Paychecks grew again in August, but the cost of living grew faster. The Bureau of Labor Statistics reported on September 11 that average hourly pay, once adjusted for inflation, fell for the typical U.S. worker last month, and measured against a full year of price increases, real hourly pay in August 2026 is lower than it was in August 2025. For a household working from a regular paycheck, the numbers confirm something many families have already suspected: a raise on paper does not always keep up in the checkout line.
Average Hourly Earnings Rose 0.3%, but Prices Rose 0.4%
Average hourly earnings for all employees on private nonfarm payrolls rose 0.3 percent from July to August, seasonally adjusted, the Bureau of Labor Statistics reported. That would normally read as a straightforward win for workers. But prices moved faster: the Consumer Price Index for All Urban Consumers (CPI-U) climbed 0.4 percent over the same month. Once BLS adjusts the wage figure for that price increase, real average hourly earnings for all employees fell 0.1 percent in August.
The drop is recorded in the agency’s Real Earnings news release, which also shows real average weekly earnings rising 0.2 percent for the month, an increase driven entirely by a longer average workweek, up 0.3 percent, rather than by hourly pay keeping pace with prices.
Free retirement updates: Social Security and Medicare change every year, and nobody sends you a memo. Our free Retirement Shield newsletter breaks down what changed and what to do. Get it free in your inbox.
A Full Year of Lost Ground: Real Pay Down 0.3% Since Last August
The monthly dip is not a one-off. Measured from August 2025 to August 2026, real average hourly earnings for all employees fell 0.3 percent, seasonally adjusted, according to the same release’s underlying earnings table. That means the average worker’s hourly pay, after inflation, moved backward over a full year, not just in a single month. Real average weekly earnings did rise 0.3 percent over that same 12 months, but again the reason was hours rather than the hourly rate: the average workweek grew 0.6 percent longer over the year, more than offsetting the decline in real hourly pay.
The distinction matters for a household budgeting off a regular schedule. A worker who kept the same hours week to week took home less real purchasing power in August 2026 than a year earlier. Any gain in take-home pay over the year came from working more hours, not from each hour of work being worth more after inflation.
Hourly and Clerical Workers Felt the Same Squeeze, from a Different Price Gauge
The BLS release breaks out a second group, production and nonsupervisory employees, a category that covers most hourly and non-managerial workers and excludes higher-paid supervisory and salaried staff. For this group, real average hourly earnings also fell 0.1 percent from July to August, the same monthly decline as the broader workforce. The mechanics were different: nominal average hourly earnings for this group rose 0.3 percent, but the relevant price measure, the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W), rose a steeper 0.5 percent.
Over the year, this group fared somewhat better than the all-employee average, according to the production and nonsupervisory earnings table. Real hourly earnings for production and nonsupervisory workers fell 0.1 percent from August 2025 to August 2026, versus the 0.3 percent decline for all employees, and their real weekly earnings rose 0.1 percent over the year, helped by a 0.3 percent increase in the average workweek.
Longer Hours, Not Bigger Raises, Are Propping Up Paychecks Right Now
Across both worker groups, the pattern repeats: whenever real weekly earnings rose, it was because people worked more hours, not because each hour of work paid more after inflation. That is a meaningfully different story for a household than a straightforward raise. Pay growth tied to extra hours on the schedule is fragile in a way that pay growth tied to the hourly rate is not, since hours can be cut back in a slower month just as easily as they were added in August.
How BLS Turns a Pay Stub Into a “Real” Number
The Real Earnings release exists specifically to strip inflation out of wage growth, using a method the agency documents in its own technical notes. BLS starts with average hourly and weekly earnings from its Current Employment Statistics survey, then divides those figures by the CPI-U, for all employees, or the CPI-W, for production and nonsupervisory workers, to produce the constant-dollar figures published each month. That is why a paycheck can rise on its face and still lose ground: the nominal wage an employer reports and the real wage after inflation can move in opposite directions within the same month.
The Next Reading Comes in October
BLS has already set the date for the next word on this trend. The agency’s own release states that the Real Earnings report covering September 2026 is scheduled for October 14, giving households and employers a little over a month to see whether August’s dip in real hourly pay was a one-month blip or the start of a longer stretch of wages losing ground to prices.
What a frozen account does not explain
Federal benefits carry protection from garnishment, but the protection has to be demonstrated before the money moves.
The Bank Account & Debt Protection Kit includes a 10-page kit, the 2-month bank protection rule and the debt-validation steps.
See the 2-month bank protection rule in The Bank Account & Debt Protection Kit.
This article was produced with AI assistance and reviewed by a human editor. Figures are linked to their primary sources; where a claim could not be verified from the public record, we say so.
