A lot of things about this current economy don’t make sense.
The University of Michigan Consumer Sentiment Index posted a final May reading of 44.8 on Friday — a record low, down five points from April, with the survey revised lower from a preliminary 48.2 and 57% of respondents spontaneously volunteering “high prices” as the thing eating their finances. The financial press ran with it. Recession imminent. Consumer breaking. Sell something. Sell everything.
The market reacted with a bit of a yawn so far this week. Why?
The UMich Index is Not Enough
Turns out the UMich Index has been a political tell more than a consumption signal for the last few cycles — Dem and Rep sub-indices diverged by historic margins post-2020, and the headline whipsaws on partisan composition more than on actual spending intent. Treating it as one input among many is the right call.
Three other indicators printed recently. They disagree with UMich. Loudly. And yet we’ve heard very little about them. “Sentiment” is a highly political measure - both by those doing the measuring and by those being measured.
To get better idea of what’s going on the economy, its worth looking at these other measures. Here they are:
The Conference Board’s Consumer Confidence Index
Conference Board CCI is more labor-market sensitive — its “jobs plentiful vs. hard to get” spread is one of the cleaner recession leads we have, and it tracks payrolls/UR better than UMich tracks PCE. There is less political contamination because the questions are more concrete.
Finding: The CCI dropped this week at 93.1, down a rounding error (0.7 points) from April and beating the 92 estimate. The Present Situation component softened — fair enough, gas prices, geopolitics — but the Expectations sub-index actually rose 1.0 point to 74.4. Forward-looking consumers are getting less worried, not more.
The New York Fed’s Survey of Consumer Expectations
NY Fed SCE is the inflation expectations workhorse — 1yr, 3yr, 5yr inflation expectations plus labor market expectations (job loss probability, reservation wage, finding-a-job probability). This is what the Fed actually watches, so it matters for policy reaction function.
Finding: The New York Fed’s Survey of Consumer Expectations, the survey the FOMC actually anchors its inflation-expectations math against, has one-year inflation expectations at 3.6% and five-year at a stable 3.0%.
UMich’s five-year inflation expectations? 3.9%.
That’s a 90-basis-point gap between two well-resourced national surveys, run on the same population (sort of), in the same month.
The NFIB Small Business Optimism Index
NFIB is the supply side of the same coin — small business sentiment, hiring plans, compensation plans, single most important problem (inflation vs. labor vs. taxes vs. sales). Hiring Plans and Compensation Plans lead BLS wage data by ~2 quarters. The “Uncertainty Index” sub-component is underrated.
Finding: The NFIB Small Business Optimism Index, meanwhile, ticked up 0.1 to 95.9 with the Uncertainty Index dropping four points and seven of ten subcomponents improving including a six-point pop in firms reporting higher earnings.
Sure my expanded panel covers different things than UMich, but that is the point. The divergence in outlook is fact and it’s actually a textbook case for why the four-indicator panel makes sense.
The findings side by side:
That 90bp gap between UMich and NY Fed on 5yr inflation expectations is the tell. Same economy, same month-ish, two well-resourced surveys, completely different signals. The NY Fed survey is anchored — that’s what the FOMC reaction function watches. UMich is unanchored and screaming “I hate Trump.”
Not Regime Shift, But Gird Your Loins
UMich is the noise.
CCI Expectations + SCE long-end + NFIB Uncertainty is the signal.
There is not regime shift, yet.
We are also all feeling some serious price-level frustration. That’s the rational response of a US population that just watched gasoline rip on Hormuz headlines straight to the gas pump. That’s all I’m picking up in the UMich Index. CCI is labor-sensitive. SCE is what many in the Fed watch. NFIB is the supply side of the same household — small-business owners. All three say the same thing: fragile and stressed, but not broken, yet.
That’s one of the reasons why the curve isn’t pricing emergency cuts off the UMich number.
The one thing to walk away with:
The consumer is sick of high prices, but not broken. The distinction can make you some money this week.
If you position for “broken” - pile into defensive staples, dump cyclicals, buy duration on the expectation of a Fed panic — you’re trading off the noisiest, most politically contaminated read. Maybe you’re early. But the other three surveys, looked at together, describe an economy that hates the gas pump and is deeply unenthused about prices, but isn’t completely shutting the wallet or changing plans.
Watch the other sentiment indicators above in the coming weeks. If all three start rolling over together, then the regime is shifting. Right now? One is rising, one is stable, and one’s uncertainty is falling.
Hope that helps.
— Matthew
Not investment advice. Macro datapoints sourced from public releases: U. Michigan Surveys of Consumers (sca.isr.umich.edu), The Conference Board (conference-board.org), Federal Reserve Bank of New York Survey of Consumer Expectations (newyorkfed.org/microeconomics/sce), NFIB Small Business Economic Trends (nfib.com). Reader is responsible for their own due diligence.