Last year, the Fed forecast that inflation would run well below 2% for the foreseeable future. Today, inflation has risen above 2%, and is likely to stay about 2% for a while. By itself, that’s not necessarily a problem. The Fed engages in “average inflation targeting”, which means offsetting periods of below 2% inflation with periods of above 2% inflation, and vice versa.
Nonetheless, there is some reason to be concerned about a possible stop-go cycle, such as we saw during the Eisenhower administration (which experienced three recessions.) The danger is that the Fed will not adjust its policy instruments to reflect changing economic conditions.
The Fed should always set its policy instrument (such as the fed funds rate target) at a position where it expects to achieve its policy goals (such as 2% average inflation.) That means that whenever the economy turns out to be stronger or weaker than expected, the Fed needs to adjust its policy instrument to keep expected future growth on course. But will it do so?
History suggests some reason for concern. George Selgin has argued that the Fed raised interest rates in 2015 not because doing so was necessary to achieve its policy goal, rather because it somehow felt that very low interest rates were undesirable for vague and hard to define reasons. Such low rates seemed “unnatural”. Policy can go off course when central bankers confuse the policy instrument with the policy goal.
Today, we may be facing the opposite scenario. While its too soon to be certain, I have some concern that the Fed is now treating low interest rates as a goal in itself, not as a means for achieving 2% inflation, on average, during the 2020s. I hope that I am wrong, but there seems to be some danger that the Fed will be reluctant to raise its interest rate target even if macroeconomic conditions call for higher short-term interest rates.
I do have some sympathy for their predicament. They rightly emphasis both sides of their dual mandate (stable prices and high employment.) Unfortunately, inept fiscal authorities have created a significant adverse supply shock, and thus it is more difficult than usual to determine whether the economy is in danger of overheating. There is some risk of stagflation, although I currently expect the problem to be transitory.
If the 5-year TIPS spread were to reach 3%, it would be a pretty clear sign that the Fed had fallen behind the curve and that we were likely to overheat. Another warning sign would be a sharp spike in long-term T-bond yields.
Fortunately, we are not there yet. Bond yields remain pretty low, and hence I’m not currently predicting anything more than a brief spike in inflation. Even so, I’d say that for the first time in 14 years the balance of risks has shifted from the threat of undershooting to a threat of overshooting in terms of nominal spending growth.
My recommendation to the Fed is two-fold:
1. Reiterate your commitment to 2% AIT for the 2020s, regardless of whether that commitment requires easier or tighter money.
2. Reiterate that your interest rate target will be 100% data dependent, and that you have no qualms about large and unexpected changes in the fed funds target, if necessary to keep average inflation at 2%.
In a better world the Fed would entirely abandon interest rate targeting and let markets set the short-term interest rate.
PS. You might wonder why I would be concerned by overshooting of the AIT target path. Overshooting increases the risk that we’ll experience a recession during the 2020s, and I don’t like recessions.
READER COMMENTS
Michael Sandifer
May 12 2021 at 6:38pm
Scott,
Market reactions seem to suggest much more concern about the Fed overreacting to inflation, and this has been going on since about mid-February. I think you’ve disagreed with this interpretation for months. Why do you think we have these periods lately, in which we see spikes in Treasury yields, while the stock indexes fall and there are other indications of reduced AD?
Scott Sumner
May 12 2021 at 10:18pm
I see zero evidence that the market fears the Fed overreacting to inflation. Where is this evidence?
Michael Sandifer
May 12 2021 at 11:52pm
Well, it’s quite often since mid-February that when interest rates spike, like the 10 year for example, the stock market sells off, commodity prices crash, etc. I interpret nominal Treasury rates as reflecting where the market thinks the Fed will set rates in the future, when the Treasury market is functioning properly. Since the Fed targets inflation, I think it’s reasonable to assume that when rates rise and demand seems to crash, that it’s thought the Fed will reverse the recovery to a degree.
Also, as I’ve pointed out elsewhere, the S&P 500 and broad commodity index futures curves are negative, and have been for months now. They get more negative on these days with spiking interest rates and signals of aggregate demand decline. Meanwhile, the NASDAQ and Russell 2000 indices have been flat.
And it’s not as if my interpretation of these events is unique. It’s seems to be the dominant interpretation in the financial media. That certainly doesn’t make it correct, as financial media gets a lot of things wrong, but they’ve gotten gradually better at covering monetary policy and its effects on the economy and markets in recent years.
