Across 34 monthly readings, private wages outpaced prices. April ended the run.

A run of 34 positive monthly readings ended with March. In April, private-sector pay started buying less than it had a year earlier in real terms — nominal wages rose, but prices rose faster — and through August it still did. The Bureau of Labor Statistics tracks average hourly earnings in the private sector two ways: for all employees, and separately for production and nonsupervisory workers, who make up roughly 80 percent of the private workforce in hourly-wage roles across manufacturing, retail, construction, and services. Read both series month-by-month against the Consumer Price Index from February 2020 through August 2026, and the full record surfaces a divergence between the two series that no single month's report shows. From February 2020 to August 2026, average hourly earnings for production and nonsupervisory workers grew from $24.03 to $32.53 — a gain of 35.4 percent (BLS CES series, via FRED series AHETPI). Average hourly earnings for all private-sector employees grew from $28.54 to $37.75 over the same period, a gain of 32.3 percent (BLS CES series CES0500000003 via FRED). The Consumer Price Index rose 28.9 percent (FRED: CPIAUCSL, from 259.3 to 334.1). Adjusted for inflation against the same pre-pandemic baseline, production workers are up 5.0 percent in real terms; all private employees, 2.6 percent. (Both real gains are computed from the unrounded monthly series levels; subtracting the rounded dollar and index endpoints shown here reproduces them to within about 0.1 point.) The gap between the two averages — the premium of the all-employees figure over the production-and-nonsupervisory figure — narrowed from 18.8 percent ($4.51 per hour) in February 2020 to 16.0 percent ($5.22 per hour) in August 2026. The absolute dollar gap widened while the percentage premium shrank, because production wages grew faster proportionally. In April 2026, year-over-year CPI growth outpaced nominal wage growth for the first time since early 2023. All-employees nominal earnings grew 3.6 percent year-over-year in April while CPI rose 3.8 percent; production workers logged 3.7 percent nominal against the same 3.8 percent CPI. Both series have remained negative in real terms through August 2026. Of the three months reported here, May 2026 showed the widest gap: CPI reached 4.2 percent year-over-year while all-employees nominal growth ran 3.3 percent — 4.17 and 3.34 percent before rounding — for a real gap of -0.8 percent; production workers logged -0.6 percent real in the same month. By August, CPI had moderated to 3.4 percent: all employees -0.3 percent real, production workers -0.1 percent real. The prior positive run for all private employees lasted 34 observable months: from May 2023, when the series first registered positive year-over-year real wages after the 2022 inflation surge (+0.2 percent), through March 2026 (+0.1 percent). (BLS did not publish a standard CPI value for October 2025; that month is excluded from the count. The production workers' positive run began two months earlier, in March 2023, and also ran through March 2026.) The preceding negative run — all of calendar 2022 and the first four months of 2023 — reached -3.6 percent in June 2022 for all employees. The BLS all-employees CES series (CES0500000003) begins in March 2006; the production-and-nonsupervisory series (AHETPI) extends to 1964. All figures are seasonally adjusted. Both series and the CPI deflator are re-pullable at the FRED identifiers cited in sources.

The following is analysis, not fact. Two patterns emerge from reading both series simultaneously across the full arc. The first is timing. Production workers registered positive year-over-year real wages in March 2023 — two months ahead of the broader all-employees measure, which first cleared zero in May 2023. That early recovery is consistent with ^sspeculative the tight hourly-labor markets of 2022-2023 — distribution, food service, manufacturing — closing faster than the more mixed-tenure supervisory workforce. The BLS CES data does not separately publish supervisory wages; the all-employees figure blends production and supervisory workers, so the divergence is observable as a gap between the two averages, not as a direct measure of supervisory compensation. The second pattern is the asymmetry of the current reversal. When CPI re-accelerated from 2.4 percent year-over-year in February 2026 to 4.2 percent in May before moderating, the shortfall was larger for all employees (May: -0.8 percent) than for production workers (May: -0.6 percent) — consistent with the production series continuing to log slightly faster nominal gains throughout 2026. The direction held into August: production workers at -0.1 percent, all employees at -0.3 percent. A third observation is cumulative. After five months of negative readings, production workers retain a net real gain of 5.0 percent over February 2020; all employees, 2.6 percent. The negative readings have reduced that cumulative gain; they have not erased it.

The strongest argument against treating this as a meaningful reversal is scale. The worst of the three months reported — May 2026 at -0.8 percent real for all employees — amounts to less than a cent of purchasing power lost on a dollar of wages. By August, the shortfall had narrowed to -0.3 percent for all employees and -0.1 percent for production workers, approaching marginal. A second counterargument targets the driver: the CPI surge that opened the gap peaked in May at 4.2 percent and had pulled back to 3.4 percent by August. If that acceleration was a one-time price-level adjustment rather than a persistent rate, nominal wage growth at 3.1 to 3.3 percent may be sufficient to return real wages to positive territory in coming months. The August figures alone cannot resolve whether the current run is transient or sustained ^sspeculative.