The Mean Core
And what the demographic cliff has to do with it
About a month ago I wrote that US Headline inflation has likely peaked, and that it would decelerate meaningfully in its month-to-month comparisons. Indeed, after clocking 1.3% for June, estimates for July Headline CPI now sit at 0.3% vs the prior month, driven by a steep decline in gasoline prices
As a consequence, financial markets have rapidly priced out inflation risk from many assets, from bonds to crypto or equities. One might argue the market’s view is now that the issue is “Over by Christmas”, akin to what European soldiers were told at the outbreak of the Great War in 1914
However, this conclusion may once again be premature. Recent data indicates continued pressure on US wages, which likely provides fuel for the stickier parts of inflation, as measured in the Core CPI
This post walks through the reasons behind this dynamic and links them to the demographic debate. As usual, the post concludes with my current views on market, where I explain why I am exiting bonds again, in light of said recent data
To start, let’s recall that Inflation can be broken down into two components:
The Headline number, which is heavily influenced by volatile food and energy costs
The Core number, which is mainly driven by Services such as rent, health care or public utilities
The Core is often described as containing the sticky parts of inflation, those that move up and down with much inertia, akin to a tanker at sea that takes time to turn
There are many ways to measure this inertial component of inflation. Aside of the Core, economists also look at the Mean (=average inflation across all components), the Trimmed Mean (=same, excluding outliers) or Sticky CPI (=slow moving components only)
The Atlanta branch of the Federal Reserve provides data on all these measures. And one can pick any of them, they all point steeply upwards:
Ok, so what? Now - as a regular reader you know that I follow a data-dependent approach
And the data continues to tells me, as laid out in the inflation post a month ago, that Goods inflation is coming down. Why? Commodity prices from gasoline to copper or aluminium are declining, and retail inventory grows
For Services however, which cover 60% of the CPI and 75% of the Core CPI, commodity prices are not informative. Here, we need to look at wage growth. Labor is the biggest cost driver for Services
And wage growth data over the past month has not been good at all (for the purposes of inflation cooling off, that is)
The Employment Cost Index by the Bureau of Labor Statistics, deemed the highest quality source on wage growth, accelerated from 1.2% in Q1 to 1.4% in Q2 (It’s a quarterly survey, so ~5-6% annualised)
The Atlanta Fed Wage Tracker accelerated to 6.7% in July
Average Hourly Earnings, as measured in the BLS monthly NFP unemployment survey, accelerated to 0.5% for July, or ~6% annualised
The same survey showed record low unemployment, and 528k jobs created in July. The NFP survey is prone to revisions, but even adjusting for that it was a very strong print
Of note, the number of job openings as measure in the JOLTS survey (to mid June) declined from 11.3m to 10.7m. However, more recent data by Indeed through end-July suggests job openings remain high
Finally, Withheld Income Tax, a very noisy data series, which however can be tracked daily, points to steady growth over the past three months, at around 10% year-on-year
Summary: Recent data indicates wage growth remains high, some sources even suggest it accelerated. This likely provides upside pressure to Core CPI - the sticky part of inflation that is much harder to tame
What’s behind this? Looking at US employment data in more detail, we notice something puzzling:
Today, the same number of people are employed as pre-COVID. However, in 2019, the labor market was far from being as “red hot” as it is today
What happened? The answer has to do with Demographics
COVID-19 put the turbo into a development that was already on the cards - the retirement of Baby Boomers
Millions of over 55 year olds retired early, and have not returned to the labor market since
The Boomers were plenty. And as they exit, they are replaced with Gen Z, which are much fewer
We are only in the early innings of the Boomer retirement cliff, so this development will be with us for the next decade
Contrary to expectations, the demographic cliff turns out to be inflationary. It tightens the labor market, which drives up wage bargaining power for workers1
The current macroeconomic dogma is one of an aging population as a deflationary force. The main reason cited is lower demand vs an historically larger production capacity.2 Japan is typically referenced as evidence
What many miss is that Japan’s deflationary period coincided with globalisation, which allowed Japanese companies to outsource abroad (i.e. China), instead of paying up domestic wages. Also, the Japanese economic model relies on very close ties between employer and employee. In crisis times, companies go out of their way to avoid layoffs (hence the constant, very low Japanese unemployment). In turn, employees hold back with wage demands3
Until the mid 2010s, the growing shortfall was compensated by immigration. However, after 2016, Trump slammed the breaks on immigration, and the Biden administration does little to reverse it
This is a reflection of the political climate. Globalisation and immigration was great for corporate profits, but hollowed out the Western working class4. The message has now resoundingly arrived in Washington
Not only Boomers and immigrants are missing from the labor market. Another big group also hasn’t returned in full - women
In particular, women from lower-income groups. They face steep childcare cost inflation, which keeps them at home for lack of alternative. This is again, in parts, linked to immigration
Even if immigration remained high, Boomers’ workplace competence, honed over decades, is not easily replaced
Take Germany as an example, where demographic trends are worse (but not the worst). The country currently faces the biggest economic crisis in decades. Yet, there is still no rise in unemployment. More so, the shortage of skilled labor is at an all time high!