Scott Sumner
May 13 2021 at 12:41pm
I think it’s far more likely that rising rates reflect the Fisher effect, not expectations of tight money. It’s the inflation premium that is rising, not the real interest rate.
Michael Sandifer
May 13 2021 at 1:19pm
Scott,
Yes, of course markets think the Fed will raise rates because inflation expectations are rising.
Scott Sumner
May 13 2021 at 5:56pm
That’s not what I said. I don’t think rates are rising because of Fed policy, rather because of higher inflation expectations.
Michael Sandifer
May 13 2021 at 8:56pm
Scott,
I’m not catching the distinction you’re making. The Fed controls interest rates, period. Sure, you can say that the Fed steers, but the road winds as it will and the Fed has to stay on the road, but the Fed doesn’t always stay on the road.
I think it’s possible to lose control of the longer end of the yield curve, but that’s happening and it’s not what we’re discussing. I don’t see why markets would be reacting the way they are if we’re just talking about a bit of short-run inflation due to supply shocks, and anyway, the Fed can certainly choose the let inflation run higher during supply shocks. The market is convinced they are capping inflation and we are bumping into that cap fairly frequently.
Michael Sandifer
May 13 2021 at 8:57pm
I meant to write, “…but that’s not happening”, referring to losing control of the longer end of the yield curve.
Scott Sumner
May 14 2021 at 1:29pm
You said:
“The Fed controls interest rates, period.”
This is a really bad way to think about the issue. You are ignoring the Fisher effect.
Michael Sandifer
May 14 2021 at 3:25pm
I don’t see how I’m ignoring the Fisher effect. I’m saying that when Treasury rates rise due to increased inflation expectations, the Treasury market is saying it expects the Fed to raise rates, because the Fed targets inflation and sets interest rates as a policy signal. Interest rate changes reflect changes in money supply versus demand.
To use interest rate language for simplicity of illustration, I understand the Fed is trying, or should be trying, to influence the neutral interest rate. The neutral rate changes in response to real or nominal shocks. The Fed should want to adjust the money supply (or demand, in some cases) such that nominal interest rate follows the neutral rate.
So, am I really missing something, or is this a case of miscommunication? I see the yield curve as reflecting where the market thinks the Fed is likely to set nominal rates in the future, sans market liquidity problems and quirks here and there, or loss of control over longer term rates due to loss of anchoring, etc.
Michael Sandifer
May 14 2021 at 3:51pm
A simpler way to make my point is to say that the Fisher effect affects the neutral rate, but the Fed sets the nominal rate.
Scott Sumner
May 15 2021 at 1:58pm
Again, interest rates are not Fed policy, but in your very first comment above you implicitly assume that higher rates are tighter money.
“Market reactions seem to suggest much more concern about the Fed overreacting to inflation, and this has been going on since about mid-February. I think you’ve disagreed with this interpretation for months. Why do you think we have these periods lately, in which we see spikes in Treasury yields, while the stock indexes fall and there are other indications of reduced AD?”
Maybe I misread your comment, but I assumed you thought interest rate spikes were Fed tightening.
Michael Sandifer
May 15 2021 at 8:15pm
Scott,
Well, in the quote from me you offered, I mention stock and commodity indices falling in conjunction with interest rate spikes. I also mentioned other indicators of falling future AD. I don’t think focusing on interest rates alone tells us anything about the stance of monetary policy.
Considering that the Fed sets nominal rates, and that interest rates in future years are rising while indicators of AD in those years are falling, it’s at least a reasonable guess that monetary policy is tightening. That doesn’t mean it’s loose or tight, per say, but tightening.
You think monetary policy is roughly in equilibrium or maybe a little loose. I think it’s tight, for my own reasons, with which you’re already familiar.
Michael Sandifer
May 16 2021 at 10:53am
Scott,
The very quote you reference from me includes indicators of future expected AD. I think that higher rates with indicators of falling AD lead to a reasonable guess that Fed policy is tightening.
That doesn’t mean policy is tight, but I think policy is tight for other reasons.
I don’t think interest rates say anything about the stance of monetary policy in isolation, except that if they’re higher than real GDP growth can imaginably be, then that suggests inflation is too high.
