It is not just that keeping interest rates low would have in all likelihood led to an overheated economy with wage-push inflation; it is also that inflation expectations looked like they might have...
Suppose we forget about timing for a moment. An increase of 0.6 per cent in the expected ten year inflation rate implies a price level 6 per cent higher than with target inflation. That's pretty much what happened in the post-lockdown period. If people expected the inflation to be transitory, then expected 10 year inflation would fall back to the pre-Covid level once that was past.
I don't think this story is consistent with full forward-looking rationality, but it seems to work.
Been thinking some more about the counterfactual expectations, ones that could've dislodged.
A way to gauge them could be to look at past episodes in the US and elsewhere of energy price shocks of similar magnitude and where the expectations DID get dislodged. That may roughly give us a sense of how much the Fed contributed by reining in the expectations to the level we actually observe.
DSGE models ought to be useful in seeing how you have to calibrate them to get results of alternative policies that correspond to seat of the pants models in the same way they can be used to examine trade policies. DeLong calls models intuition pumps.
That calibration is one of the issues. Choice of parameter values can deliver just about any result one might want. We discovered that very early on in the 1990s when DSGE modeling was being taught more widely than before. It reduces a debate about what might be going on in the economy to what parameter value is appropriate. As students back the, we found that to be a dissatisfying way of gaining knowledge about the economy and about discovery.
"Suppose, on April 21, 2022, (a) you had warrant to think that bond-market inflation expectations were both rational and representative of broader economy-wide expectations, ..."
There's a strong probability this was true. Global energy prices surged (after having risen already in prior months expecting an economic recovery) when Russia invaded Ukraine February 24, 2022. Ditto wheat prices. The doves on the FOMC were then talking like uber hawks and the hawks on the FOMC were talking like uber uber hawks. They were all united in one message. It was critical.
Having acknowledged that, regarding the expectations channel, I still feel uncomfortable about relying on a notion that "expectations might've been worse if the FOMC had not." We will never know. How will we know?
Th seems to be taking the numerical results of the market too seriously. Do we reallu think that that "the market" today believes the Fed will produce inflation averaging only 2.18% over the next 10 years?
I'm dubious that market yields can tell that clear a narrative in a 2-month window. This bond price falls faster than that bond price for a little while - that can happen for many reasons, and a hell of a lot of trading isn't with buy and hold investors.
That said, should the Fed have raised rates in 2022? Of course. Interest rates shouldn't stay at zero after the pandemic crisis. The question is how fast and how long. But maybe overnight interest rates don't matter as much in the US because banking and borrowing has changed too much. There are too many non-bank lenders, wealth isn't stored at banks, and bank reserve requirements and availability are 10X more complicated than pre-GFC.
There are supply shocks and supply shocks. Some are permanent until the change. :) Others are temporary in that they immediately start to be offset by equilibrating forces. The contraction of the labor force/productive capacity by COVID and then the reopening and shift from services t goods are the former. Shortage of truck drivers and port delay and oil prices after the Ukraine invasion are the latter. The former needs the Fed to engineer enough inflation to accommodate; the latter may not be worth the trouble.
Suppose we forget about timing for a moment. An increase of 0.6 per cent in the expected ten year inflation rate implies a price level 6 per cent higher than with target inflation. That's pretty much what happened in the post-lockdown period. If people expected the inflation to be transitory, then expected 10 year inflation would fall back to the pre-Covid level once that was past.
I don't think this story is consistent with full forward-looking rationality, but it seems to work.
Been thinking some more about the counterfactual expectations, ones that could've dislodged.
A way to gauge them could be to look at past episodes in the US and elsewhere of energy price shocks of similar magnitude and where the expectations DID get dislodged. That may roughly give us a sense of how much the Fed contributed by reining in the expectations to the level we actually observe.
Well said about the DSGE models. More people should say that aloud.
DSGE models ought to be useful in seeing how you have to calibrate them to get results of alternative policies that correspond to seat of the pants models in the same way they can be used to examine trade policies. DeLong calls models intuition pumps.
That calibration is one of the issues. Choice of parameter values can deliver just about any result one might want. We discovered that very early on in the 1990s when DSGE modeling was being taught more widely than before. It reduces a debate about what might be going on in the economy to what parameter value is appropriate. As students back the, we found that to be a dissatisfying way of gaining knowledge about the economy and about discovery.
"Suppose, on April 21, 2022, (a) you had warrant to think that bond-market inflation expectations were both rational and representative of broader economy-wide expectations, ..."
There's a strong probability this was true. Global energy prices surged (after having risen already in prior months expecting an economic recovery) when Russia invaded Ukraine February 24, 2022. Ditto wheat prices. The doves on the FOMC were then talking like uber hawks and the hawks on the FOMC were talking like uber uber hawks. They were all united in one message. It was critical.
Having acknowledged that, regarding the expectations channel, I still feel uncomfortable about relying on a notion that "expectations might've been worse if the FOMC had not." We will never know. How will we know?
Th seems to be taking the numerical results of the market too seriously. Do we reallu think that that "the market" today believes the Fed will produce inflation averaging only 2.18% over the next 10 years?
I'm dubious that market yields can tell that clear a narrative in a 2-month window. This bond price falls faster than that bond price for a little while - that can happen for many reasons, and a hell of a lot of trading isn't with buy and hold investors.
That said, should the Fed have raised rates in 2022? Of course. Interest rates shouldn't stay at zero after the pandemic crisis. The question is how fast and how long. But maybe overnight interest rates don't matter as much in the US because banking and borrowing has changed too much. There are too many non-bank lenders, wealth isn't stored at banks, and bank reserve requirements and availability are 10X more complicated than pre-GFC.
There are supply shocks and supply shocks. Some are permanent until the change. :) Others are temporary in that they immediately start to be offset by equilibrating forces. The contraction of the labor force/productive capacity by COVID and then the reopening and shift from services t goods are the former. Shortage of truck drivers and port delay and oil prices after the Ukraine invasion are the latter. The former needs the Fed to engineer enough inflation to accommodate; the latter may not be worth the trouble.