Real Wage Growth: US Average Hourly Earnings Minus CPI Inflation Monthly Since 1964

Real Wage Growth is an Eco3min monthly composite that measures whether American workers’ purchasing power is actually improving. It is calculated as the year-over-year change in average hourly earnings of all private employees (FRED: CES0500000003) minus year-over-year CPI inflation (FRED: CPIAUCSL), available since March 1964. Unlike headline wage numbers, Real Wage Growth isolates the net gain in purchasing power once consumer price increases are subtracted, providing a direct read on the labor share of macroeconomic outcomes.

Dataset: US Real Wage Growth (1964–2026) · Updated —

Source: BLS Average Hourly Earnings (CES0500000003) & CPI (CPIAUCSL) via FRED · Eco3min composite

Macro Takeaway

Construction & Components

Real Wage Growth subtracts the inflation deflator directly from nominal wage growth at the YoY level, producing a series that is comparable across decades regardless of the underlying inflation regime. The Eco3min composite uses headline CPI (not core) because households face headline prices — the deflator should match the inflation they actually experience.

Formula:

Real Wage Growth (YoY %) = Average Hourly Earnings YoY (%) - CPI YoY (%)

Components:

• Average Hourly Earnings of All Private Employees (FRED: CES0500000003) — monthly, BLS Employment Situation release on the first Friday of the month. Year-over-year percentage change is the nominal wage component. Covers all private-sector workers; excludes government employees.
• Consumer Price Index for All Urban Consumers (FRED: CPIAUCSL) — monthly, BLS release in the second or third week of the month. Year-over-year percentage change is the inflation deflator subtracted from nominal wage growth.

Frequency reconciliation: Both components are monthly with synchronized BLS release windows. No interpolation is needed. The composite inherits the slower component’s latency — typically the CPI release, which lags the AHE release by 1–2 weeks each month.

Coverage: March 1964 to present. The start date is constrained by CES0500000003; CPI is available much earlier but the wage series only begins in 1964.

Dataset Overview

Dataset Variables

The CSV and Excel files contain the following columns.

Column names match the CSV headers exactly.

Download the Complete Dataset

The full dataset is available in CSV and Excel formats.

Source FRED Series Used for This Composite

This Eco3min composite is built from two underlying FRED series, both freely accessible:

• CES0500000003 — Average Hourly Earnings of All Private Employees:

  https://fred.stlouisfed.org/graph/fredgraph.csv?id=CES0500000003

• CPIAUCSL — Consumer Price Index for All Urban Consumers:

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

Eco3min computes the YoY transformation and the differential and publishes the aligned composite below.

Direct CSV Access — Eco3min Composite

https://eco3min.fr/dataset/us-real-wages.csv

This URL returns the complete pre-computed composite (nominal earnings YoY, CPI YoY, real wage growth) in CSV format. Use directly in pandas, R, curl, or any data tool — no need to fetch and align the two FRED series manually.

Using the Dataset in Python

import pandas as pd

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

print(df.head())
print(df["real_wage_yoy"].describe())

Using the Dataset in R

library(readr)

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

head(df)
summary(df$real_wage_yoy)

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

Methodology

The Real Wage Growth composite is recomputed monthly by an Eco3min pipeline that pulls CES0500000003 and CPIAUCSL from the FRED API, computes the 12-month percentage change for each series, and subtracts the CPI YoY from the nominal earnings YoY. The pipeline runs on the second Friday of each month, by which point both the AHE release (first Friday) and the previous month’s CPI release have typically been published.

Eco3min publishes both the input components (nominal_earnings_yoy, cpi_yoy) and the derived series (real_wage_yoy) so users can audit the computation or substitute alternative deflators (PCE, core CPI) using the published nominal series.

Data Quality & Provider Notes

Latency is governed by the slower of the two BLS releases. Average Hourly Earnings publishes first (first Friday of the month, covering the prior month); CPI publishes 1–2 weeks later. The composite’s effective latency is therefore the CPI release date plus one business day. Real-time consumers should expect the latest Real Wage Growth reading to refer to a month ending roughly 5–7 weeks before the present.

Revisions to either input propagate to the composite. CPIAUCSL is revised in February of each year (annual seasonal adjustment update); CES0500000003 receives benchmark revisions every February with the QCEW reconciliation. Eco3min republishes the historical composite when either component is revised — historical values may therefore shift modestly compared to earlier downloads.

