Science1 distinct publisher3 min readPublished
The University of Michigan's August survey puts sentiment at 51.7 with declines in every political group, and the current-conditions reading has fallen far enough to close its usual gap with expectations.
The Scientist · Science desk

invest
Only 8% of Americans expect income to beat inflation, and pricing power is the casualty1 distinct publisher
science
Flow control gets a shared benchmark, and a 38% friction cut nobody had to simulate first2 distinct publishers
leadership
Mamdani is buying credibility in asphalt before he asks New Yorkers to fund the hard part1 distinct publisher
product
The reply is the ranking signal, and the reply is where disagreement lives1 distinct publisher
Compiled by The ScientistSomething wrong?How this is made
The distance between this release's two halves is more informative than the headline index itself. Last August the Current Index stood at 61.7 and the Expectations Index at 55.9, a gap of 5.8 points [2][3]. This month they read 51.9 and 51.5, a gap of 0.4 [2][3][8]. Across twelve months the current-conditions reading fell 15.9 percent against 7.9 percent for expectations [7]. The present-tense assessment did most of the work of the annual decline, and expectations have caught down to it. Month over month the order reverses: expectations off 7.0 percent, current conditions off 5.3 percent [9]. An index 11.2 percent below a year ago is doing something less interesting than that convergence [11].
On mechanism, U-M economist Joanne Hsu points to policy uncertainty including the Iran conflict, with consumers anticipating further gasoline price increases in both the short and long run [12]. The release adds that any reescalation of trade tensions would likely worsen the trend [13]. Both are stated attributions inside a survey of opinion. The instrument asks people what they expect; it does not run the counterfactual in which fuel prices hold flat, so it cannot separate a fuel-price read from a general loss of confidence.
The breadth claim deserves the same care. The release reports declines across all political groups, particularly acute among Republicans [4]. On income it does not say every band fell; it says declines were stronger among older consumers, lower- and middle-income consumers, and those without a college degree [5]. Direction looks common wherever the release reports direction, but magnitude varies sharply across groups, and nothing here decomposes how much of that is a partisan re-rating versus a shared price shock landing hardest on thin household balance sheets.
What people fear has also reordered. In January, 23 percent named inflation and 14 percent unemployment as the more serious hardship; in August it was 36 percent and 6 percent [6]. The ratio moved from roughly 1.6 to 1 to 6 to 1 [10]. That is a change in perceived risk, not a measurement of the labour market, and it will shape how a price increase reads to a customer relative to a layoff headline.
What this leaves unanswered is what any of these households will actually do. Fifty-three percent volunteered that high prices are weighing on their current finances, up eight points since January [14][15], and only 10 percent expect their purchasing power to rise over the year ahead [16]. Hsu says a growing majority now expect inflation to outstrip income gains [17]. None of that is transaction data. Treat the August release as a well-measured attitude series with an unusually broad base, which justifies stress-testing a weaker-demand case for next quarter rather than booking one.
Ranked by verification strength, evidence, and original report placement.
The University of Michigan Consumer Sentiment Index fell to 51.7 in the August 2026 survey, down from 55.2 in July and below last August's 58.2.
The Current Index fell to 51.9, down from 54.8 in July and below last August's 61.7.
The Expectations Index fell to 51.5, down from 55.4 in July and below last August's 55.9.
Sentiment declines in August were seen across all political groups and were particularly acute among Republicans, according to the University of Michigan Surveys of Consumers.
Groups typically less equipped to absorb increases in the cost of living exhibited stronger declines in sentiment, including older consumers, lower- and middle-income consumers, and those without a college degree.
Asked which factor will present more serious hardship, 23% of consumers named inflation and 14% unemployment at the beginning of the year; this month the share naming inflation climbed to 36% while 6% selected unemployment.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · August 28, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One issuer, but the arithmetic holds
Everything on this page traces to a single University of Michigan release relayed by phys.org — and to the release's credit, it prints the raw index levels next to its own summary, so the numbers can be checked against each other. They survive it: 51.9 against last August's 61.7 is the 15.9% fall, and the 5.8-point cushion current conditions held over expectations really does shrink to 0.4. The collection window is stated. What is absent is anything from outside the issuer — no sample size, no error bands, no second gauge to place 51.7 against, and no spending data to test whether the mood shows up in behaviour.
Attitudes measured, behaviour unobserved
A sentiment survey records what people say, and this reporting stops at the saying. Nothing here follows the 53% who volunteer that high prices are squeezing them into actual spending, nor shows markets, retailers or policymakers reacting to the August print. A survey's own publication is not uptake, so we are not scoring one.
The issuer undersells its own table
The release rounds itself down — 'about 6%' for a 6.3% monthly drop, 'around 11%' for 11.2% — and skips the most arresting item in its own data: current conditions have fallen to within 0.4 points of expectations, from 5.8 points a year ago. Against that restraint sits one flourish, the line that renewed trade tension 'will likely exacerbate these trends,' asserted with nothing behind it. Net, the framing is quieter than the numbers.
University press release, republished intact
phys.org reproduces the U-M announcement essentially as written — headings, quotes and sequencing — so every framing choice here belongs to the survey's own communications, not to a reporter weighing it. The pull is limited by what the Surveys of Consumers is selling: decades of methodological continuity rather than any particular month's reading. The exception worth flagging is the partisan line, singling out Republicans, a detail that travels widely while arriving with no subgroup numbers attached.
Trust the levels, not the causal story
The index readings are close to primary — the surveys' director is named and quoted, the interview window is stated — so 51.7, 51.9 and 51.5 are safe to work with, as are the derived percentages. Confidence drops as the claims move away from the table: that Iran-driven gasoline expectations and cost-of-living pressure explain the fall rests on Joanne Hsu's reading alone, and with one publisher carrying the story there is nothing to triangulate.