US Consumer Sentiment August 2026 — Michigan Index 51.0

After two months of tentative improvement, household mood broke lower in early August. The headline index sank to 51.0, undershooting forecasts near 54.5, with the expectations component tumbling nearly five points to 50.6 and current conditions falling to 51.8.^[1][2][3]

The inflation picture drove the pessimism: year-ahead expectations edged up to 4.3% while long-run expectations stayed at an elevated 3.3%. Survey director Joanne Hsu noted the decline crossed political lines and hit hardest among older, lower-income, and less-educated households most exposed to rising prices.^[1][2][3]

A reading of 51 invites the reading that consumers are at 'half' strength, or that 51% feel positive. Neither is what the number means. The index is anchored to 1966 = 100, so it expresses sentiment relative to that base year, not as a percentage of anything.^[5]

The calculation is specific: five survey questions each produce a relative score — «the percent giving favorable replies minus the percent giving unfavorable replies, plus 100» — and the sum is divided by «the 1966 base period total of 6.7558» before adding a constant of 2.0. Because the base period sets the divisor, the index can run well above 100 in good years; it is a ratio to 1966, not a share.^[5]

Historical placement also needs care. The commonly cited record low of 50.0 from June 2022 has been surpassed: the published series shows 44.8 in May 2026. The 2026 path ran 56.4 in January down to that May trough, back to 55.2 in July, then 51.7 in August. Citing 2022 as the floor is now out of date.^[4][6]

The mid-month preliminary figure and the end-of-month final are the same calculation on different amounts of data. The survey targets «approximately 420 cases for the preliminary reading», while the final «typically reflects approximately 1,000 interviews». The preliminary therefore rests on roughly 40% of the eventual sample.^[1][5]

August 2026 shows the effect: the preliminary print was 51.0 and the final came in at 51.7. Revisions of this size are routine rather than a sign of error, which is why citing a preliminary number weeks later — after the final exists — is a common way to quote a figure that has since been superseded.^[4]

Michigan publishes sentiment separately by party identification, and the gap is large. In May 2026 the readings were 32.8 for Democrats, 40.6 for independents and 84.6 for Republicans — a spread of more than 50 points on a single index.^[6]

It is tempting to conclude that the headline number therefore measures partisanship rather than the economy. The survey's own analysis pushes back on exactly that reading: «national estimates are consistently aligned with that of independents, a group that is less likely to be influenced by partisan extremes… National estimates are not dominated by one end or the other of the political spectrum».^[6]

So the useful version of the caveat is narrower than 'the index is political'. Movements in the overall index track the least partisan group closely, while the subgroup figures can swing on political alignment. Quoting a party-specific number as though it described consumers generally is where the distortion enters.^[6]

The preliminary reading for September 2026 is scheduled for Friday, September 11, 2026 at 10am ET, with the final at month's end. The preliminary lands in the second week of the month and the final in the last week — a cadence worth checking before quoting whichever figure appears first in search results.^[1][4]