Summary: The Demographic Cliff is one of the main drivers for the very tight labor market. This development likely intensifies over the next decade, as more Boomers retire
If we take all of the above together, it is easy to conclude that unemployment and thus Core inflation will likely evolve very differently during the coming recession (yes, there will be one)
Many companies still struggle to fill roles, and some sectors remain severely understaffed, such as Hospitality or Teaching
With regards to Public Services more broadly, they are in a tough spot, needing to compete for scarce talent
As a consequence, US States need to raise wages. Full State treasury coffers permit this after a surge in tax intake. Expect more, similar headlines:
While there are certainly layoffs in parts of the economy (e.g. Tech), these workers - for now - get rapidly absorbed elsewhere:
Summary: The job market likely remains firmer for longer, and would likely require a severe recession to drive up unemployment and to break the inflationary price-wage spiral
It is just common sense - if a company struggles to find the right talent, they will hold on to everyone they have for longer
So something else has to give - corporate profit margins, which are on historical highs
Finally, what about commodities? Hasn’t any commodity slowdown in the past also preceded a slowdown in Core Inflation? They are lead indicators that shows that the Fed has done enough, and the medicine is working?
This is a very important point. If we take a step back, we know there is way too much debt in the world, aided by a decade of low interest rates. So higher rates should indeed slam the brakes on the economy
However, much of this debt is outside of the US, while the US Private Sector has delevered5. This is what caused inflation in the first place. I’ve shown the chart below several times, it is an important one:
So as the Fed increases interest rates, it crushes the overlevered Rest of the World first, from Emerging Markets like Pakistan or Sri Lanka to Europe and China’s Real Estate bubble. The US Private Sector determines the pace of tightening, however it is also the most resilient against it
Commodities are global goods. Just take oil as an example, 75% of its demand is outside the US. In my view, the steep commodity price decline reflects an asymmetric global slowdown, where the Fed’s tightening measures hit the Rest of the World much harder than the US, in addition to local issues (Europe gas, China zero-Covid)
What does this mean for markets?
In conclusion, the analysis laid out in this post leads me to believe that Core inflation will be much harder to defeat than consensus currently believes. The market has - in my view - become too complacent on inflation risk, and as such vulnerable to negative surprises
This translates into my following views, which may as always be wrong:
Bonds - I continue to believe that Bonds yields are in the part of the cycle that drives them lower. The economy slows and the Fed fights inflation. However, the complacency around recent wage growth data leads me to believe that a poor Core CPI print might shock yields upwards, to test recent highs once more. This would be lead by the short end, but also push up the long end. I am therefore taking profits here on TLT and ZROZ after a decent run, and would look to re-enter (1) if they reset to a higher level, or (2) if I am wrong and Core inflation moderates further (in which case I miss out a bit)
Equities - I remain short with a generous stop, acknowledging that there may still be overly bearish positioning that needs to be cleared, while at the same time signs of FOMO are now quite evident. In fact, the return of retail-FOMO trades in AMC, Gamestop and BBBY is an indication that long-term bond yields are too low. Returning to Core CPI6, if it surprises, then bond yields go up which will hurt equities. If it doesn’t, equities might rally, however eventually a decline in earnings should catch up with them (see NVidia profit warning today). Finally, pricing in a declining CPI with corporate earnings staying high at the same time is, in my view, the current Schrödinger’s Cat of financial markets, i.e. a paradox that eventually gets resolved as only one can be true at the same time
I hope you enjoyed today’s Next Economy post. If you do, please share it, it would make my day!
This leaves aside other drivers such as deglobalisation or re-shoring
Many other reasons in addition, from excess savings to lower consumption in age. Details on the demographic debate will be the topic of a future post
More on this in Goodhart’s and Pradhan’s excellent work “The Great Demographic Reversal”
This was also reflected in stagnant real wages for lower income groups, while top-quintile wages grew fast. Some observers expect this to reverse now, I have sympathy with that view
With this, please bear in mind that the lowest quintiles have reverted to their pre-COVID position. However, their share in consumption is smaller
Published this coming Wednesday
Great post. One quick question - probably a dumb one:
"However, much of this debt is outside of the US, while the US Private Sector has delevered5. This is what caused inflation in the first place."
Why has the de-levering caused inflation in the first place? I'm struggling to make the connection.
I assume what you're saying it's due to government transfer payments into the pockets of corporates and consumers? But then if they used that money to de-lever they wouldn't have used it on consumption so why would it be inflationary?
Sorry if you previously talked about it and I missed it.