When I complain that the Fed has seemed to abandon average inflation targeting (FAIT), people sometimes respond that the Fed doesn’t seem to interpret the new policy in the same way that I do. After all, the first letter in the acronym “FAIT” stand for flexible.
My problem with this is that I cannot think of any reasonable interpretation of FAIT that is consistent with recent statements by Jerome Powell. Yes, the policy does not require an exact mathematical average for the inflation rate. But if the term “average” is to mean anything, it must mean that there are periods where you’d want to overshoot 2% to make up for past undershoots, and vice verse. And a recent Dallas Fed paper by Enrique Martínez-García, Jarod Coulter and Valerie Grossman seems to confirm my assumption:
By comparison, average inflation targeting means that policymakers would consider those deviations and can allow inflation to modestly and temporarily run above the target to make up for past shortfalls, or vice versa. [Emphasis in original]
So “vice versa” it is. The policy is symmetric. The term “average” really does mean something.
Still, I am haunted by nagging doubts that I have missed something. Maybe FAIT is like one of those ancient mystery cults, where only a few are initiated into the secrets of the temple. Perhaps the Fed refuses to spell out a clear definition for FAIT because they don’t want the public to know; they’d prefer that only top Fed officials understand how the regime is supposed to work. In that case, any criticism of Fed policy can be easily deflected by Fed officials who insist that we outsiders just don’t understand the nuances of FAIT.
Chicago Fed President Charles Evans is clearly an insider, one of the top officials at the Federal Reserve. And he has a radically different view than Martínez-García, Coulter, and Grossman. Here’s Evans:
With flexible average inflation targeting (FAIT), the FOMC allows for greater discretion as they aim for their average goal of 2%. The window over which the average is taken is not specified yet. Furthermore FAIT is asymmetric: If the FOMC finds themselves undershooting for any extended period of time, they are prepared to overshoot to compensate, but without the same worries for combatting high inflation. Excessively high inflation in the past is not compensated by low inflation in the near future. The reason is that we do not know how to deal with low inflation because of the effective lower bound. There is a second asymmetry depending whether the shock comes from the supply or demand side.
Now there’s no longer a vice versa. The policy has gone from being symmetric to asymmetric. So what’s going on here?
I suspect that Martínez-García, Coulter and Grossman were never initiated into the secrets of the temple. Like me, they are looking at things from the outside. They read the Fed’s new policy directive adopting average inflation targeting, and assumed that average meant average. But if Evans is correct, then average doesn’t mean average. (Perhaps the FTC should investigate the Fed for misleading advertising.) Evans seems to be suggesting that the Fed adopted something closer to temporary price level targeting.
In any case, it’s now pretty clear that whatever the Fed was trying to do, they adopted the wrong form of FAIT. A serious commitment to undo the effects of inflation overshooting would have largely prevented the sort of excess inflation that we’ve recently experienced.
To be clear, I am not claiming that they need to commit to undoing 100% of excess inflation. I agree with the “flexible” part of the mandate. There’s no need to offset supply shocks. But surely the Fed should offset at least the portion of the recent inflation overshoot that is due to excessive NGDP growth. Alas, they don’t even seem to be willing to do that, which is a minimal requirement for “average” inflation targeting to have any coherent meaning at all.
HT: Jeff
READER COMMENTS
dlr
Feb 21 2022 at 3:37pm
how much do you think the semantic inconsistency matters relative to the actual policy? that is, let’s say the fed had set forth a more straightforward/candid version of their asymmetric make-up for misses policy, as evans pretty much does above. is most of the damage from the lack of a level-type upside inflation anchor in the policy itself, or is there more damage from the potential uncertainty/distrust between the plain language of the policy and the reaction function the market actually comes to expect?
Scott Sumner
Feb 21 2022 at 5:03pm
I definitely think both factors are a problem, but I don’t have firm view as to the relative proportions.
In my view, the biggest problem is that the lack of inflation averaging made the initial inflation overshoot much worse than it would otherwise have been.
Rajat
Feb 21 2022 at 6:27pm
Now that Fed Governors are celebrities in their own right and due to the Fed’s past incompetence, which gave rise to the original cult of MMT, the Fed is a victim of the very time-inconsistency problem it was designed to address. If Evans is so concerned about ELB problems that FAIT – or, presumably, also TPLT – won’t overcome them, he should support an increase in the inflation target to 4%. Maybe he already does? My guess about why the Fed didn’t adopt TPLT is that it didn’t want to be shackled to any sort of timeline regarding the meaning of ‘temporary’. As Evans said, the window over which the average in FAIT is taken is “not specified…yet” (hmm).