Alternative deflators exist (PCE, core CPI, MIT Billion Prices, ATL Fed Wage Tracker for the wage side). Bloomberg and Refinitiv publish similar wage-deflated series under various proprietary tickers, but none are freely accessible. The Eco3min composite is the free, source-transparent equivalent. The two are set against each other in comparing nominal wages and real wages across regimes.

What This Index Captures (And What It Doesn’t)

Real Wage Growth is a powerful headline number that disguises significant compositional and definitional caveats. Used correctly, it answers a precise question; used loosely, it can mislead.

What it captures:

• Net purchasing power change for the typical private-sector worker, expressed in YoY terms
• The gap between nominal wage gains and consumer price inflation in real time
• The labor share of macro outcomes across business cycles and inflation regimes

What it does NOT capture (common misinterpretations):

• Composition effects. Average hourly earnings can rise simply because higher-paid sectors gain weight in employment. The 2020 collapse in low-wage hospitality employment mechanically lifted AHE by 4-5 percentage points without any individual worker getting a raise. The Atlanta Fed Wage Growth Tracker controls for this by following the same individuals over time.
• Wage growth for specific workers. The series is a cross-sectional average. Workers in different sectors, regions, and skill brackets experience very different wage trajectories. The headline number cannot answer “did my wages keep up with inflation.”
• Total compensation. Employer-provided healthcare, retirement contributions, paid leave, and equity compensation are excluded. The BLS Employment Cost Index (ECI) is the more complete measure of total compensation; it has shown a wider gap from CPI than AHE-based real wage growth in some periods.
• Differential inflation exposure. CPIAUCSL is a national average. Lower-income households spend a higher share of income on food, energy, and rent — categories that often run hotter than the headline. The “inflation experienced” by the bottom income quintile can exceed the headline by 1–2 percentage points during food and energy shocks.

Use this composite to track the macroeconomic labor-versus-prices contest over time. To answer questions about specific workers or income brackets, pair it with the Atlanta Fed Wage Tracker and the BLS CE Survey distributional inflation data.

Historical Regimes

Reading Real Wage Growth across business cycles reveals the inflation-versus-labor regimes that have shaped post-war US economic experience.

• 1965–1973 (Late Bretton Woods, productivity boom). Sustained positive Real Wage Growth (often 2-3% YoY) as productivity gains and tight labor markets combined with moderate inflation.
• 1973–1982 (Great Inflation / Stagflation). Two oil shocks drove CPI above wage growth repeatedly; real wages stagnated for nearly a decade, falling sharply during the 1974–1975 and 1979–1980 episodes.
• 1983–2000 (Great Moderation). Slow but persistent positive Real Wage Growth, especially in the late-1990s productivity acceleration when real wages briefly outpaced 2% YoY.
• 2001–2007 (Pre-GFC expansion). Modest real wage gains masked by widening dispersion — the median worker saw considerably less real wage growth than the aggregate.
• 2008–2014 (Post-GFC recovery). Weak demand kept nominal wages compressed; falling inflation kept real wage growth near zero or modestly positive, despite high unemployment.
• 2015–2019 (Late-cycle tightening). Real wages rose ~1-1.5% YoY as labor markets tightened and inflation remained subdued.
• 2020–2022 (Pandemic shock and inflation surge). Nominal wage growth surged above 5% but CPI ran 7-9%, producing the largest negative Real Wage Growth episode since the 1970s. Cross-references with CPI inflation and breakeven inflation show how rapidly inflation expectations adjusted during this period.
• 2023–2026 (Disinflation, real wage recovery). Disinflation outpaced nominal wage deceleration; Real Wage Growth returned positive, partially recovering pandemic-era losses. Whether the full cumulative loss has been recouped depends on the deflator chosen.

Related Macroeconomic Datasets

Real Wage Growth sits at the intersection of inflation, labor markets, and household purchasing power. The natural cross-references are the inflation deflators themselves (to substitute headline CPI for alternatives) and forward inflation expectations (to gauge whether nominal wages anticipate price changes).

Macroeconomic Dataset Hub

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

Explore the Eco3min Dataset Hub

Sources

• US Bureau of Labor Statistics — Average Hourly Earnings of All Private Employees (CES0500000003), via FRED
• US Bureau of Labor Statistics — Consumer Price Index for All Urban Consumers (CPIAUCSL), via FRED
• Eco3min Research — YoY transformation and composite calculation

Dataset Reference

Eco3min Research. (2026). US Real Wage Growth (1964–2026): Eco3min composite of CES0500000003 and CPIAUCSL.<br> https://eco3min.fr/en/us-real-wage-growth-dataset/