Why Larry Summers is Wrong on the Fed & CPI
Larry Summers on Friday claimed that the fed will not lower rates this year. We discuss why he's likely wrong & provide our takes on inflation into tomorrow's CPI report.
Why Larry Summers is Wrong on the Fed & CPI Sneak Peak
Larry Summers is not exactly someone who people trust today. Maybe that’s for a good reason. Trust is critical in today’s world. Maybe people should distrust Summers more. If it wasn’t enough that Summers met with Jeffrey Epstein on many occasions in addition to directly asking him for funding for not only Harvard but also his wife, perhaps the fact that Summers is a former Treasury Department chair and one of the many political elites with a revolving door to the corporate world is enough to be skeptical of Summers.
If that won’t convince you, maybe Summers involvement & advisory board position with DCG, the now failed crypto scam-firm, will convince you. Obviously Summers has gotten away scot-free.
And if that is unconvincing, perhaps the allegation that Summers used his influence to guide policy to help DCG when Silicon Valley Bank was collapsing will convince you. If the government-corporate revolving door had a financial face, Larry Summers (and Steven Mnuchin) would be its face.
When we see headlines from Larry Summers, our first questions are ‘why is Summers saying this? Who is Summers speaking on behalf of?’ We try to be skeptical. In this article we explain why we think Summers is off-base on inflation & his recent statements around the fed.
Larry Summers The Inflation Hawk
For most of 2021 & 2022, Larry Summers has been the face of inflation hawks. Kudos to Summers for calling out Biden’s reckless fiscal stimulus. Kudos for calling out the fed’s mistakes and Powell for not raising rates sooner; however, much like many talking-heads, Summers is finding it difficult to come to terms with reality on the fed and inflation. In fact, one could argue that Summers represents a large group of market participants, such as Jim Bianco, who see inflation unable to get below 3%. The argument goes something like this: the economy is running hot, unemployment is very low, inflation is still higher than target, and so the fed won’t cut rates this year and will delay rate cuts to 2025.
We sympathize with Summers and Bianco on their views. In fact, we share their views in part. The economy is currently growing above trend and the unemployment rate is low. Additionally, there are clearly concerns when it comes to inflation—the printing of money, too many dollars chasing too few things to purchase.
We think that the picture has changed on inflation for now. But to Larry Summers, the picture has not changed. In fact, we think that Summers will be proven wrong by the very next inflation reading, which is tomorrow, March 12th.
In this article, we explain what we think Summers is missing and provide a view for what we expect. This article will be the first of our many articles covering inflation & federal reserve expectations when it comes to the fed funds rate. In sum: one month does not make a trend.
Larry Summers Says More Hikes Possible
Our story begins with Summers’ recent statements regarding inflation and the fed. Last month, Larry Summers stated that he estimates there’s a 15% chance that the fed will raise rates next, not cut them.
“There’s a meaningful chance — maybe it’s 15% — that the next move is going to be upwards in rates, not downwards,” Summers said on Bloomberg Television’s Wall Street Week with David Westin. “The Fed is going to have to be very careful.”
He argued this after the January CPI reading, which came in “hot.”
“That’s not the only [shelter costs] disturbing indication,” he said. Another key concern is core services prices — which leave out food and energy costs — excluding housing, which have been pushed up by higher wages. “It sure looks like super-core was explosive in January,” he said in reference to that measure… “The assumption that inflation was headed down to 2% in a tranquil, healthy, real economy has certainly been called into question by these data.”
Here is the video of the full interview. The argument is summed up as:
Last mile inflation is coming in sticky & stickier than economists thought
Housing was supposed to come down but isn’t
Core service inflation exploded in January
“Inflation” won’t fall below 3%, the fed needs to become more flexible with its 2% inflation target
As such, the neutral rate of interest is higher than the 2.5% estimate
This argument is at least accurate at a surface-level. Core services inflation did move higher in January. Economists had forecast an annualized inflation rate 20bps lower than it came in and a month-over-month reading which was .1% lower than it came in; however, as always, the devil is in the details. In fact, Summers acknowledges this.“It’s always a mistake to over-interpret one month’s number — and that’s especially true in January, where calculating seasonality is difficult,” said Summers, a Harvard University professor and paid contributor to Bloomberg TV. “But I think we have to recognize the possibility of a mini-paradigm shift.”
The Problem with Summers’ Arguments
The problem with Summers’ arguments is exactly the point he made in the above quote. “It’s always a mistake to over-interpret one month’s number — and that’s especially true in January, where calculating seasonality is difficult.” This is the problem. Recall that back in 2022 when inflation was peaking many market participants argued that ‘one month of disinflation doesn’t make a trend’. But then one month turned into two, and two months has led to 19 months of disinflation. To put the question to Summers, does January’s CPI reading really mark a “mini-paradigm shift” in the ongoing disinflationary trend? We strongly believe that it does not.
CPI vs PCE
The first thing to note in Summers’ argument is the use of CPI and not PCE. The fed claims that it’s PCE which it uses to gauge inflation, not CPI. ‘PCE is the fed’s preferred measure of inflation.’ It’s difficult to make the argument Summers’ makes using PCE. Unlike CPI, PCE did not beat expectations last week, it was in-line. Another key point that inflation hawks like Jim Bianco make is that inflation bottomed summer 2023 and this can be seen in CPI which is effectively unchanged since the summer 2023, demonstrating the struggle of last mile inflation.
The problem with this argument is cherry-picking. Yes, annualized CPI is essentially flat since June 2023, but this is the only measure of inflation which shows this: PCE, Core PCE, and even Core CPI are all down from June 2023. Core CPI is down from 4.8% to 3.9% today, Core PCE is down from 4.1% in June 2023 to 2.8%, and even annualized PCE is down from 3% to 2.4% today (after declining in January as well). To argue that the disinflation process is bottoming in January is just incorrect.
