BUSINESS
Michigan Sentiment Falls as Households Mark Down the Economy
Michigan consumer sentiment fell to 51 as households marked down business hopes after already pricing high costs at home.
The University of Michigan’s consumer sentiment index fell to 51.0 in early August, a 7.6% drop from July that missed economist forecasts. Households barely changed how they rated their own money, then marked down the business outlook for the year ahead and the longer haul.
Joanne Hsu, director of the school’s Surveys of Consumers, said people are still bracing for their pay to lose ground to prices. The early reading lands after July retail sales fell for the first time in nine months, which is the lag the monthly mood score has been slow to show.
The Preliminary August Reading Is 51.0
Michigan’s preliminary August reading of 51.0 compares with 55.2 in July and 58.2 in August 2025, a 12.4% decline from a year earlier. Economists in a Reuters poll had looked for 54.5, and a Bloomberg survey had centered on 55. Both of the index’s main parts weakened, and the expectations side fell harder than current conditions.
| Index | August 2026 | July 2026 | August 2025 |
|---|---|---|---|
| Consumer sentiment | 51.0 | 55.2 | 58.2 |
| Current economic conditions | 51.8 | 54.8 | 61.7 |
| Consumer expectations | 50.6 | 55.4 | 55.9 |
Current conditions are down 5.5% on the month and 16.0% from a year earlier. Expectations are down 8.7% on the month and 9.5% from last August. The final August figure is due Friday, August 28, at 10 a.m. Eastern.
Trading Economics puts the long-run average of the series, which starts in 1952, at 84.34, with a high of 111.40 in January 2000. Advisor Perspectives, which tracks the same history, said a print of 51 is below the 1st percentile of the record and below the index’s level at the start of all six recessions since it began.
THE 2026 PATH BACK TO 51
- February 2026: The Iran conflict starts, and Michigan later compares this month’s 3.4% year-ahead inflation reading with everything that followed.
- May 2026: The final index hits an all-time low of 44.8, under the previous floor of 50 in June 2022.
- June 2026: The index rebounds to 49.5, about 10% above the May low, according to the Federal Reserve Bank of St. Louis FRED series.
- July 2026: The final index rises to 55.2, the high for this bounce.
- August 14, 2026: The early August index falls back to 51.0, and Census issues the July retail report the same morning.
Hsu has said a modern reading near 50 is not the same object as a 50 from decades ago, because the index is built to track the trend. That trend has been flat or down since early 2025, and the June-July bounce did not take the series far from the floor it set in May.

Personal Finances Held While Business Outlook Sank
The split inside the August survey is the move that the headline 51.0 conceals. Views of personal finances saw only minor declines, Hsu wrote in the release, while expected business conditions sank 11% for the short run and 17% for the long run.
Consumer sentiment fell about 8% this August, ending two consecutive months of improvement. While views of personal finances saw only minor declines, expected business conditions sank 11% for the short run and 17% for the long run.
Joanne Hsu, Director, University of Michigan Surveys of Consumers
In interviews around the release, Hsu said a solid majority of respondents brought up high prices on their own when they talked about their personal finances, a share that rose to 56% this month from 50% in July. She also said households had already folded higher gasoline costs into those assessments over the past few months, so the fresh damage showed up more in how they rated the wider economy than in how they rated their own books.
“People are still feeling relatively negative about the economy,” Hsu said. “And, specifically, just very few consumers are expecting any wage gains, income gains, in the next year to outpace inflation. So, they are bracing for further erosion of their purchasing power in the year ahead.” She added that real income expectations are softer than in the recent past, and that this “feeds into this idea that consumers just don’t feel like they’re thriving right now.”
That is a second turn, not a first shock. The price complaint was already in the answers. What fell hard in August was the bet on business conditions, the channel through which a long sour mood eventually shows up in orders, hiring, and the checkout lane.
July Retail Sales Posted the First Drop in Nine Months
The Census Bureau’s advance report put July sales of $763.6 billion for retail and food services, after seasonal adjustment but not after price changes. That is the cash-register print that arrived the same morning as the Michigan survey.
https://x.com/uscensusbureau/status/2088251668957647146
THE JULY SALES PRINT
- Month-to-month: Sales fell 0.6% from June, with a margin of ±0.4%, the first decline since last October in the Reuters account of the same report.
- From a year earlier: Sales were still up 5.0% (±0.5%) from July 2025, and the May-through-July total was up 6.3% from a year earlier.
- June revision: The May-to-June gain was left at 0.2% (±0.3%).
- Underlying demand: Reuters said sales excluding autos, gasoline, building materials, and food services fell 0.4%, against a forecast rise of 0.3%.
The Associated Press wrote that households had spent through the World Cup and Amazon Prime Day, then cut back in July by the most since May 2025. CNN said gasoline-station sales fell 0.9% last month as energy prices eased, and that the rest of the ticket was still down 0.6% even after that drag. Some economists told Reuters that consumer spending, more than two-thirds of the economy, could slow to a pace under 2% annualized in the third quarter after a 3.2% pace in April through June.
The Bureau of Economic Analysis, which measures a broader basket than the Census retail report, had consumer spending up 0.3% in June and 0.9% in May. Those monthly gains are why the sour surveys of 2022 through early 2026 did not, on their own, prove a household pullback. July is the first hard print in this cycle that rhymes with the mood.
The Michigan Survey Lost Its Tie to Spending
A June Chicago Fed Letter by Scott Brave, a senior economist at the Federal Reserve Bank of Chicago, and colleagues, documented a near-zero link to real spending for the Michigan index over the past 10 years of data. Before 2020, the rolling 120-month correlation with annual real personal consumption growth averaged 0.69 for Michigan and 0.6 for the Conference Board. After 2020, that link fell sharply, and the drop was largest for Michigan.
