Board of Governors of the Federal Reserve System

April 24, 2025

Tracking consumer sentiment versus how consumers are doing based on verified retail purchases

Sinem Hacıoğlu Hoke, Leo Feler, Samantha Mitchell, and Jack Chylak¹

Despite low unemployment, moderating inflation and anchored inflation expectations, and rising incomes since mid-2022, surveys at the end of 2024 continued to report that consumer sentiment remained unusually low, below levels at the onset of the pandemic and on par with levels during the Great Financial Crisis. Why was there this discrepancy between consumer sentiment and the real economy?

This note summarizes the results of a household survey that aims to understand how changes in household incomes and spending between 2019 and 2024 and how behavioral changes households made to adapt to the economic environment shaped their economic sentiment. It includes nearly 10,000 representative panelists' responses to various questions collected over two waves in October and November 2024.² Using micro spending data, we link survey responses to respondents' total annual spending to gauge how actual spending and household-specific price levels for everyday retail purchases changed between 2019 and 2024. Being able to connect sentiment with verified spending is a novel contribution, above and beyond previous studies that explored economic sentiment in the post-pandemic period.

Our results indicate that:

1. The more people thought the prices they paid rose faster than their incomes, the worse they said they were doing.
2. Consumers were more likely to overestimate than to underestimate the inflation they experienced. Those consumers who overestimated their verified inflation said they felt worse about economic conditions.
3. Most respondents reported higher household incomes in 2024 versus 2019 but still said they did not feel good about the economy due to the effort they exerted to adapt to the economic environment.
4. After adjusting for inflation, verified spending on everyday retail items remained strong even among those who reported having lower incomes or among those who said they felt worse about the economy in 2024 compared with 2019.

Taken together, we show that what consumers have been saying differs from what they have been doing during the post-pandemic period; consumers say they feel worse, but through the end of 2024, they are buying more – not just spending more – than they did in 2019. This disconnect between what consumers have been saying and doing suggests that consumer sentiment surveys on their own have become weaker indicators of future consumer behavior and of the health of US consumers. While it is important to recognize how consumers feel, we should exercise caution when using consumer sentiment surveys to infer future consumer behavior given this recent disconnect between what consumers say and do.

Consumer sentiment about current conditions has remained low despite rising wages and moderating inflation

Over the past few years, there has been a change in how overall consumer sentiment corresponds with sentiment regarding incomes and prices. Usually, consumer sentiment moves closely with sentiment regarding income growth. More recently, sentiment has been moving closely with sentiment regarding price levels.

Figure 1 plots three lines from the University of Michigan's Survey of Consumers. Consumer sentiment fell to historic lows in mid-2022, lower than during the Great Financial Crisis and during the depths of the pandemic. In late-2024, consumer sentiment remained substantially lower than pre-pandemic levels. Historically, consumer sentiment moves in tandem with concerns regarding lower income and higher prices on household finances, and sharp drops in consumer sentiment tend to precede or coincide with recessions. This time, the sustained drop in consumer sentiment following the pandemic has not preceded or coincided with a recession. Perhaps one of the reasons for this anomaly is that the link between overall consumer sentiment and sentiment regarding income seems to have broken in mid-2022, which is evident in the inset box that shows the series from 2019 onwards. Since mid-2022, consumer sentiment has been moving more closely with sentiment regarding higher prices than with sentiment regarding income. This situation similarly occurred during the inflationary episodes of the 1970s and early 1980s. In other words, during inflationary episodes, consumers seem to put more weight on higher prices than on higher incomes when assessing their current economic conditions, and as long as income growth remains robust, consumers continue spending even though in sentiment surveys, they say they feel pessimistic about the economy.

Figure 1. The index of consumer sentiment and percentage of consumers who mention high prices and lower incomes as reasons for their financial situation being worse than a year ago

Linking household surveys with panel data to better understand the rationale for low consumer sentiment

To better understand households' sentiment regarding current economic conditions, we use a custom household survey run by Numerator, a consumer panel data and survey company.³ The survey received nearly 10,000 responses in two waves in October and November 2024.⁴

To understand how respondents' economic sentiment is linked to changes in income and spending, the survey first asked panelists how they feel in 2024 compared with 2019, approximately how much they earned in 2024 compared with 2019, and how much they are spending in 2024 compared with 2019.⁵