Andrew_FL
Feb 21 2022 at 6:52pm
The Fed’s policy is Asymmetric Discretionary Inflation Targeting
Michael Sandifer
Feb 22 2022 at 1:04am
Very good comments, but lets not forget about the full-employment mandate which, given my reading of Fed statements since FAIT was adopted, seem to indicate that there will be more attention paid to metrics such as the labor force participation rate than in the past. A charitable notion of why the Fed’s kept its targeting regime vague is to reflect uncertainty about the how labor markets recover post-recession.
Hence, there are those of us, including some on the FOMC it seems, who wouldn’t mind seeing inflation rates that are temporarily a bit higher than in the past to see if labor force participation can be brought up to higher sustainable levels more quickly. There seems to be an increasing number of people, including Jay Powell, who are shrugging off common estimates of full employment and RGDP growth potential and would rather explore the problem space. I see this as healthy, as long as we don’t go too far.
This is not the 1970s, in which shortly after leaving the gold standard, bad monetary policy models and political pressure, coupled with higher velocity due to women entering the workforce, and the confounding influence of real shocks, led to a cycle of overshooting and recession. History might rhyme a bit too much for some, but it will not repeat in any significant way. As bad as the Fed is, Scott is correct to often point out that it’s improved, on average, over time.
Dale Doback
Feb 22 2022 at 2:19am
We’ll never know how high the labor force participation rate can go if the Fed can’t engineer a soft landing. The current overshoot makes a soft landing less likely.
Michael Sandifer
Feb 22 2022 at 7:02pm
The Fed doesn’t need to engineer a soft landing. The 5 year breakeven is around 2.5%. They can keep their policy stance as is and be fine
Dale Doback
Feb 23 2022 at 1:57am
A 2.5% 5 year breakeven is consistent with near term excess inflation forcing an aggressive tightening.
Michael Sandifer
Feb 23 2022 at 12:39pm
Why?
Radford Neal
Feb 22 2022 at 2:07pm
My understanding of the usual argument for why a bit of inflation could be good, especially for reducing unemployment, is that employers/employees don’t reduce nominal wages quickly when needed (they’re “sticky”), and in that situation inflation can produce a necessary reduction in real wages despite this stickiness.
I don’t see how this translates into inflation increasing labour force participation. The people not in the labour force presumably see employment as optional, and are hence *less* likely to take a job when the real wage has been lowered by inflation.
Michael Sandifer
Feb 22 2022 at 7:06pm
The problem is with your assumption that the unemployed don’t want to work. On average, it’s lower productivity workers with fewer employment options who get laid off during recessions. They also tend to be the last rehired.
Michael Rulle
Feb 22 2022 at 8:46am
When Scott finally admits we do not understand FAIT (as he says, even Powell seems to almost deny its existence) we have a problem—-not with Scott, but with monetary policy. In the beginning, Scott took the long term view of FAIT. He used the announcement date as a starting point around 2019–and would say—-“if in 10 years the ave 10 year inflation was 2% The Fed will have succeeded——if it was 2.5%—it will have failed.
But many of his readers were unsatisfied with that concept of success or failure——myself included. First, 2.5 versus 2 technically “failed” but would it’s failure really have mattered?
Scott reasonably said yes because unexpected inflation is a problem—-not expected inflation. But .5%?.
Milton Friedman believed a constant growth of money supply was what was required.
Somewhere along the way, I believed Scott was like Friedman—-except adjusting for velocity. Under that theory—-we would not want FAIT.
Powell, who I believed was a defacto monetarist, would not have brought up FAIT——undefined in operational terms—-if he was a Friedman style monetarist.
BUT——we had Covid. Then a massive spending bill. Then a declining work force—-Then we had work from home. Then we fought about mandates. Of course confusion would reign. Powell should have stated these issues required a “temporary” FAIT ——that might have at least made sense.
But or Fed head is now mumbling.
robc
Feb 22 2022 at 9:37am
The gold standard seems much easier. If someone thinks its in their best interest to do it, they can dig gold out of the ground and expand the money supply by that much.
If no one thinks it is worth it, the supply doesnt expand.
And historically, it looks like it achieved FAIT too, as the **average** inflation rate is pretty constant. The gold standard does a pretty good job of balancing out high inflation and deflation over the long run.
Scott Sumner
Feb 22 2022 at 3:53pm
“And historically, it looks like it achieved FAIT too, as the **average** inflation rate is pretty constant.”
