No, I don't think those forecasts make sense. The US economy is clearly overheated, and needs a period of sub-potential growth, which the Fed needs to let happen 1/

Olivier Blanchard
1. To Brad and Paul: Do you agree with the SEP unemployment forecasts of the FOMC, 3.8% for 2022 and 3.5% for 2023? In other words, do you really believe we can get inflation back to target with a very hot labor market for the next two years? ...

But it's not clear that this will require a drastic, Volcker-type monetary tightening. Fiscal drag as the ARP fades in the rear-view mirror plus gradual rate rises may be enough to bring growth down and allow a modest rise in unemployment ... 2/
... which in turn, together with supply improvement, can bring inflation down to tolerable levels. Of course we don't know that, but I think it's premature to demand that the Fed hit the economy over the head with a 2X4. 3/
That's the sense in which I disagree with claims that the Fed is obviously far behind the curve. I want to see step by step interest rate increases until we see the whites of disinflation's eyes — not a Big Bang of rate hikes 4/

1.ブラッドとポールに:2022年3.8%、2023年3.5%というFOMCの失業率予測に同意するか? 換言すれば、今後2年間に労働市場が非常に過熱していながら、インフレを目標に戻せると本当に信じているか?



People won't be surprised that I agree with Brad here. The Fed might be behind the curve, but people who think it's obvious are making what I consider careless parallels with the 70s 1/
The big difference between now and the big disinflation of the 1980s is that back then everyone expected high inflation to be more or less permanent. Now you get ridiculed if you say "transitory", but every measure we have says people expect inflation to be ... transitory 2/
Compare Michigan Survey results for short- and longer-term inflation from 1980 and today 3/
We don't have TIPS yields from back then, but current breakevens (using 4/15/2023 bond for 1-year) tell basically the same story 4/
So expected inflation is somewhat elevated over the medium to long term, but mostly that appears to reflect higher inflation over the next year or 2 — that is, expected inflation is front-loaded in a way it wasn't back then. This matters for both supply and demand 5/
On the supply side, inflation probably isn't baked into wage- and price-setting yet, so that we probably don't need a large bulge of excess employment to bring inflation down 6/
I tried to write about the demand side yesterday, but I suspect nobody understood my point. So here's another try 7/
Why do we talk about real, not nominal, interest rates as determinants of investment? Let me start with an Austrian-style example — deliberately silly and unrealistic, plus I'm going to ignore compound interest for simplicity. Bear with me 8/
Suppose you have point-input point-output investment — say, you can plant a tree and sell it 10 years later. Suppose planting costs $100, and in a world of stable prices you'd expect to sell it a decade from now for $110. Then the investment is worth it if you can borrow @<1% 9/
(I said I was going to ignore compounding). Now suppose you expect 1% inflation, so the tree can be sold for $120. Then it's worth doing if interest rate <2%; it's the real interest rate, nominal minus expected inflation, that matters 10/
But wait: suppose it takes time to plant a tree — you can't borrow the $100 and immediately put it in the ground. Instead, suppose that it takes a year from conceiving of the investment to actually laying out the money 11/
In that case, the real interest rate that matters isn't the rate over the 10 years starting today; it's the real rate over the 10 years that start *one year from now*. Ordinarily that's not an important distinction. But times are unusual 12/
Expected inflation is very front-loaded; real interest rates over the next year are very negative, but not so much so if we look at the implicit rates starting 1 or 2 years from now. And I'd argue that those are the relevant real rates for investment in structures 13/
So if you say something like "Interest rates are only 0.25% but inflation is 6, so we have minus 6.25% real interest rates" you're not talking about the real rate that matters, which isn't nearly that low 14/
I still don't know whether the Fed's proposed trajectory for rate hikes is right. But right now , once you take front-loading into account, it implies a real interest rate for, say, late 2023 that isn't far out of line with where rates were pre-pandemic 15/