Sources: BLS Real Earnings, September 2026; U.S. DOL Unemployment Insurance Weekly Claims; BLS Consumer Price Index, September 2026

September pay versus prices

% monthly, SA

-0.50.050.60.1Nominal hourly pay0.6Consumer prices-0.5Real hourly pay
All private-sector employees; CPI-U deflator. BLS preliminary September 2026 observations, released October 14.

Sources: [1] BLS Real Earnings, September 2026 · October 14, 2026; September preliminary

WASHINGTON

Average real hourly earnings for U.S. private-sector employees fell 0.5% in September, the Bureau of Labor Statistics reported October 14. Average nominal hourly pay rose 0.1%, but the seasonally adjusted consumer-price index rose 0.6%. Real weekly earnings also declined 0.5%, with the average workweek unchanged. The preliminary result describes the buying power of an average private-sector pay measure; it does not say every employed person received the same nominal increase or experienced the same inflation rate. [1]

More dollars, less purchasing power

The published nominal hourly average increased from $37.76 in August to $37.81 in September. Average weekly earnings moved from $1,298.94 to $1,300.66 as the reported workweek stayed at 34.4 hours. In constant 1982–84 dollars, however, weekly earnings declined from $388.75 to $386.76. The two descriptions are consistent: a small rise in money income was insufficient to match the measured increase in consumer prices. Differences in rounded growth rates should not be treated as unexplained discrepancies. [1]

The annual comparison

Real hourly earnings were 0.7% lower than in September 2025. A 0.6% increase in average weekly hours partly offset that deterioration, leaving real weekly earnings down 0.1% over the year. For production and nonsupervisory employees, real hourly earnings fell 0.6% in September and real weekly earnings fell 0.5%. That group uses CPI-W as its deflator, whereas the all-employee measure uses CPI-U. The different populations and price measures limit simple one-for-one comparisons between the two series. [1]

Claims measure a different margin

Initial unemployment-insurance claims were 199,000 in the week ended October 10, on a seasonally adjusted basis, according to the Labor Department’s October 15 release. The prior week was revised to 198,000 from 197,000. The four-week average rose only 250 to 198,500. This evidence does not show an abrupt national surge in new benefit applications. It also does not establish that hiring was strong or that people already unemployed found jobs quickly. [2]

Continuing claims and eligibility

Insured unemployment was 1.710 million in the week ended October 3, down 4,000 from a revised 1.714 million. Its insured unemployment rate remained 1.1%. That rate applies to covered employment and is not the unemployment rate from the BLS household survey. Continuing claims also carry a one-week observation lag relative to initial claims. Benefit eligibility, claims processing and the end of entitlement affect what is counted; the series is not a census of all people without work. [2]

Energy’s pressure on the paycheck

September CPI-U rose 3.7% over twelve months. Gasoline increased 9.1% during the month and accounted for more than half the overall monthly rise; food prices were unchanged. [3] A worker’s transport needs therefore matter when interpreting the average. The combination of relatively steady initial claims and falling real pay is not contradictory. One series concerns a route out of employment into benefits; the other concerns what the pay of those employed can buy. Neither alone describes a household’s complete balance sheet.

Sources & methodology

Bracketed numbers refer to the sources below. Analysis is original editorial interpretation, not a personalized recommendation.

  1. BLS Real Earnings, September 2026 · October 14, 2026; September preliminary
  2. U.S. DOL Unemployment Insurance Weekly Claims · October 15, 2026; October 10 / October 3 observations
  3. BLS Consumer Price Index, September 2026 · October 14, 2026; September observations