Matt Yglesias has an excellent post on how to think about the current overheating in the US economy. Here are a few excerpts:
So NGDP, even though it’s a bit of an unfamiliar indicator, is also a very simple one. Counting up the total amount of dollars spent across the economy isn’t a totally trivial task, and the official numbers are subject to error and revision. But it is absolutely the simplest part of the whole process. Calculating the rate of inflation is a much more conceptually difficult (how do you adjust for the changing quality of restaurant meals or haircuts?) and fraught enterprise. So even though the official names tend to call inflation-adjusted quantities “real,” NGDP has always struck me as considerably more real than RGDP. The former is an actual count of defined quantities (dollars spent) while the inflation-adjusted figures are necessarily an abstraction.
Yglesias cites David Beckworth’s chart showing NGDP rising above trend, and then suggests:
I spent more than 10 years consistently thinking that the Fed was erring too much on the side of inflation aversion, and I now think the reverse. Not because I’ve changed, but because this very simple indicator — NGDP — is now giving us a different message.
Unfortunately, the Fed has set policy at a level where they are expected to fail:
What I do feel strongly about is that if I had Jay Powell’s job, I would be saying to the staff, “I’m not an economist and I don’t know how to build a macroeconomic model, but we should be choosing the policy that, when plugged into our model, gives us the policy outcome we want to see.” This concept is called targeting the forecast.
There’s only one part with which I slightly disagree:
By the book, then, they should be trying to average out the inflation of 2021 and 2022 with future years in which inflation is really low. It seems like they’re not actually going to try to do that because you’d probably have to deliberately engineer a recession in order to make it happen. So the Fed is making the correct policy choice, but that just shows that FAIT is a bad framework.
I wouldn’t call that policy framework “bad”, although I agree that FAIT is not optimal in the sense that other frameworks like NGDPLT are better. But even if the Fed had adopted a 4% NGDP level target from the beginning of 2020, the recent high inflation would end up being partly (not entirely) offset by lower than 2% future inflation, even if there were no recession. And I also agree with Jim Bullard that FAIT is pretty similar to NGDPLT, as the “flexible” part of FAIT suggests they’d allow small deviations for supply shocks.
For instance, let’s suppose that Covid permanently depressed RGDP by 1%. Then under NGDPLT you’d have 1% extra inflation over the long run. That’s not very much. Something similar could have occurred under a well managed FAIT.
The fact that suddenly dropping inflation to below 2% would trigger a recession is not an indication that FAIT is a bad policy, it reflects the fact that the Fed has not been doing FAIT, and has allowed the economy to overheat in a way that never would have happened under FAIT. If the Fed had told the markets a year ago that they were serious about FAIT, and acted like they were serious, then the markets would have done the tightening even before the Fed took any “concrete steps”. Instead we had all this happy talk about “running the economy hot”, and that’s what we got.
I mistakenly assumed the happy talk was empty rhetoric, and that FAIT was the actual policy. I was wrong.
PS. Among reporters who are not professional economists, Matt Yglesias and Ramesh Ponnuru are the two best people to read on monetary policy.
PPS. Powell has a new speech on monetary policy, which contains absolutely no discussion of average inflation, nor any sort of reference to the Fed’s new FAIT policy. The new policy is as dead as John Cleese’s parrot. Powell attributes the recent surge in inflation to supply problems, with little discussion of excessive nominal spending. He also suggests that the Fed has been tightening policy, which is incorrect.