U.S. Employment Costs Rise 0.8% in Third Quarter, Annual Growth Reaches 3.4%
Key Facts
Civilian worker compensation costs in the United States rose 0.8% in the three months ended in September 2026, according to the Bureau of Labor Statistics, in a reading that combines wage growth and benefit costs. Wages and salaries increased 0.7% from June, while benefit costs rose 0.9%, on a seasonally adjusted basis. On a non-seasonally adjusted annual basis, civilian compensation growth reached 3.4% in the 12 months ended in September. Over the same period, wages and salaries rose 3.1% and benefits increased 3.9%. The Employment Cost Index measures changes in the cost of labor after excluding the effect of workers moving between occupations and industries, according to the bureau’s definition.
In the private sector, compensation costs rose 0.8% quarter on quarter, after a 0.9% increase in the second quarter, according to a data correction noted in the bureau’s statement. The third-quarter increase comprised a 0.8% rise in wages and salaries and 0.9% growth in benefits. On an annual basis, private-sector worker compensation rose 3.4% through September. Their wages increased 3.1%, while benefits rose 3.9%, reflecting a difference between growth in direct pay and other compensation costs. The breakdown shows that benefits grew faster than wages in the private sector over the year.
Despite nominal compensation growth, inflation-adjusted wages and salaries in the private sector fell 0.6% in the 12 months ended in September, according to the bureau. That puts the reading for real wages in the opposite direction from the 3.1% nominal wage growth in the private sector over the same period. In the government sector, compensation for state and local government workers rose 0.6% in the three months ended in September. Annual growth reached 3.5% through September. These figures offer a comparison between the course of compensation for government workers and that of their private-sector counterparts, with the two not matching in their quarterly or annual pace.
Private-sector data also showed differences in compensation growth by union status. Compensation for union workers rose 4.3% in the 12 months through September, compared with 3.3% for nonunion workers, according to the Bureau of Labor Statistics. Union workers’ wages and salaries increased 4.3%, while the increase for nonunion workers was 3.0%. The comparison highlights that the gap in wage growth was wider than the gap in compensation growth between the two groups. Across the economy, gross domestic product estimates published a day before the Employment Cost Index report showed real growth at an annual rate of 3.4% in the third quarter.
In price data accompanying the GDP estimate, the personal consumption expenditures price index rose 2.1% in the third quarter, compared with a 5.0% increase in the second quarter, according to the Bureau of Economic Analysis. The personal income and outlays report published on October 29 showed the index rising 3.6% year on year in September, compared with 3.4% in August. The Federal Reserve kept its target interest-rate range at 3.75% to 4% on October 28 and said inflation remained elevated. These data place the labor-cost reading in the context of an economy that saw activity grow while price pressures remained under scrutiny. The Fed is scheduled to hold its next meeting on December 8 and 9, 2026, with its news conference on December 9.
The Employment Cost Index report for the fourth quarter of 2026 is scheduled for release on January 29, 2027, at 8:30 a.m. Eastern Time, according to the Bureau of Labor Statistics’ calendar. It will be the first release covering December 2026 data after the announced methodological change to the index. Starting with the release of December data, the bureau will update the fixed employment weights used to calculate the index. It will also exclude workers’ compensation costs from the calculations. The change identifies methodological elements to consider when reading subsequent releases and comparing them with earlier data.