Michigan 5 Year Inflation Expectations

Michigan 5 Year Inflation Expectations is the expected price change over the next 5 to 10 years, from the University of Michigan Surveys of Consumers. It fills the table headed Expected Change in Prices, Next 5 Years. The 5 year label is shorthand for 5 to 10 years, asking about prices in general, not a defined basket. It is not actual inflation, CPI or PCE, nor a professional forecast or sentiment index.

Answers come from rotating panel households across 48 states plus DC, with equal selection chance. The program began in November 1952 and runs monthly since January 1978, with the first long-run reading in February 1979. Monthly readings run unbroken since April 1990.

Interviews span a 27 to 28 day field period: 10 months use 4 weeks and 2 use 5 weeks. Two Friday releases come at 10am ET: preliminary around the second Friday from early interviews, final around the last from all interviews. Preliminary is superseded by final in the same month, leaving one monthly value with no routine revision.

The Federal Reserve uses the median among inputs to its Index of Common Inflation Expectations, with no buy or sell cutoff. It appears with sentiment indexes and 1 year expectations, read by level and direction.

It has no weighted parts because one question drives it. Salient price and trade news, fuel prices and partisan, income and age splits shape responses without fixed weights.

• Direction and buckets: share answering down, same or up, with up split into percent bands.
• Median: middle response and headline gauge of long-run expectations.
• Percentiles and uncertainty: 25th and 75th percentiles, interquartile range as uncertainty and 15% plus tail share.
• Mean and cases: average shown for completeness only, plus case counts.

With no numeric threshold, readings are judged against past ranges: 2.8% to 3.2% through 2024 versus 1.9% to 2.8% for web interviews before the pandemic.

In the April 25, 2025 final release, year-ahead expectations rose from 5.0% to 6.5%, the highest since 1981 across all political groups. Long-run expectations rose from 4.1% to 4.4% on a large move among independents.

About 60% named tariffs unprompted, up from about 44% in March 2025, with weaker sentiment and higher expectations among them. Seven-day averages rose after the worldwide-tariff announcement of April 2, 2025, then ebbed after the partial pause of April 9, 2025 while staying highly elevated. The 1 year gauge moved about twice as far as the 5 to 10 year gauge, showing anchoring through long-run stability. The expectations index fell 32% since January 2025, the steepest three-month drop since the 1990 recession.

1. Direction question: Respondents say whether prices will be higher, same or lower over the next 5 to 10 years.
2. Same probe: Since March 1982, those saying same say whether they mean the same rate of rise or no rise.
3. Percent probe: Each respondent gives an annual percent for up or down, forming one complete response with direction.
4. Missing answers: Direction-only cases are split proportionally across complete answers with fractional cases, not dropped. Dropping them would bias results down because unmeasured rises always exceed unmeasured falls.
5. Adjustments: Procedures adjust extremes and correct the pre-March 1982 same-code misreading, keeping stated decline percentages.
6. Median: The headline is the median by interpolation, since extremes pull the mean and tails drive variance. In the publisher's illustrative example of 12 interviews including 0%, 12% and 20%, the median is near 3.5%. The mean reaches about 5.4% because of the 12 and 20.
7. Weighting: Household and adult weights correct selection, coverage and nonresponse by raking to Census surveys, fresh and recontact separately.
8. Collection and seasonal adjustment: Interviews ran by telephone through March 2024, then moved online, with all official estimates web-based since July 2024. There is no seasonal adjustment step, so medians publish as collected.