions behind the Skewer position also leave room for an overheating story.
What am I talking about? Let’s look at some numbers.
Textbook macroeconomics relies a lot on the concept of “potential output,” the maximum level of real gross domestic product consistent with stable inflation. There are problems both with that concept and with the estimates of potential output produced by official bodies like the Congressional Budget Office and the European Commission, but we can ignore those issues today and focus on what the official estimates are telling us.
Here’s the official U.S. output gap — the percentage difference between actual G.D.P. and the budget office estimate of potential G.D.P. — since 2000:
Ignore the huge gap during the height of the pandemic, which was more about lockdowns than about conventional economic forces. Focus instead on where we are now and where we were in the aftermath of the 2008 financial crisis.
According to these numbers, the U.S. economy is still operating slightly below capacity, which if true would mean that we have no inflationary overheating at all. In reality, the budget-office estimates probably overstate potential output right now, especially because they don’t take into account the Great Resignation, the drop in labor force participation during the pandemic. So it’s likely that we’re actually running somewhat above capacity, maybe even by several percent.
But historical experience suggests that the effect of the output gap on inflation, while clear, is fairly modest. You can see this in the aftermath of the financial crisis: We had huge shortfalls in output relative to potential, but all we saw was a moderate reduction in the inflation rate, not a slide into rapid deflation.
You can also see this in the research literature. The best available estimates suggest that a 1 percentage point fall in the unemployment rate — which closely tracks with the output gap — should raise the inflation rate by only about one-third of a percentage point.
So how could what looks to be at most a fairly modest amount of overheating explain the big surge in inflation? Well, as I understand the arguments of thoughtful Overheaters like Jason Furman, they believe that the effect of the output gap on inflation is much bigger when the economy is running hot than when it’s running cold. I’d summarize that schematically with a picture like this:
That’s a reasonable argument, supported by some statistical analysis, such as recent work by Kristin Forbes, Joseph Gagnon and Christopher Collins. But here’s the thing: Basically the same argument underlies the Skewer view of inflation driven by a sudden shift in the composition of demand.
After all, why do we imagine that an unusual shift of demand from services to goods raises the overall rate of inflation? Obviously it tends to push the prices of goods up and the prices of services down, at least relative to trend. But why don’t these movements in prices more or less cancel each other out once you average them? The answer — sometimes implicit, but always in the background — is that we expect soaring demand to raise prices more than slumping demand cuts them.
This means that if demand is skewed — if people are buying much more of some things than usual, but less of others, so some parts of the economy are running hot while others run cold — we expect price increases in the booming sectors to outweigh price cuts in the slumping sectors, raising the average rate of inflation above what you’d expect given the overall level of demand. Schematically, it looks like this:
As you can see, I sketched in the same hypothetical curve here that I used to explain the Overheater position. And that is, of course, intentional. The point I’m trying to make is that the things you need to believe about inflation to reconcile an overheating view with the amount of inflation we’re seeing also make the Skewer position credible. That is, I don’t see any coherent way to insist that it’s all overheating, that skewed demand is playing no role.
The converse is also more or less true. You could, under certain circumstances, argue that elevated inflation is all skew with no overheating. But given data like quit rates, which point to a very tight labor market, it’s hard to avoid the conclusion that the economy is at least somewhat overheated. And given a nonlinear relationship between output and inflation, that overheating could explain a significant part of the inflation we see.
So is it all intellectual kumbaya? Are we all saying the same thing? Well, no. Even if we agree that both overheating and skewed demand are driving inflation, their relative importance matters. If you think it’s mainly overheating, you want the Federal Reserve to slam the brakes hard, choking off economic growth and maybe even pushing us into recession. If you think it’s mainly skew — which I do — you want a much milder response, easing off the gas pedal but mostly counting on supply and demand adjustments to bring inflation down.
But an argument about relative magnitudes isn’t the same as a fight over fundamentals. At this point, as I see it, there’s far more consensus about inflation among serious thinkers than many people, including some of the participants in this debate, may realize.
And now back to the war news.