📚 What nonfarm labor productivity measures, and how to read unit labor costs

Nonfarm labor productivity is real output per hour worked; unit labor costs are hourly pay divided by productivity. In Q3 2026, US productivity rose 3.6% annualized, unit labor costs fell 0.7%, and labor's share of output hit 53.0%, a record low.

What is nonfarm labor productivity

Labor productivity is real output per hour worked. A bakery that turns out more loaves from the same staff hours has raised productivity, even if nobody looks busier. The national series applies the same ratio to the nonfarm business sector, which leaves out general government, nonprofits, private households and farms.

The Bureau of Labor Statistics (BLS) divides an index of real output by an index of hours. Output comes from the Bureau of Economic Analysis (BEA). Hours come mainly from the Current Employment Statistics survey, with the Current Population Survey covering the self-employed. Unit labor costs (ULC) are hourly compensation divided by productivity, which gives the labor cost of producing one unit of output.

Why it matters

Over long periods, output per hour is what lets real income per hour rise. Over a quarter or a year, ULC shows whether pay is outrunning that gain. The Fed has no productivity target. It watches ULC because labor costs feed the services-inflation debate, and because faster trend productivity raises potential growth and arguably the neutral rate.

How to interpret it

The Q3 2026 preliminary release shows the mechanics. Real output rose 3.7% annualized and hours rose 0.1%, so productivity rose 3.6%. Hourly compensation rose 2.9%, so ULC fell 0.7%. The headline numbers all follow from those three inputs.

Quarterly rates are annualized, so they swing widely. This same release revised Q2 productivity from 1.4% to 2.1%. The year-over-year pace of 2.1% is the steadier signal, and ULC is up 2.4% over four quarters.

FRED line chart of YoY percent change in nonfarm labor productivity and unit labor costs, ~2015–2026, with 2020 recession shading. — After the 2021–22 spike, year-over-year unit labor costs have receded toward the 2% area even as nonfarm productivity growth stayed above its pre-pandemic pace, easing the labor-cost channel the Fed watches for services inflation. Source

Three checks help:

Output or hours? A gain from rising output is healthier than one from cutting hours. This quarter was output, but with hours flat, growth created almost no additional work.
Labor share versus real pay. The labor share hit 53.0%, the lowest since the series began in 1947, yet real hourly compensation still rose 1.1% in the quarter. The two don't conflict. The share compares pay with current-dollar output, while real pay adjusts for consumer prices.
Trend over print. Productivity has grown at 2.2% annualized since Q4 2019, in line with the post-1947 average and above the 1.5% of the previous business cycle. Comparisons across 2020 and 2021 are distorted by the collapse and rebound in hours.

Don't confuse nonfarm business output with GDP, either. It excludes government, households and nonprofits, so the two growth rates differ.

Key takeaways

Productivity is output divided by hours; ULC is pay divided by productivity. Check both against their inputs.
A single annualized quarter is provisional. The year-over-year figure and the multi-year trend carry the signal.
A falling ULC print weakens the case that wage costs are driving inflation. It does not map cleanly to lower yields, though, because stronger trend productivity can lift the neutral rate.
The test comes on December 8, when BLS publishes revised Q3 data. If output is revised down materially, the 3.6% becomes a story about revisions rather than a productivity pickup.