WHY THE SPENDING LINK BROKE
- The survey shift: Michigan moved the Surveys of Consumers fully to the web in April 2024, which raised the sample size and, in work cited by the Fed letter, lowered the average respondent’s optimism.
- The vibes gap: Commentators have called the other break a “vibes” story, people feeling worse than hiring, spending, and output implied, and the Fed letter says that gap is still in the Michigan series.
- A composite fix: The authors built a Composite Consumer Sentiment Index that blends Michigan, the Conference Board, and daily polls, and they say it tracks spending more closely than Michigan alone.
Using the Michigan index, the letter forecast annual real spending growth of 1.3% by October 2026, six months after the April data then in hand. The same exercise with the composite index produced 2.0%, close to the 2.1% pace of real spending in the 12 months through April. Both paths are slower than the recent past. The Michigan path is the weaker one.
The Conference Board’s separate survey, which samples a different mix of questions and people, has been less gloomy. Its confidence index eased to 90.8 in July from a revised 92.2 in June, according to the group’s July 28 release, with the present-situation gauge at 114.9 and the expectations gauge unchanged at 74.7. Dana M. Peterson, the Conference Board’s chief economist, said confidence has been on a downward slope since late 2021, and that the expectations index stayed in negative territory. That is a cooler instrument than Michigan’s 51.0, and it is one reason the Fed letter treats a single survey as a noisy guide to the register.
Only 8 Percent Expect Pay to Beat Inflation
The durable line in Hsu’s August note is not the 4.2-point drop in the headline index. Across all consumers, only 8% expect their income growth to exceed inflation in the year ahead, down from 18% in December 2024. Hsu called that a belief that high prices will keep weighing on households.
WHAT HOUSEHOLDS SAY ABOUT PRICES
- Year-ahead inflation: Expected inflation ticked up to 4.3% from 4.2% in July, above the 3.4% reading in February before the Iran conflict and above every 2024 figure.
- Long-run inflation: The five-to-ten-year reading held at 3.3% for a third month, a bit above the 2024 range of 2.8% to 3.2%.
- Official CPI: The Bureau of Labor Statistics said prices up 3.4 percent over the year ending in July, after 3.5% through June, with the core index up 2.5%.
- Energy in the CPI: Energy rose 14.7% over those 12 months, and gasoline rose 24.6%, even after a 2.9% drop in gasoline in July itself.
The national average for a gallon of gas was $4.07 on the Friday of the Michigan release, according to AAA, which is $1.09 more than the national average the day before the war with Iran started in late February. Food in the CPI was up 3.0% over the year, and shelter was up 3.2%. Core inflation at 2.5% is closer to the Federal Reserve’s 2% goal than the headline is, and it is still above that goal. Hsu’s respondents are not treating that gap as closed. They are treating it as a year of pay that fails to catch prices.
Todd Belt, a professor of political management at George Washington University, noted that inflation has eased and is still above 3%, and that purchasing power has not kept pace. The survey’s 8% figure is the household version of that same claim, stripped of party labels.
Older and Lower-Income Adults Took the Hit
Hsu wrote that the early-month weakening ran across political and other groups, with the largest reductions among older consumers, lower-income consumers, and those without a college degree. Those groups, she said, are particularly exposed to any loss of buying power from inflation. Decreases showed up across the political spectrum, with Republicans posting the strongest month-to-month decline. Sentiment among Republicans is now 19% below readings just before the Iran conflict and the lowest since the 2024 election.
WHO THE AUGUST DROP HIT HARDEST
- Older adults: Hsu flagged this group for a notably large reduction, a cohort that spends a larger share of income on fuel, food, and care.
- Lower-income households: The same note tied their drop to vulnerability when prices take a bite out of pay.
- Adults without a college degree: They posted a sharp decline alongside the other two groups.
- Republican identifiers: They had the steepest month-to-month fall and now stand 19% below their pre-conflict readings.
Belt said those groups had backed Donald Trump on a promise to turn the economy around, and that tariffs and the Iran war have instead kept the inflation problem alive. “When it really starts to affect your pocketbook, when it’s really, really getting to the nitty-gritty and you can’t make ends meet, you just can’t out-message that,” Belt said. He argued that discouraged Republican voters are more likely to stay home in the fall than to switch parties, while Democrats look more fired up to vote.
That turnout claim is a political read of a household survey, and it is the frame most of the first-day coverage used. The survey’s own table is narrower. The people who lost the most ground in August are the ones Hsu called most exposed to prices, and they are the ones whose pay-versus-inflation math is already in the 8% statistic.
A Record Market and a Floor-Level Survey
The live argument after the print was not whether Americans feel bad. It was whether that gloom had reached the register, and whether it was already in the price of stocks. The 8% share who expect pay to beat prices is the figure that kept getting repeated, more than the 7.6% monthly drop in the headline index. A prediction-market post the same afternoon put a 23% chance on a U.S. recession by the end of next year, well below a coin flip, which is another way of saying traders have not treated 51.0 as a hard-landing signal.
That split is the same gap the Chicago Fed letter was written to explain. Michigan can print numbers that look like recession starts, and spending can still grow, until a month like July shows up. The July sales drop is one month, it is not adjusted for prices, and it still leaves sales 5.0% above July 2025. It is also the first time in this stretch that the cash register moved in the same direction as the business-outlook collapse Hsu recorded in August.
The early August index will be revised on August 28. What will not be revised is the 8% who expect their income to outrun inflation, or the 11% and 17% markdowns in expected business conditions after personal finances had already stopped falling much. Those are the second-order readings, and they are the ones July’s receipts just started to match.