1. How they feel: 43 percent of respondents said they were doing worse or much worse, 32 percent of respondents said they were doing about the same, and 25 percent of respondents said they were doing better or much better.
2. Changes in self-reported income: 26 percent of respondents said they have lower household income in 2024 compared with 2019. Some of these respondents reported that they or someone in their household lost their job or retired. 31 percent said their incomes increased 0-10 percent. 22 percent reported income increases of 10-30 percent. And the remaining 21 percent reported income increases greater than 30 percent.⁶
3. Changes in self-reported versus experienced inflation: We compare how much more households said they were spending now for the items they buy versus what their transaction data, collected by Numerator, shows for experienced inflation. To calculate experienced inflation, we take the items consumers bought in 2019 and track the price changes for those items through 2024 while keeping consumers' baskets constant. We use each consumer's 2019 purchases to calculate weights and construct an individualized consumption basket for each consumer.
   
   Figure 2 compares self-reported versus experienced inflation, providing a matrix for how households responded to the survey question (on the left axis) versus what their verified purchases showed (on the top axis). 61.8 percent of consumers experienced inflation of 20-30 percent, in line with the cumulative CPI increase for everyday retail purchases like groceries and general merchandise between 2019 and 2024, but only 28.8 percent of consumers said they experienced inflation of 20-30 percent in their survey responses. Most consumers over-estimated the extent of inflation they experienced. For example, 24.0 percent of consumers said they experienced inflation greater than 40 percent, on average, across their everyday retail purchases. Based on verified purchases, we see only 1.7 percent of consumers experienced inflation greater than 40 percent between 2019 and 2024. Some consumers also under-estimated the extent of inflation they experienced, but a greater share of consumers over-estimated than under-estimated their experienced inflation.

Figure 2. Survey responses for perceived inflation versus experienced inflation based on verified purchases between 2019 and 2024

4. Changes in perceived purchasing power: We next combine changes in self-reported income and inflation to capture how households think their purchasing power changed relative to 2019. In Figure 3, the horizontal axis shows the difference between how much households said their income increased versus how much they said their spending increased. If the change in income and spending is the same, most respondents should fall in the 0 percent to 10 percent bin. If a respondent's change in perceived spending is greater than the change in their income, they will be in the negative bins; conversely, if their change in income is greater than their change in spending, they will be in the positive bins. The widths of the bins are proportional to the share of surveyed panelists in each bin. We segment the height of the bins based on how panelists said they were doing in 2024 relative to 2019.

Figure 3. Sentiment in 2024 compared with 2019 by the difference between the change in annual income and change in self-reported spending

Most respondents felt their annual income did not keep up with their spending but recall that consumers tend to over-estimate the extent of retail price inflation they experienced. Only 23 percent of respondents reported that their incomes grew at the same rate or faster than their spending. The more consumers said their income did not keep up with spending, the greater the share who said they were doing worse or much worse in today's economy compared with 2019.

Let's look at what consumers do, not just what they say

We saw from Figure 2 that consumers tend to over-estimate their inflation experience. Do consumers also under-estimate the extent to which incomes kept up with inflation? The survey asked panelists to choose between the following three options to gauge their perceptions of how much their incomes changed compared with retail prices: i) Our incomes went up by more than the prices we pay for items we buy; ii) Our incomes went up about the same as the prices we pay for items we buy; iii) Our incomes went up by less than the prices we pay for items we buy.

Only 14 percent of respondents said their incomes went up more than prices, 33 percent said their incomes went up about the same as prices, and 53 percent said their incomes went up less than prices. Of those who reported their incomes went up less than prices, 65 percent said they felt worse or much worse in 2024 compared with 2019. On average, respondents felt the prices they paid have increased more than their incomes, and they feel worse or much worse than they did in 2019.

The Numerator data allow us to link panelists' survey responses with their retail spending.⁷ In the top panel of Figure 4, we show box plots for the change in real annual spending (adjusted for inflation) by sentiment from 2019 to 2024. Despite substantial variation, most panelists are buying more in 2024, after adjusting for inflation, compared with 2019. Panelists' sentiment tends to improve the more they are able to buy. In the bottom panel of Figure 4, we show panelists' change in real annual spending (adjusted for inflation) by the categorical change in their income based on our calculations using their self-reported incomes in 2019 and 2024. Spending increases with income, as expected, but even when respondents reported lower income compared with 2019, they are still buying more. These two figures and the survey responses collectively show that despite deteriorating sentiment, retail spending remained strong across panelists, even when they said they experienced declines in their incomes since 2019.⁸

Figure 4. Change in real (inflation-adjusted) annual spending between 2019 and 2024 by sentiment and income

Why are consumers unhappy?

