US consumer sentiment fell to 46.3 in the University of Michigan’s preliminary October 2026 survey, from 48.1 in September. Households also raised their year-ahead inflation expectation to 4.7%, from 4.6%. The combination leaves consumer-facing businesses with a difficult planning problem: shoppers feel squeezed, while lenders and investors still need evidence of how that pressure affects purchases and repayment. Michigan’s October results establish the deterioration in confidence; they cannot establish a decline in October spending.
The latest spending figures tell a different, earlier story. In August, Americans increased inflation-adjusted consumption by 0.6% from July, according to the Bureau of Economic Analysis. That gap between attitudes and recorded activity deserves attention. Businesses that treat a gloomy survey as proof of an immediate sales collapse could cut too much inventory. Businesses that extrapolate a strong summer spending month could miss customers beginning to postpone larger purchases.
A weaker assessment of the present
Michigan’s current-conditions index dropped to 44.7 from 50.9, while its expectations index rose to 47.3 from 46.3. The headline therefore combines a worse assessment of today’s economy with a small improvement in the outlook. Survey director Joanne Hsu linked poor durable-goods buying conditions to prices and borrowing costs, and reported weakness among lower-income households and consumers with smaller stock portfolios. Long-run inflation expectations increased to 3.5% from 3.4%. The university’s release supports these details.
For a retailer, the distinction between present conditions and future expectations changes the questions to ask. Customers might believe their income will improve yet decide that a refrigerator or car costs too much today. An improvement in expectations would then offer little help to this quarter’s sales. A finance provider could face weaker demand for new loans even before existing borrowers experience payment trouble.
The index is a measure of attitudes. Its level has no direct conversion into dollars of lost consumption, a delinquency rate or a probability of recession. Analysts need transaction and loan-performance data to make those assessments. They should also distinguish an index-point change from a percentage-point change in an inflation expectation; the two measures describe different things.
Spending growth complicates the slowdown story
BEA’s August income and outlays report, released September 30, showed current-dollar consumption rising 0.9%. Disposable income grew 0.3% in nominal terms and was unchanged after inflation. The personal saving rate was 4.1%. The PCE price index rose 3.4% from a year earlier, or 3.0% excluding food and energy.
Those figures require careful timing. August spending predates the October interviews, so it cannot settle whether households have since retrenched. It does demonstrate why confidence and spending need separate treatment. A household may resent higher prices while continuing to pay for necessities, replace a broken appliance or honour a planned trip. Total expenditure can remain firm while satisfaction with purchasing power falls.
The nominal and real measures also answer different business questions. A supplier receives nominal revenue, but revenue growth caused by higher prices can coexist with weak unit demand. A seller whose costs rise faster than its selling prices may face a margin squeeze even if receipts increase. Investors assessing consumer companies should compare volumes, discounts and gross margins alongside sales growth.
Flat real disposable income in one month provides a reason to examine household buffers, without proving that consumers have exhausted them. Aggregate saving figures cannot identify the cash available to a particular borrower. A lender needs income verification, liquid balances and payment history before translating the macro data into an individual credit decision.
Consumers may change the mix of purchases
A separate Michigan study, Gasoline Prices and Expected Consumer Spending, published October 9, gives more detail on intended responses to expensive goods. Its Table 1 reports that 31% expected to buy as usual, 54% to cut back and 16% to stop buying items with large price increases. Rounding explains why the shares exceed 100%. The study used 3,492 interviews conducted June 23 to September 21, rather than the preliminary October sample.
Table 2 shows a substantial income divide: 20% of the lowest income third expected to spend as usual, compared with 42% of the highest third. These are intentions about specific items, rather than measured changes in total expenditure. The report also cautions that consumers can substitute between products.
That distinction gives businesses a practical way to test the survey’s implications. A grocery chain could examine whether customers move toward cheaper brands, change package sizes or respond more to promotions. A durable-goods seller could track quote requests that fail to become orders. Each measure would reveal a different adjustment, with a different consequence for revenue and working capital.
Suppliers should avoid assuming that an economy-wide change affects their customer base evenly. A premium service aimed at households with ample liquid assets may face a different demand path from a retailer serving customers with little room after rent and food. The relevant exposure depends on who buys the product, the purchase’s urgency and the customer’s financing needs.
The Federal Reserve faces competing signals
The Federal Reserve raised its policy-rate target by a quarter percentage point to 3.75%–4.00% on September 16. In its FOMC statement, the committee described domestic spending as resilient and inflation as elevated. It reaffirmed its 2% inflation goal. That is the documented policy backdrop to the October survey, rather than a forecast of the next decision.
For monetary-policy analysis, weak confidence and higher expected inflation can pull in different directions. Slower purchases could reduce demand pressure. Households expecting higher prices might seek wage increases or bring some purchases forward, although the survey does not establish either behaviour. Officials would need to assess actual inflation, employment and spending before deciding which channel dominates.
An expectation of 4.7% is also different from an observed inflation rate of 4.7%. Respondents describe their outlook; statistical agencies measure price changes using defined expenditure baskets. Treating the survey figure as an official inflation reading would exaggerate what the release shows. Likewise, a small monthly increase alone cannot prove a lasting loss of confidence in price stability.
Businesses planning financing should therefore consider more than one rate path. A project that works only if borrowing costs fall soon carries a different risk from one that can service debt at its present rate. The October confidence reading offers a reason to revisit assumptions, while the September policy statement explains why easier financing cannot be taken for granted.
Analyst’s View
The following implications are analytical judgments drawn from the verified releases. They are conditional exposures, rather than evidence that credit losses or market repricing have occurred.
For consumer credit, the priority is to distinguish weaker loan demand from weaker repayment capacity. Borrowers who postpone a major purchase may take on less new debt. Borrowers who keep spending while their financial buffer shrinks may become more vulnerable to an income interruption. Lenders can track utilisation, missed payments and hardship requests by income and product, then compare those trends with the portfolio’s usual seasonal pattern.
For corporate credit, consumer exposure should be assessed through both earnings and cash collection. If sellers offer more discounts while holding expensive inventory, operating cash flow could deteriorate before headline revenue does. Suppliers extending trade credit should examine customer concentration, receivable ageing and refinancing dates. A stress scenario can combine lower sales volumes with slower collections, without assigning an unsupported probability to that outcome.
For market positioning, a weak sentiment number alone provides insufficient grounds for a broad allocation decision. Investors can ask which issuers depend on discretionary purchases and which have the liquidity to withstand softer orders. Rate-sensitive exposures also need separate inflation and demand scenarios: slower growth can support some bond prices, while persistent inflation can keep yields elevated. The survey does not resolve that trade-off.
The next evidence to watch
Michigan’s methodology FAQ says preliminary readings target about 420 interviews, with the exact count varying; final monthly readings typically use about 1,000. These are general sample targets, not the verified October interview count. The university lists October 23 for the final October release. Revisions can change the month-to-month comparison, so analysts should retain the preliminary label.
BEA schedules September income and outlays for October 29. That release will extend the spending record by one month, although it will still precede much of the October survey period. Together with company sales and borrower payment data, it can help test whether households are reducing purchases, changing their mix or continuing to spend despite dissatisfaction. Until then, businesses have evidence of strained confidence and a reason to examine exposure, while the scale of any spending response remains open.
