Look at how many deflators are in use and the ambiguity becomes concrete. Inside a single Bureau of Labor Statistics release, two different price indexes are applied to two different populations: "the Consumer Price Index for All Urban Consumers (CPI-U) is used to deflate earnings for the all employees series, while the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) is used to deflate earnings for the production and nonsupervisory employees series." The two indexes cover different population groups and are constructed with different expenditure weights, so the two real series are not simply the same measure at different levels of detail.
A third choice is in use at the Federal Reserve Board, which deflates a different wage measure by a different index again. Its Monetary Policy Report discusses real wage growth "as measured by the Federal Reserve Bank of Atlanta's Wage Growth Tracker and deflated by the personal consumption expenditures price index," a series the same report describes as reporting "the median 12-month wage growth of individuals responding to the Current Population Survey." So the two most-cited institutional real-wage series in the United States differ in both the numerator, an establishment average against a matched-individual median, and the denominator, a consumer price index against the personal consumption expenditures price index. Neither is doing anything wrong. They are answering different questions, and their numbers should not be expected to match. What each price index is and how they are built belongs to the pages on the consumer price index and the PCE price index.
Precisely because the parameters differ, one document can carry two real-wage statements that look inconsistent and are not. The Board's Monetary Policy Report of July 2026 reports both. On the aggregate: "Solid nominal wage gains have nonetheless been outpaced by overall price inflation over the past year. With the recent jump in energy prices boosting the change in PCE prices to 4.1 percent over the 12 months ending in May, the purchasing power of workers' wages, in the aggregate, declined somewhat over this period. That said, for the year ending in May 2025, wage gains outpaced price inflation." Both halves of that belong together; a two-year story told with one of them becomes a trend that neither sentence supports.
On groups, the same report says that since late 2024 real wage growth "for workers in the bottom quartile has fallen below that of the other quartiles but has remained positive, on average, over the past 12 months," while "real wage growth among the highest income quartile remained robust through mid-2025 but has since slowed sharply." A positive group reading alongside a declining aggregate is not a contradiction, and the Board supplies the reason in the same passage: the group series are built on a different wage measure and deflator, and they "are a 12-month moving average and are therefore lagged relative to actual real wage growth." A moving average of a falling series is still positive for a while after the series turns.
The honest reader payoff is the gap between any national figure and a household's own position, and the Board states it rather than leaving it implied: "The effects on individual households, though, depend in part on workers' circumstances—because nominal wage changes vary significantly across industry and occupation and because households consume different baskets of goods than the one represented in the aggregate PCE price index." Two things vary at once. A worker in an industry where pay rose faster than average, who also spends less than average on whatever rose most in price, can have gained purchasing power in a year when the aggregate figure fell, and the reverse is equally possible.
