The pandemic has broken a closely followed survey of sentiment

Americans are gloomy about the state of the economy. Since the covid-19 pandemic began, consumer sentiment has been in the doldrums, hitting its lowest level ever in June 2022. Such negativity has prompted claims that the country is suffering a “vibecession”—although the market appears healthy, good vibes are lacking.

Changes in consumer sentiment are normally a useful economic benchmark. The longest-running measure comes from a survey by the University of Michigan, which began in 1946. Each month it asks a representative sample of 600 Americans a set of five questions probing their opinions on their own finances and spending, the state of the wider economy, and the outlook for both. For most of the past four decades the index has moved in lock-step with changes in current consumer spending, and also loosely predicted spending a year ahead.

Economists worry that current economic pessimism will become a self-fulfilling prophecy. Although a despondent public would have been a grave warning sign before the pandemic, since 2020 the mix of economic indicators that shape sentiment has changed, and as a result the measure has lost its predictive power.

The first study highlighting this pattern was published last month on X (formerly Twitter) by a researcher using the handle “quantian1”, who chose to remain anonymous. Extending this analysis, we built a statistical model to predict the monthly consumer-sentiment index between 1980 and 2016 using a broad battery of economic data. A combination of 13 variables, including inflation, unemployment and petrol prices explained 86% of the variation in the index in this period, a very good fit.

Before the pandemic, the relationships between these indicators and consumer sentiment were relatively stable. When tested on data from 2017-19, the model trained on the 1980-2016 period reliably predicted sentiment, with only small errors. However, covid seems to have severed this link, making the model’s projections wildly inaccurate. If the pre-2020 associations still held, today’s score would be 98, some 30 points above the actual value.

Although Americans report being worried about their finances, they are behaving as flush as ever—and in economic forecasting, actions speak louder than words. When used to project future spending rather than consumer sentiment, the same battery of economic variables has fully maintained its forecasting power since 2020. In contrast, since covid began, the correlation between sentiment and both current and future spending has vanished.

Our results should assuage concerns that a vibecession today spells a recession tomorrow. The gap between sentiment and economic reality has finally stabilised after growing steadily from 2020-22. Bad vibes may be the new normal. ■