OPHNFB: US Labor Productivity YoY Growth from FRED (1948–2026)

The US Labor Productivity dataset serves year-over-year growth in output per hour worked for the nonfarm business sector, quarterly since 1948 — the single variable that ultimately governs living standards, the inflation-safe speed limit of wage growth, and the r-versus-g arithmetic of debt sustainability. The eras are stark: 2.8% average growth in the postwar golden age (1948–73), 1.5% in the long slowdown, 3.1% in the IT boom (1996–2004), 1.5% in the 2005–2019 stagnation — and 2.5% on average since 2023, the strongest sustained run in two decades.

Dataset: US Labor Productivity Growth (1948–2026) · Updated 2026-01-01

Source: US Bureau of Labor Statistics, Productivity & Costs · FRED series OPHNFB

Macro Takeaway

Dataset Overview

Dataset Variables

The CSV and Excel files contain the following columns.

A value of 2.80 means output per hour rose 2.8% over four quarters. Note: the chart above displays the underlying index level from FRED; the CSV serves the year-over-year growth rate.

Download the Complete Dataset

The full dataset is available in CSV and Excel formats.

FRED Direct CSV Access

The underlying data is available from FRED under series code OPHNFB:

https://fred.stlouisfed.org/graph/fredgraph.csv?id=OPHNFB

Direct CSV Access — Eco3min Structured Dataset

https://eco3min.fr/dataset/us-labor-productivity.csv

This URL returns the complete dataset in CSV format. It can be used directly in pandas, R, curl, or any data tool.

Using the Dataset in Python

import pandas as pd

url = "https://eco3min.fr/dataset/us-labor-productivity.csv"
df = pd.read_csv(url, parse_dates=["date"])

print(f"Latest productivity growth: {df['productivity_yoy'].iloc[-1]:.2f}% YoY")
print(df.tail())

Using the Dataset in R

library(readr)

url <- "https://eco3min.fr/dataset/us-labor-productivity.csv"
df <- read_csv(url)

tail(df)
summary(df$productivity_yoy)

Both examples load the dataset directly from the URL — no download or API key required.

Methodology

Labor productivity is real output per hour worked in the nonfarm business sector, computed by the BLS from national accounts output and hours from the payroll and household surveys. It measures average output per hour, not worker effort — capital investment, technology, and workforce composition all move it.

This dataset serves the year-over-year growth of the index (2017 = 100). The BLS headline quarter-over-quarter annualized figure is among the noisiest statistics in macro; the YoY series is the standard for reading trend.

This dataset is updated daily (Mon–Sat, 08:00 UTC) via automated pull from the FRED API; new observations appear with each source release.

Data Quality & Provider Notes

• Release latency. Published with the quarterly Productivity and Costs report, ~5 weeks after quarter end.
• Revisions policy. Substantial — output and hours both revise, and benchmark revisions have historically rewritten entire productivity eras (the late-1990s boom was initially underestimated).
• The 2020 artifact. Measured productivity spiked in 2020 because hours collapsed faster than output and low-productivity sectors shut down — a composition effect, not efficiency gains. Treat 2020–2021 YoY readings with caution.
• Trend vs noise. Even YoY growth swings ±1.5pp on measurement; multi-year averages are the meaningful unit for structural claims.
• Known gaps. None; continuous since 1947 (YoY from 1948).

Common Pitfalls When Using Productivity Data

1. Calling a regime change on two years of data. Every productivity ‘boom’ since 1995 was declared within eight quarters; only one survived revision and time. The 2023–2026 run clears the bar of past false dawns only marginally so far.
2. Reading the 2020 spike as gains. The pandemic surge was arithmetic — hours fell faster than output. It reversed as hours normalized.
3. Confusing labor productivity with efficiency or effort. Output per hour rises with more capital per worker, sectoral shifts, or labor shedding; it is not a measure of how hard people work.
4. Quarterly prints as signal. QoQ annualized productivity is noise-dominated; the YoY series served here is the minimum viable smoothing, multi-year averages the honest one.

Historical Regimes

1948–1973 — The golden age. Growth averaged 2.8% for a quarter century — the postwar capital deepening and technology diffusion that doubled living standards in a generation.

1974–1995 — The great slowdown. Average growth halved to 1.5%; the causes (oil shocks, measurement, exhausted catch-up) remain debated. The era that broke the wage-price balance of the golden age.

1996–2004 — The IT boom. Growth averaged 3.1% as computing diffused through the economy — the productivity dividend that let the late-1990s run hot without inflation.

2005–2019 — The stagnation puzzle. Back to 1.5% average, with the 2011–2019 stretch among the weakest on record despite the digital revolution — the ‘Solow paradox’ redux that defined secular-stagnation debates.

2020–2022 — Pandemic whiplash. A composition-driven spike, then payback: measured growth swung from +6% to negative as hours normalized — two years of statistical fog.

2023–2026 — The tentative boom. Growth averaging 2.5%, running at 2.8% in early 2026 — the strongest sustained stretch since the IT era, coinciding with the AI investment cycle. Whether it is the third great acceleration or another cyclical rebound is unresolved.

Related Macroeconomic Datasets

Productivity converts wages into unit costs and growth into fiscal capacity; the series below are its main interfaces.

Macroeconomic Dataset Hub

This dataset is part of the Eco3min macro-financial data repository.

Sources

• US Bureau of Labor Statistics — Productivity and Costs, nonfarm business sector
• Federal Reserve Bank of St. Louis — FRED series OPHNFB

Dataset Reference