Consumer sentiment hit a record low in April, with the University of Michigan index falling to 49.8 in its final reading, the weakest print since the survey began in 1952. Yet what consumers actually do, spending, working, paying their bills, still reads healthy in aggregate. The gap between mood and behavior is the story, and the visible strain at the lower end of the income distribution helps explain why the headlines feel worse than the underlying data.
The 2026 Iran war and the late-February closure of the Strait of Hormuz drove the largest inflation-adjusted quarterly oil price move in data going back to 1988, according to the U.S. Energy Information Administration. Brent crude rose from $61 per barrel at the start of the year to $118 by the end of the first quarter. A Dallas Fed working paper published in April estimates the shock will add roughly 0.6 percentage points to 2026 Q4-over-Q4 headline PCE inflation in the baseline scenario, with a prolonged closure pushing the impact as high as 1.8 points. Energy is a larger share of low-income household budgets, so this shock is K-shaped before any other channel kicks in.
The K-shape shows up in retail data as well. The top 20% of earners, those making more than $175,000, now account for roughly 59% of all consumer spending, according to Moody’s Analytics, the highest share in the 36-year history of the data going back to 1989. Headline retail sales rose 1.7% in March 2026, up from 0.7% in February. Strip out gasoline and other fuel, though, and the gain falls to 0.6%. With the national gasoline average roughly 38% above year-ago levels per AAA, some of the headline number reflects consumers paying more at the pump, not buying meaningfully more goods. Wage growth has favored the lower end of the income distribution for several years, but that advantage is now narrowing, an unsurprising cooling after the unusually rapid catch-up that followed the COVID-19 recovery.
Recent layoff announcements at large technology firms complicate the picture, but they are not the distress signal they once were. Meta announced cuts of 8,000 workers, or 10% of its workforce. Microsoft is offering voluntary buyouts to up to roughly 8,750 employees, about 7% of its U.S. workforce. Amazon has eliminated 30,000 corporate jobs since October 2025. The national unemployment rate still sits at 4.3%. Historically, mass layoffs signaled margin defense at companies under pressure. These three are doing the opposite, committing record capital expenditure to cloud infrastructure, automation, and data center buildouts. The cuts read as a reallocation toward artificial intelligence, not retrenchment.
Debt-service capacity is another clean read on consumer health. The Federal Reserve series “Household Debt Service Payments as a Percent of Disposable Income” sat at 11.3% in the fourth quarter of 2025, drifting higher and closer to pre-pandemic levels, though still well below the 15.9% peak of 2007 and 2008. Pandemic-era stimulus, loan forgiveness, and forbearance helped lower-income households materially reduce balances, while low borrowing costs benefited consumers across income cohorts.
The 30-plus day delinquency rate at the major card issuers, including American Express, Bank of America, Capital One, Citigroup, and JPMorgan, averaged 1.30% in February 2026, up just 3 basis points from January and roughly flat year over year, according to S&P Global Market Intelligence. The series has held close to historical norms, setting aside the unusual pandemic and recovery period.
On April 30, the Bureau of Economic Analysis released its advance estimate of first-quarter 2026 GDP, showing real growth of 2.0% annualized. Spending, employment, and debt service all read as stable, even as pressure builds for households at the lower end of the income distribution. The layoff cycle looks more like a reallocation of capital toward artificial intelligence than a sign of corporate stress. How that reallocation flows through to long-term growth, productivity, and employment is the longer-term question; the more immediate one is how long the gap between consumer sentiment and consumer behavior can persist.
