Does Consumer Confidence Predict Spending?
Consumer spending accounts for approximately 68% of U.S. GDP. If investors could accurately predict changes in consumer spending, they would have a direct window into the trajectory of the economy and, by extension, corporate revenues. Consumer confidence surveys exist precisely for this purpose, measuring the attitudes and expectations of households about their financial situation and the broader economy. The Conference Board's Consumer Confidence Index and the University of Michigan's Consumer Sentiment Index are the two most prominent surveys, and both generate market reactions on release days.
The relationship between what consumers say and what consumers do is less straightforward than it appears. Consumer confidence has predicted some recessions, missed others entirely, and sometimes signaled distress that never materialized in actual spending patterns. Between 2022 and 2024, consumer sentiment dropped to levels consistent with past recessions while actual consumer spending remained remarkably resilient. This divergence raises a pointed question: does consumer confidence actually predict spending, or is it measuring something else entirely?
The Two Major Surveys
The Conference Board Consumer Confidence Index is based on a mail survey of approximately 3,000 households conducted monthly. It has two components: the Present Situation Index, which measures consumers' assessment of current business and labor market conditions, and the Expectations Index, which measures their outlook for the next six months. The index uses a base year of 1985 = 100.
The Conference Board survey is weighted toward labor market perceptions. Its questions ask consumers to assess whether jobs are "plentiful" or "hard to get," which makes it a useful proxy for labor market conditions that aligns well with the official unemployment data. When the percentage of respondents saying jobs are "plentiful" rises, it typically correlates with a falling unemployment rate and vice versa.
The University of Michigan Consumer Sentiment Index is based on telephone surveys of approximately 500 households, with preliminary results released mid-month and final results at month-end. It also has two components: the Index of Current Economic Conditions and the Index of Consumer Expectations. The Michigan survey places more emphasis on personal financial situations and buying conditions.
The Michigan survey also includes inflation expectations measures, both one-year and five-to-ten-year horizons, that the Federal Reserve monitors closely. A spike in consumer inflation expectations can trigger hawkish Fed communication because entrenched expectations risk becoming self-fulfilling through wage demands and pricing behavior. The five-to-ten-year inflation expectations number, in particular, is watched as a gauge of long-term inflation credibility.
The Predictive Track Record
The evidence on whether confidence predicts spending is mixed, and the answer depends on the time horizon, the component of confidence measured, and the economic context.
The Expectations component has a better track record than the headline index. Forward-looking questions about income expectations, business conditions, and employment prospects correlate more strongly with future spending than questions about current conditions. This makes intuitive sense: current assessments largely reflect information already available from other sources, while expectations contain the consumer's own forecast of their future financial capacity.
Confidence predicts large declines in spending better than normal variation. During recessions, consumer confidence drops significantly and the drop tends to precede or coincide with the actual reduction in spending. The Conference Board index fell from 111 in July 2007 to 25 in February 2009, and consumer spending declined roughly 3% in real terms during the same period. The Michigan index fell from 96 in January 2020 to 71 by April, capturing the pandemic shock's impact on consumer psychology.
Confidence is a poor predictor of month-to-month spending changes during expansions. In the normal course of economic growth, confidence fluctuates based on gas prices, political events, stock market moves, and media coverage, while spending follows a much smoother trajectory driven by income, employment, and credit availability.
The 2022-2024 disconnect was the starkest example of confidence failing as a spending predictor. The Michigan sentiment index dropped to 50 in June 2022, the lowest reading in the survey's history dating to the 1950s. This reading was lower than during the 2008 financial crisis, lower than during the early 1980s recession, and lower than during the depths of the COVID shutdown. Yet real consumer spending grew approximately 2.5% in 2022, 2.2% in 2023, and continued expanding in 2024. Consumers said they felt terrible about the economy while continuing to spend at a healthy pace.
Why Confidence and Spending Diverged
Several factors explain the post-2022 disconnect between sentiment and spending.
Inflation sours sentiment without reducing spending power. Rising prices make consumers feel worse about the economy even when their nominal income is rising fast enough to maintain spending in volume terms. A worker earning 5% more who faces 5% higher prices reports feeling worse off (the dollar buys less) but can maintain the same level of purchases. Confidence surveys capture the psychological response to inflation, while spending data captures the economic reality. These concepts are explored further in the economics guide.
Wealth effects offset sentiment effects. Despite low sentiment, household balance sheets entering 2022 were the strongest on record. Excess savings from the pandemic period, rising home values, and elevated stock portfolios provided a financial buffer that enabled spending to continue regardless of how consumers reported feeling.
The labor market remained strong throughout. Unemployment stayed below 4% for the longest sustained stretch since the 1960s. As long as consumers had jobs and paychecks, they spent, regardless of their stated confidence level. Income is a more reliable predictor of spending than sentiment.
Political polarization distorts sentiment surveys. Research has shown that consumer confidence has become increasingly correlated with partisan affiliation. Supporters of the party holding the White House report much higher confidence than opponents, and this gap has widened significantly over the past two decades. This political signal introduces noise that reduces the economic signal in confidence data.
When Confidence Data Is Most Useful
Despite its limitations, consumer confidence data still provides useful investment information in specific contexts.
