Guide · Markets
Consumer Confidence vs Sentiment: Why Two Surveys Disagree
The Conference Board's consumer confidence index and the University of Michigan's sentiment index both survey households but ask different questions, so they can point in opposite directions. Here is how each is built, why October 2026 gave 91.7 on one and 48.0 on the other, and which parts markets actually watch.
Two surveys of American households are released every month, and both are often described as measures of consumer mood. The University of Michigan's Index of Consumer Sentiment comes out twice a month, a preliminary reading on the second Friday and a final one at the end of the month. The Conference Board's Consumer Confidence Index comes out on the last Tuesday. They are not two versions of the same number. They ask different questions, of different samples, and weigh different worries.
Michigan's survey, run by the university's Surveys of Consumers since the 1940s, asks five core questions about personal finances now and a year ahead, business conditions over one and five years, and whether it is a good time to buy large household items. Because two of the five questions are about the household's own finances and one is about buying conditions, it is sensitive to prices: when inflation erodes purchasing power, sentiment falls even if jobs are plentiful. Its index is set to 100 in the first quarter of 1966. The survey also produces the one-year and five-to-ten-year inflation expectations that the Fed and bond traders follow closely.
The Conference Board, which began its survey in 1967, puts more weight on the labour market and the wider economy. Its index, set to 100 in 1985, combines a Present Situation Index, built from views on current business conditions and jobs, and an Expectations Index for the next six months covering business conditions, jobs and income. A reading of the Expectations Index below 80 is often cited by the Conference Board as a signal of recession within a year. Its most closely watched detail is the labour differential: the share of people saying jobs are plentiful minus the share saying they are hard to get, which tends to move with the unemployment rate.
October 2026 shows how far apart the two can be. On October 23 Michigan's final sentiment index came in at 48.0, with one-year inflation expectations of 4.7% and long-run expectations of 3.5%, both the highest since May. Four days later, on October 27, the Conference Board's index rose 7.5 points to 91.7, its highest since June and well above the forecast of about 81; the Present Situation Index rose to 120.4, the Expectations Index to 72.6 and the labour differential widened to 5.3 points. Both surveys found inflation worries, the Conference Board's respondents expecting prices to rise 5.8% over the next year, but only Michigan's design turned those worries into a gloomy headline. Easing gasoline prices and a steadier job market were enough to lift the Conference Board's measure.
Markets rarely trade the headline of either survey for long, because spending data matter more than what people say about it. The parts that do move prices are narrower. Michigan's inflation expectations can shift rate expectations, since the Fed watches whether households start to believe high inflation is permanent; the Conference Board's labour differential is read as an early hint for the next jobs report. A big gap between the two surveys usually says something about the economy itself: in 2026, a strong job market and persistent inflation made people feel secure about work and squeezed on prices at the same time.
When reading either release, compare the result with the consensus, look at the revision to the previous month, and go straight to the components rather than the index. Check Michigan's one-year and long-run inflation expectations, the Conference Board's jobs plentiful and hard-to-get shares and its 12-month inflation expectations, and whether the Expectations Index is above or below 80. Then wait for the hard data: retail sales, personal spending in the BEA's monthly report and the jobs report will show whether consumers acted on their mood.