Garrett
May 21 2021 at 12:21pm
I’m a bit late to the party, but a backwardated stock index futures curve is driven by dividend yields and interest rates. As I type the SPX index is at 4160 and the December contract is at 4148. The December dividend future is at 59 and the annual dividend index is at 24.5. So you have to take (59-24.5) 34.5 points and subtract that from the spot level because you don’t get paid dividends on index futures. That gets you to 4125.5, or about 23 points below the December future. Bullish or bearish views can’t be priced into the futures curve due to arbitrage.
Mark Barbieri
May 12 2021 at 6:55pm
I’m getting nervous. I shifted more of my bond holdings to TIPS. I was wrong when I got nervous in 2007, which lead me to your blog and NGDP targeting views. I hope that I’m wrong again.
It seems so strange to see rapid and dramatic increases in home prices, commodity prices, stock prices, and cryptocurrency prices but not see much in the way of consumer price inflation. On the other hand, the TIPS spread is about as clear of a market signal as you can get. And my backup indicator is the foreign exchange rates. I’m confused and nervous.
Thomas Lee Hutcheson
May 13 2021 at 8:07am
It’s time for the Fed to stop speculating about what they might need to do with one possible instrumental variable to achieve their mandate. They could do dot graphs of inflation rate expectations and employment ratios at 1, 2, 3, 5 year out intervals to keep themselves focused on business.
MarkLouis
May 13 2021 at 9:14am
This is a Fed that’s rapidly losing confidence, in my opinion.
They say they won’t rely on inflation forecasts, yet they are relying on their own forecast that the current spike in inflation is temporary. Either you rely on forecasts, or you don’t.
They give no indication what they will do if their goals of price stability and full employment are at odds with each other. And that could easily be the case in an age of weak Phillips Curve, emergency unemployment benefits, etc.
They highlight the need to help certain cohorts of the labor market, yet that falls outside their mandate nor do they have the tools to do so.
They brush aside the regressive-tax nature of inflation while at the same time pretending to be most concerned about low-wage labor.
They claim to follow “average inflation targeting” yet they’ve given no indication that they will make up for an overshoot in any timely manner.
Scott Sumner
May 13 2021 at 12:49pm
You said:
“They say they won’t rely on inflation forecasts, yet they are relying on their own forecast that the current spike in inflation is temporary. Either you rely on forecasts, or you don’t.”
Very good point.
Thomas Lee Hutcheson
May 14 2021 at 7:11am
“They claim to follow “average inflation targeting” yet they’ve given no indication that they will make up for an overshoot in any timely manner.”
Inflation expectations were below target for over a decade since 2008. Five-year expectation have now been above target for about 2 months. I think it is a little bit too soon to worry about the Fed being unwilling to achieve its inflation target.
Max Goedl
May 13 2021 at 3:50pm
“Last year, the Fed forecast that inflation would run well below 2% for the foreseeable future. Today, inflation has risen above 2%…”
It follows that the “foreseeable future” is one year.
Scott Sumner
May 13 2021 at 5:57pm
Unfortunately, that’s correct.
Econymous
May 14 2021 at 9:09am
What’s the mechanism for an AIT overshoot increasing risk of recession?
Scott Sumner
May 14 2021 at 1:30pm
The overshoot will cause the Fed to tighten, in order to bring inflation back on target. If they don’t tighten, we’ll eventually end up with hyperinflation.
David S
May 15 2021 at 6:50am
How would it even be possible for hyperinflation to occur in the U.S. economy and monetary system? Wouldn’t the Fed go full Volcker if we had a few quarters of rapidly rising prices in 2022—say in the range of 5% and beyond? Or would the data come to late for a modest reaction?
I accept that it’s impossible to predict extreme scenarios, but haven’t we all been conditioned to except some modest tightening in the event that things seem to be going out of control? I don’t want a recession either, but I could accept some tapping on the brakes at the end of this year.
Rodrigo
May 15 2021 at 11:45am
I don’t believe we have ever had a “soft landing” a tightening cycle usually ends in a recession.
Scott Sumner
May 15 2021 at 1:54pm
You said:
“How would it even be possible for hyperinflation to occur in the U.S. economy and monetary system? Wouldn’t the Fed go full Volcker”
That’s exactly my point, overshooting risks a reaction that pushes us into recession.
Comments are closed.