Seasonal Adjustments, Survey Response Rates, & Revisions
Summers makes a note that would likely perk up the ears of any economist: “It’s always a mistake to over-interpret one month’s number — and that’s especially true in January, where calculating seasonality is difficult.” He is correct here. Not only is this correct, but the fed itself has noted this. In fact, one of the problems with relying on first-release data today is exactly this: seasonal adjustments & revisions play a large part. Seasonal adjustments will be made using pre-COVID or 2021 data when in reality we live in a novel world, as most will readily admit today. A theory for this is that with more people working from home, the seasonal layoffs that occur due to weather occur to a lesser extent. This affects employment, consumption, and inflation. Here is an article examining this. The takeaway is that January, just one month, certainly does not change the disinflationary trend. One month does not make a trend.
Another point to make here is that revisions to data are larger than in recent history due to the lowest response rate in recent history. Recent revisions to non-farm payrolls show that the prior two months’ (December & January) actually added ~167,000 less jobs than initially reported. The discrepancy is due to response rates. Response rates were above 60% pre-COVID, now we’re in the low-40s%. On the JOLTs, response rates are in the low-30s%! It’s impossible to use first-release data to guide policy expectations when the number of survey respondents is this low. This is one of the faults of survey-based measures.
Shelter Inflation
In the Bloomberg interview, Summers makes the argument that shelter inflation is hot, is not falling, and is a meaningful contributor to inflation. He says that because the CPI is not showing shelter disinflation that it’s concerning that inflation will not cool. This is one of the factors preventing inflation from getting below 3%, according to Summers.
Shelter has not fallen to levels that most real-time indicators are showing. The Cleveland fed actually introduced a real-time gauge of rents called the New Tenant Repeat Rent Index. This index and most alike it suggest that new rental inflation is coming down and it’s seen as a leading indicator to view the overall rent measure published in monthly official inflation data. Some argue that rent inflation will come down for this reason. We tend to agree, but the extent to which it disinflates is the bigger question: do we go from 6% today to 2-3% or 4-5%? History suggests that rent inflation should come in at around 3%.
While we could argue that rent inflation should come down due to market observed rent inflation coming down, we don’t necessarily need to make this argument. To make our argument, we just need to take Summers’ argument.
Summers says that “my co-authors at the NBER still looking for 3-4% Owner’s Equivalent Rent (OER) inflation through the remainder of the year, that’s 30% of Core CPI inflation.”
The problem with this argument is that if this is correct, then this is all that’s actually needed to bring Core CPI back to under 3%. Shelter at 3.5% and not 6% would remove as much as 1% from headline inflation. In other words, Summers estimates that inflation this year is where it’s been historically. It’s surprising that he is so concerned about shelter. Perhaps he is just waiting to see it in the official data (and so is the fed).
Today, shelter inflation is running at 6% according to the CPI and that’s down from its peak of 8.1% last summer. Being ~1/3rd of CPI (shelter as a category is 36% of CPI, OER is 26% of CPI), if OER were to fall from 6% to ~3.5% this year, about 100bps would come off the headline CPI measure. In other words, if the shelter reading on the CPI declined to where Summers estimates it’s running at, CPI would be nearly at 2% (2.1% using January’s reading of 3.1% on CPI).
Can Inflation Get to 2%? It already is at 2%
In this article, we’ve put “inflation” in quotations and we’ve done it for a reason. When market commentators say “inflation” they tend to mean CPI. When the fed says “inflation” they tend to tell which measure they use. Everyone knows that they prefer to use PCE. When we look at PCE, PCE is already near 2%. There has been no problem going from 3% in the PCE to 2.4%. Even with energy & oil prices up in February, this will only delay the eventual 2% reading. Energy is not up from last year other than a couple of upcoming months of base effects. WTI crude is between $70-$80 and unless that changes to the upside, energy inflation alone is not going to reverse disinflation.
Let’s repeat that for Larry Summers: today, “inflation,” as seen in the headline annualized PCE, is just 2.4%, down from 4.4% this time last year. Core PCE is already under 3% and that is before we see shelter meaningfully disinflate to where Larry Summers estimates it’s running at. If and when we see shelter disinflate, using these numbers, it’s possible that inflation move well under the 2% target.
On the idea that we are already at the fed’s target, this is true using month-over-month data but, I suppose, not quite true about annualized data. That said, most economists will admit that we are at or near the 2% target. In fact, the IMF did admit this in a recent study of inflation data. In this analysis, the IMF qualifies inflation being back at target as inflation readings coming within 100bps of reading before the inflation shock. Today we are 50bps away from our Pre-COVID inflation readings. In other words, we’re at target according to the IMF.
“We categorize an inflation shock as having been “resolved” if inflation falls to within 1 percentage point of its pre-shock rate by the end of our 5-year window (i.e., Π𝑇+5 ≤ Π𝑇−1 + 1%).” — IMF Authors
A note: this analysis flies in the face of Summers who has previously compared 2022 inflation to 1976. The IMF clearly demonstrates that that was a novel shock due to an inflationary oil shock. Today is very different than 1976 and Summers is not recognizing this.
Neutral Rate of Interest
The final argument that Summers makes is that the neutral rate of interest is too low. It has to go higher. The argument is based off the assumption that inflation cannot return to 2%. While we believe we’ve demonstrated why this argument is incorrect, we wanted to address this final aspect of the argument as well.
Keep reading with a 7-day free trial
Subscribe to FVCK THE FED to keep reading this post and get 7 days of free access to the full post archives.