As a labor market proxy. The Conference Board's "jobs plentiful" minus "jobs hard to get" differential has a strong correlation with the unemployment rate and sometimes provides earlier signals of labor market turning points than the official monthly employment data.
For durable goods spending. Confidence correlates more strongly with spending on big-ticket items (cars, appliances, furniture) than with spending on nondurables or services. Large purchases are more discretionary and more affected by how consumers feel about their financial prospects. When confidence drops sharply, auto sales and furniture purchases tend to follow within a quarter or two.
At extremes. Very low confidence readings, even if they do not predict an immediate spending collapse, have preceded recessions more often than not. The signal is probabilistic rather than deterministic: readings below certain thresholds increase the probability of future economic weakness without guaranteeing it.
For inflation expectations. The Michigan survey's inflation expectations data has genuine monetary policy significance. If consumers begin expecting higher inflation, the Fed takes notice because expectations can influence wage negotiations and price-setting behavior. A spike in long-term inflation expectations can shift the Fed's policy stance, which in turn moves markets.
The Savings Rate Connection
One underappreciated dimension of consumer confidence data is its relationship to the personal savings rate. When confidence is high, consumers tend to save less and spend more freely. When confidence drops, precautionary saving increases and spending growth slows. The personal savings rate dropped to approximately 3% in mid-2022 even as sentiment was at record lows, which seems contradictory. The explanation is that inflation was eroding real income, forcing consumers to draw down savings to maintain their standard of living. They were spending not out of confidence but out of necessity.
During the 2008-2009 recession, the savings rate spiked from roughly 3% to 8% as consumers retrenched. This surge in precautionary saving deepened the recession because every dollar saved was a dollar not spent at a business. The relationship between confidence, savings behavior, and spending is mediated by financial conditions. When consumers have access to credit and their balance sheets are healthy, low confidence may not translate to spending cuts. When credit is tight and balance sheets are stressed, even modest confidence declines can trigger significant spending reductions.
The Federal Reserve tracks related indicators through its Survey of Consumer Finances and its quarterly Financial Accounts data. These provide a more comprehensive picture of household financial health than sentiment surveys alone. A household with $50,000 in savings, manageable debt, and a stable job is unlikely to dramatically cut spending regardless of what it reports on a sentiment survey. A household living paycheck to paycheck with rising credit card balances is a very different story.
Business Confidence and the CEO Perspective
Consumer confidence has a business-side counterpart worth comparing. The CEO Confidence Survey from the Conference Board, the NFIB Small Business Optimism Index, and the Business Roundtable CEO Economic Outlook Survey all measure confidence from the employer's perspective.
Business confidence surveys tend to have a more direct connection to investment and hiring decisions than consumer surveys have to spending. When CEOs report declining confidence in future economic conditions, they are more likely to delay capital expenditure, slow hiring, and build cash reserves. These actions directly reduce economic activity. A CEO who reports pessimism and then cuts her capital budget has made a decision that affects GDP. A consumer who reports pessimism but continues buying groceries and paying rent has not.
The NFIB Small Business Optimism Index is particularly useful because small businesses account for roughly half of U.S. employment. When small business owners report declining expectations for sales, difficulty finding qualified workers, or plans to reduce hiring, the signal often leads the official employment data by several months. The NFIB survey's "plans to increase employment" component has a documented leading relationship with nonfarm payroll growth.
Comparing consumer and business confidence provides a richer picture than either alone. When both are declining simultaneously, the economy is more likely facing a genuine downturn. When consumer confidence is low but business confidence remains solid (or vice versa), the discrepancy often resolves in favor of the business side because business decisions are more directly tied to economic outcomes.
Investment Applications
Consumer confidence data is best used as one input among many rather than as a standalone signal.
For consumer discretionary stocks, confidence provides directional context. A sharply declining trend in confidence suggests caution on retailers, restaurants, travel companies, and auto manufacturers. A recovering trend supports cyclical consumer exposure. But the correlation is loose enough that other data, particularly employment, wages, and credit conditions, should carry more weight.
For the homebuilder and auto sectors, which depend on large, financed purchases, confidence has a more direct connection to demand. Home-buying intentions and vehicle-buying plans within the surveys provide sector-specific signals that are more useful than the headline indices.
For macro positioning, the expectations component of both surveys provides a useful cross-check against other leading indicators. If confidence expectations, the yield curve, initial claims, and manufacturing new orders all weaken simultaneously, the probability of recession increases substantially. If confidence drops in isolation while other indicators remain healthy, the signal is less reliable.
For contrarian opportunities, extremely low confidence readings often coincide with stock market bottoms. Consumer sentiment hit its all-time low in June 2022, within two months of the stock market bottom. The March 2009 confidence trough came within weeks of the generational stock market low. This is not because low confidence causes stock recoveries, but because both confidence and stock prices tend to overshoot on the downside during periods of maximum pessimism.
Consumer confidence surveys capture something real about the psychological state of American households. They just do not capture the full picture of what those households will actually do with their money. The most reliable predictor of consumer spending remains consumer income, particularly labor income from employment. Confidence is the mood. Income is the means. When the two diverge, the means tend to win.
Put these principles into practice. Track fundamentals, build portfolios, and analyze stocks with AI-powered insights.
Start Free on GridOasis →