We have shown that sentiment deteriorates as changes in spending outpace changes in income. But people are not only spending more – we adjust for inflation – they are buying and consuming more in 2024 compared with 2019.

Recent papers explore the potential reasons behind deteriorating sentiment in high inflation periods. Stantcheva (2024) (PDF) documents that people dislike inflation because of perception. People perceive that prices increase while wages do not seem to catch up. People do not associate increases in their income with inflation-adjustments but instead with personal achievements they have earned or deserve. Although the labor market has been strong, with wages rising faster than inflation, the negative sentiment associated with inflation has been an offsetting factor in judging today's conditions and how they compare with the past. More recently, Lewin, Orchard and Dasgupta (2025) show that consumer's education-group specific inflation rate is more predictive of their well-being than the education-group specific unemployment rate or wage growth. Guerreiro, Hazell, Lian and Patterson (2024) document that actions workers take to earn more income affect their well-being, and inflation adjustments on wages do not fully reflect the effort workers exert even when incomes catch-up with price changes.

Here we show that lower sentiment about the economy is also related to changes people made to adapt to economic circumstances. To understand changes people made between 2019 and 2024 to cut costs, the survey asked two questions: i) how much effort have you put into cutting back on expenses since 2019; ii) have you made changes to your shopping behaviors such as clipping coupons, buying items when they are on promotion, shopping at lower-priced retailers, shopping around to find the best prices, buying in bulk to save money, or buying more store brands instead of brand name items.⁹

First, 80 percent of panelists said they put a lot or some effort into cutting expenses. Only 5 percent percent said they had not been trying to cut back on expenses. Second, as Figure 5 shows, negative sentiment increases with the number of "Yes" answers to the changes households made to cut back: the more changes households made, the worse they feel.
 

Figure 5. How many changes have people made to their spending between 2019 and 2024? Those who are doing worse say they've done more to cut back

Next, the survey asked panelists if they changed jobs in order to earn a higher wage, if they took on additional jobs, and if they regularly work more hours per week now.¹⁰ As we show in the left panel of Figure 6, respondents who changed jobs to earn a higher wage were more likely to say they were doing better. But as the right panel of Figure 6 shows, respondents who said they took on additional jobs or now regularly work more hours per week were more likely to say they were doing worse.

Figure 6. Changes that may have impacted household incomes between 2019 and 2024: Taking on additional jobs or working more hours is associated with doing worse now than in 2019

Concluding remarks

In this note, we summarize our observations from an economic sentiment survey linking panelists' responses to verified purchases. Consumer sentiment deteriorated since 2019, but spending remained strong even among those who felt they were doing much worse or experienced income losses as of 2024.

People's negative sentiment seems to be driven by the perception that incomes have not kept up with prices, even though real spending has increased, and by the effort they exerted to adapt to rising prices. Although sentiment improves with higher incomes, the more people said they had to make changes to their behaviors since 2019 to reduce spending, the worse is their sentiment. Moreover, those who experienced increases in their incomes still reported negative sentiment, citing the need to work more hours or take on additional jobs to earn extra income. In contrast with much of the history of consumer sentiment measures since the 1960s but similar to the history of consumer sentiment during other inflationary episodes, consumer sentiment in 2024 was more closely related to concerns about higher prices than to concerns about lower income. This situation may reverse as inflation continues to decline or if the labor market weakens.

References

Cavallo, Alberto and Oleksiy Kryvtsov (2024). "Price Discounts and Cheapflation During the Post-Pandemic Inflation Surge," Journal of Monetary Economics.

Guerreiro, Joao, Jonathan Hazell, Chen Lian, and Christina Patterson (2024). Why Do Workers Dislike Inflation? Wage Erosion and Conflict Costs. NBER Working Paper 32956.

Hacioglu Hoke, Sinem, Leo Feler, and Jack Chylak (2024). "A Better Way of Understanding the US Consumer: Decomposing Retail Sales by Household Income," FEDS Notes. Washington: Board of Governors of the Federal Reserve System, October 11, 2024.

Lewin, Nicolas, Jacob Orchard, and Kabir Dasgupta (2025). "Do People Care More About Inflation or Wage Growth?," FEDS Notes. Washington: Board of Governors of the Federal Reserve System, March 28, 2025.

Stantcheva, Stefanie (2024). Why do We Dislike Inflation? Spring 2024 Brookings Papers on Economic Activity (BPEA).