Is the vibecession real — or is the survey broken?

SubscribeSign in

Politics<br>Is the vibecession real — or is the survey broken?<br>A shift to online polling and undersampling of Republicans are skewing America’s most-cited measure of consumer sentiment.

Joel Wertheimer<br>Jun 29, 2026

84

Share

Black Friday at Garden State Plaza in Paramus, New Jersey in 2025. Eduardo Munoz Alvarez/Getty Images<br>Today’s newsletter is a guest post from Joel Wertheimer. Joel is a civil rights attorney in New York and has served as a policy consultant to political candidates. He worked for President Obama from 2015 to 2017 as Associate Staff Secretary. He irregularly writes about data, soccer and politics at wertwhile.substack.com.

For nearly four years, the internet has debated whether we’ve been mired in what Kyla Scanlon dubbed a “vibecession” — whether people feel worse about the economy than the underlying data suggests they “should” feel. People as esteemed as Nobel laureate Paul Krugman have frequently posted about the vibes mystery. Nate wrote a whole piece about the divergence in the New York Times two years ago.<br>But there’s one big problem with the discussion: most of the participants are relying on a broken survey, the University of Michigan’s consumer sentiment survey (“Index of Consumer Sentiment” or “ICS”), that is in dire need of being repaired. Failure to correct for these issues has led to plenty of pet theories — but they explain a trend that may not even exist.<br>On Friday, the survey reported a 49.5 reading for consumer sentiment, a number that, while slightly improved from last month, puts it below the depths of the Great Recession. But without properly correcting for the survey’s issues, the headlines are misleading. No, Americans are not as unhappy about the economy as they were during the Great Recession.

Subscribe

What’s wrong with the University of Michigan survey

The University of Michigan ICS is the gold standard sentiment survey measuring consumer sentiment.1 The survey has historically shown a very strong correlation with “hard” economic data such as inflation and unemployment. But before more bad analysis gets done on the vibecession, people need to know they’re working with dubious data. As with election polls, the ICS has struggled amid a shift away from telephone polling. There are issues both with partisan nonresponse, with some political groups more likely to respond than others, and partisan expressive response, with survey-takers using questions about the economy to express political sentiment.<br>So the problems with the ICS are these:<br>The switch to online polling made responses more negative and,

There are too many Democrats in the sample.

Thus, ICS data since mid-2024 is not comparable to past periods. Here is the partisanship of the sample over time:

And Democrats right now say they hate the economy with Trump in charge.

Share<br>When adjusted for these issues, the ICS should be substantially higher than the Great Recession lows we have witnessed over the past year. Weighting the survey to Pew’s National Public Opinion Reference Survey — what Nate Cohn called “perhaps the most important poll you’ve never heard of” — would place the ICS at a level more like that in 2013, when the economy was growing after the Great Recession but unemployment remained stubbornly high and wages stubbornly low. This adjustment would bring the survey in line with other measures of consumer confidence, such as those from the Conference Board, Gallup, and YouGov.<br>The online version of the ICS finds more negative sentiment

Some background on the ICS: for most of its existence, the survey was conducted by phone and weighted to demographic variables: age, sex, race, education, income, and region. Beginning in April 2024, the University of Michigan began a phased transition from phone to web polling, becoming fully web-based by July 2024, as response rates to telephone polls plummeted. Michigan was meticulous about the switch and polled both by phone and online for a period to establish the validity of the online polling method, finding a .97 correlation between the phone and web series.<br>But as Ernie Tedeschi and Ryan Cummings have noted at Briefing Book, that came with a seemingly permanent level shift in the survey, of nearly nine points down. A high time-series correlation only means the phone and web series moved together; it does not rule out a persistent level shift, with web respondents consistently registering lower sentiment. If every group’s economic outlook declines roughly uniformly, the correlation will hide that.<br>After controlling for respondents’ demographic characteristics across the web and phone versions, Tedeschi and Cummings found that the mode effect was very strong. The correlation that the University of Michigan found is still there, but web survey takers persistently expressed much more negative views than those reached on the phone (a well-documented phenomenon in other surveys). They...