Not really.
robc
Feb 22 2022 at 8:27pm
The gold standard averaged much closer to 0% ( the de facto target) than the Fed has to 2%.
Michael Rulle
Feb 22 2022 at 9:49am
@robc
—that’s true—-good observation—-
art andreassen
Feb 22 2022 at 3:52pm
Scott: Having just finished reading “Greenspan’s Bubbles ” by Fleckenstein and “Fed Up” by Booth I have to say this discussion about an annual difference in a CPI target of 0.5% over 10 years is other worldly. It is impossible for any FMOC to ever agree and then to work to hit such a target.
However, my purpose in writing is to discuss the major reason that Fleckenstein wrote this overlooked book, i.e., the input that Greenspan had in the savings and loan, the dot.com, and the Great Financial Collapse. Even before he was Fed Chairman from 1987 to 2006 he had a hand in praising and lobbying for a lowering of regulation financial regulations. As Fed Chairman he rationalized that technological advances were the explanation for inflated p/e ratios and financial products but not a sign of speculation and so did little to discourage them.
I retired from the BLS Office of Occupational Statistics and Employment Projections in 2004. My responsibility was for PCE and a major tool in our projections was input/output analysis. The BEA periodically publishes these I/O tables, only not annually and in nominal dollars. One of our functions was to produce tables for the intervening years and put all in real dollars. This latter required the use of the BLS PPI deflators. For those not familiar with input/output it is a strict accounting procedure requiring the exact balance of inputs and outputs, income and spending, expense and revenues; this is its main purpose and requirement.
It is, however, the use of BLS deflators that unbalanced the tables. At this time the BLS had become enamored with hedonic deflation. The expansion of high tech opened a new vista of plenty and it required means of measuring it. A perfect example of this is Morre’s Law which states that the number of transistors in a semiconductor would quadruple every eighteen months. With hedonic deflation the BLS sees if the good that is being deflated has changed in quality over the period that is being studied. If changes in quality have been determined the BLS estimates the dollar value of these changes and adds that to the nominal value of the good to derive the real value. This hedonic estimate of quality changes sometimes resulted in the real value of the good being four times its nominal value. The BLS creates output at nobody’s expense and with nobody’s revenue. So stocks were increasing in price on the basis of nothing tangible.
Booth’s book takes up where Flekenstein ends and shows why we are where we are today.
David S
Feb 22 2022 at 6:22pm
I’m not sure what to be more annoyed by–the inconsistent messaging from the Fed in recent months or the painfully slow action. A true commitment to FAIT would have required clear statements by the Fed in November or December that they would do at least these two things:
-Wind down QE faster and telegraph a rate increase schedule for 2022
-Be clear about how there would be a period of sub 2% inflation for several quarters in ’23 and ’24
Granted, we could very well return to sub 2% PCE inflation for several years post 2022 but that doesn’t excuse the Fed from it’s irresponsible messaging and actions RIGHT NOW.
bb
Feb 23 2022 at 12:23pm
Scott,
It’s becoming very clear that they don’t really have trust/faith in AIT. AIT was used as cover for allowing higher inflation last year. It’s clear now that what they believe in is discretion.
I find this very distressing. “Good” (accidentally good) monetary policy helped us avoid a tremendous amount of suffering during the last several years. If they don’t get back on track it will lead to unnecessary suffering, and a lot folks learning the wrong lessons from the whole experience. Very sad.
Spencer Bradley Hall
Feb 26 2022 at 12:15pm
Jerome Powell deliberately lied to Congress and the public saying money and inflation were not connected. Powell has changed the charter of the Federal Reserve from an inflation fighting institution into a fiat funding institution (years of undershooting followed by years of overshooting to stimulate *growth*). The result: “Real Disposable Personal Income Has Fallen 8 Out of the Last 9 Months”
Medicare insolvent by 2024. Social Security must reduce payments by 2033. Fait is an ill-designed policy to bankroll the Democratic agenda.
Fait runs the economic engine in reverse (making up for secular stagnation by increasing inflation). It is much more desirable to promote prosperity by inducing a smooth and continuous flow of monetary savings into real investment, than to rely, as we have done c. 1965, on a vast expansion of bank credit with accompanying inflation to stimulate production.
Spencer Bradley Hall
Feb 26 2022 at 6:35pm
Powell: “The connection between monetary aggregates and either growth or inflation was very strong for a long, long time, which ended about 40 years ago”
You wonder about the economics profession when there’s no railing against this moron.
Comments are closed.