Consumer sentiment surveys ask people how they feel about their finances and the economy. Their results can help explain the mood behind economic news, but a low reading is not, by itself, proof that households will cut spending or that a downturn is coming. The distinction matters: surveys record answers, not transactions.
The University of Michigan Surveys of Consumers began in 1946. Their founding premise was that household decisions about spending and saving help shape the national economy. That makes people’s expectations relevant to economic analysis, even when those expectations do not match other measures of the economy.
The survey asks people about their own financial prospects and their expectations for the general economy in both the near term and the longer term. Other questions cover business and buying conditions, expected changes in inflation, unemployment and interest rates, and views of government economic policy. Respondents can also explain their answers in their own words, giving context that a single index number cannot show.
That context can include a household’s experience with prices or job prospects. But a survey answer is still a person’s reported view at a particular time. It does not verify how much that person earns, owes or spends.
What the survey captures
The Michigan Index of Consumer Expectations brings together views on three broad subjects: a respondent’s personal financial prospects, the economy in the near term, and the economy over the longer term. The survey’s description from the University of Michigan lays out the topics and the role the survey assigns to consumer decisions. The questions cover more ground than whether people feel optimistic or pessimistic.
For example, a person may expect their own finances to improve while thinking the broader economy will weaken. Another may say current buying conditions are poor because prices feel high, even while expecting inflation to slow. Those answers can point in different directions. A headline index compresses many responses into a summary, so it cannot preserve every household’s reasons or show why individual answers differ.
How interviews become a monthly reading
Michigan conducts about 1,000 interviews each month and includes roughly 50 core questions. Its sample is designed to represent U.S. households, excluding Alaska and Hawaii, and participants are recruited using postal addresses. The survey has moved to web-based interviewing. Its technical documentation includes information on that transition and comparisons of web and phone data from 2017 to 2024.
The schedule matters when readers compare headlines. Preliminary results typically come out in the middle of the month, followed by final results at month-end. The initial reading can therefore change. A preliminary figure is useful as an early snapshot, but it should not be treated as the final monthly result.
Changes in survey methods also deserve attention when comparing readings across years. A switch in how interviews are conducted may affect how people respond or who completes the survey. The technical documentation is the place to check the details before assuming that every change in the index reflects a change in public opinion alone.
Why household mood and economic data can diverge
Sentiment is not the same as a household’s financial condition, and neither is identical to the broader economy. A Federal Reserve survey of household economic well-being asks people directly about their own situation, including whether they are doing okay financially and whether they feel worse off than a year earlier. Those questions are narrower than a broad assessment of the economy.
The gap between public mood and economic indicators drew attention in 2024. Brookings described a “paradox” in which many measures of the economy looked strong while consumer sentiment remained weak. For example, the unemployment rate was 4.1% in September 2024, below the 21st-century average of 5.7%. Over a different period, household wealth rose by about $50 trillion in nominal terms from the first quarter of 2020 to the second quarter of 2024, while real median weekly earnings were only 0.3% above their level in the fourth quarter of 2019. Brookings’ analysis sets those figures against continued weak sentiment.
These measures describe different experiences. A household can see asset values rise without having more cash available for groceries or rent. Someone whose pay has barely changed after inflation may feel squeezed even as national employment figures look healthy. Survey responses can reflect those pressures, but they do not tell us how common any one experience is without looking at the underlying results and other data.
What sentiment can, and cannot, forecast
Michigan expectations measures are included in leading-indicator composites, including an OECD composite for the United States. A leading indicator is a measure used as one clue about what may happen later. That inclusion shows that analysts use expectations data in forecasting frameworks. It does not establish how accurately sentiment predicts later spending, saving or recessions, or how much it improves a forecast compared with other information.
To judge predictive power, readers would need evidence that compares survey readings with later outcomes over time. They would also need to know how the index is calculated, how results change when survey methods change, and whether it adds useful information beyond employment, income, prices and other measures. The survey description explains the questions and their purpose, but a summary reading alone cannot answer those questions.
For everyday readers, the careful use is as a measure of reported expectations and concerns. A decline can tell us that surveyed households feel less confident at that point in time. It cannot forecast what a particular family will spend, determine its financial position, or guarantee that a recession will follow. For a fuller picture, compare sentiment with official economic data and check the survey’s methods and release notes. Individual financial decisions depend on a household’s own circumstances, not a national mood index.
Please seek a professional. This article is general information, not financial, tax or insurance advice. Please seek a licensed financial, tax or insurance professional about your own situation.
