CPI or Trumpian BS? Leaning to the latter

If you believe the numbers released by the Bureau of Labor Statistics today, you’d think INFLATION WATCH may be coming to an end. But anyone honest shouldn’t be taking these numbers at face value.

Inflation unexpectedly – and sharply – slowed in November, a seemingly welcome change for Americans weighed down by the persistently high cost of living.

However, economists were quick to caution Thursday that the Consumer Price Index slowing to 2.7% from 3% in September was likely the result of shutdown-related distortions of economic data.

“It’s hard to read too much into the November inflation data. The shutdown clearly had a big impact on data collection,” Heather Long, chief economist at Navy Federal Credit Union, wrote in a note on Thursday. “Inflation did not suddenly improve a lot between September and November. Anyone who has been to the grocery store or paid a utility bill knows this.”

And Long was far from the only person not buying that reported decline in inflation.

“I don’t take it at face value,” Stephanie Roth, chief economist at Wolfe Research, told CNN. “It seems like the government shutdown had a big impact.”

Or, as Wells Fargo economists quipped: “Take it with the entire salt shaker.”

With this in mind, I did a crude calculation of how much prices would have had to have changed in order to meet the 12-month change that was listed in the November 2025 CPI report. The archives of the older CPI reports are here, at least till they get pulled away.

The calculation I made is:

1. 12-month change in Sept 2025 CPI report (last one before the shutdown).

2. MINUS total change of Oct-Nov 2024 CPI, since that won’t be part of the 12-month Nov 2025 CPI

3. PLUS X = 12-month CPI change Nov 2025

For example, let’s look at food at home (“some may call it groceries”).

2.7% Sept 2025 year-over-year change
MINUS 0.6% Oct-Nov 2024 change
2.1% + X = 1.9% Nov 2025 year-over-year change.
-0.2% = X, the Oct-Nov 2025 price change.

Do we believe grocery prices dropped by 0.2% between September and November? As a person who got outside in those 2 months, I say no.

What’s interesting is that food away from home (some may call it “going out to eat”), went up by 0.5% between September and November. Are we to believe that the increase was all because of overhead, salaries and profits for restauranteurs? I’m doubting it.

Likewise, do we really think the rent and other housing prices didn't go up in October or November? That's what the CPI report was saying.

Also, let’s note that the typical survey dates are in early-to-mid November, while 2025’s CPI survey likely went a week or two later. Given what I remember my email inbox looked like in mid-to-late November, that may be an important difference.

Sam Tombs, chief US economist at Pantheon Macroeconomics, pointed out another wrinkle: “A higher proportion of price quotes than usual for November likely were sourced during the Black Friday discount period,” he said in a note to clients.

“November’s CPI data have to be treated cautiously, given that CPI data collection resumed only on the 14th after the end of the shutdown,” Tombs said.

And if inflation really is slowing down, then there’s a serious crunch going on with businesses in this economy, as S&P mentioned in their recent overview of the US economy.

Input cost inflation accelerated sharply in November, hitting the fastest rate for three years barring the jump in costs seen in May. Tariffs were again the predominant reason cited by companies for increased costs, alongside reports of higher wage rates. Service sector costs rose at the fastest rate since January 2023. In contrast, manufacturing input price inflation cooled to the lowest since February but remained well above the average seen over the past three years.

While higher input cost inflation fed through to a steeper rise in average prices charged for goods and services in November, competitive pressures restrained pricing power and meant selling price inflation remained below recent peaks. Overall, the increase in prices charged was the second lowest since April. Divergent trends were apparent at the sector level: selling price inflation slowed in manufacturing but reaccelerated in services.

So that likely will translate into a significant slowdown in profits for firms, and an increased risk of cutbacks on orders and layoffs. But hey, maybe an inability to sell stuff could keep inflation in check!

But let's hold off on the party hats on the inflation front. The nice version of how to take this November report is “let’s wait until December’s report to see if prices really are leveling off, as there could be a lot of snapback from an artificially low reading in November.”

The cynical version is something I hinted at in my post about the jobs report that dropped earlier this week.

This is the first report where I think TrumpWorld messed with the BLS numbers. Almost no specifics for Oct and Nov, and no way groceries are up less than 2% over the last 12 months, with barely any increase between Sept and Nov. And low CPI encourages lower interest rates. @mchinn.bsky.social

— jakemadtown.bsky.social (@jakemadtown.bsky.social) December 18, 2025 at 8:17 AM

With this crew, the dumbest and simplest reason is often the right one.