Why is labor’s share of GDP declining globally?

The short answer

For most of the 20th century, the labor share was thought to be one of the great macroeconomic constants. Keynes called it “one of the most surprising, yet best-established, facts in the whole range of economic statistics.” Across decades and business cycles, roughly two-thirds of US national income flowed to labor compensation, with the remaining third to capital. Cobb and Douglas built their famous production function around exactly this assumption.

This stylized fact broke. Starting in the 1980s and accelerating sharply after 2000, the labor share fell across nearly every advanced economy. The compensation of US nonfarm workers as a share of business sector output dropped from a postwar plateau near 64% to roughly 57% today.

What makes this puzzle distinctive is the concentration of the decline. Three-fourths of the entire post-1947 fall happened in just 16 years (2000-2016). This timing rules out simple stories about technology or trade — both have been operating for much longer.

→ New to macro economics? Financial education hub

What the data shows

The most cited evidence comes from BLS labor share series and the McKinsey Global Institute decomposition.

The empirical context (BLS, McKinsey, IMF, 1947-2024):

• US nonfarm business labor share: stable around ~64% from 1947 through the mid-1980s
• Drop from 63.3% in 2000 to 56.7% in 2016 — McKinsey/BLS
• Three-fourths of entire post-1947 decline concentrated 2000-2016 (BLS)
• 35 advanced economies aggregate: from ~54% in 1980 to 50.5% in 2014 (IMF)
• Developing economies: from 39% in 1993 to 37.4% in 2014
• Roughly one-third of measured US decline reflects accounting treatment of self-employed income (Elsby-Hobijn-Şahin)

The exception worth noting: France and Germany saw labor share recover slightly after 2008, partly reflecting wage moderation policies. The US, UK and southern Europe have seen no such recovery. The pattern is global but heterogeneous.

→ Dataset: US Real Wage Growth Dataset

Why it happens — the macro mechanism

Four mechanisms compete to explain the decline, and their relative weights have shifted over time.

Channel 1 — Globalization and offshoring. When firms offshore labor-intensive production to lower-cost countries, the share of value-added paid to domestic labor falls mechanically. Elsby-Hobijn-Şahin estimate that increased reliance on imported intermediate inputs accounts for a meaningful portion of the US decline, particularly in manufacturing. The pivot here is that the China shock (2001 WTO accession) coincides exactly with the start of the steepest decline.

Channel 2 — Superstar firms and rising markups. Autor-Dorn-Hanson-Katz-Van Reenen 2020 documents that industries that have become more concentrated in “superstar” firms have seen the largest labor share declines. These firms have very high productivity but operate with low labor share — their growth pulls down the aggregate. This dovetails with the De Loecker-Eeckhout-Unger 2020 finding that aggregate markups rose from 21% to 61% above marginal cost between 1980 and 2016. The angle that distinguishes this view: the decline is not uniform across firms but driven by reallocation toward a small number of dominant ones. See our FAQ on monopoly concentration.

A third mechanism operates through capital intensity.

Channel 3 — Automation and the falling price of capital. When investment goods become cheaper relative to labor, firms substitute capital for labor. Karabarbounis-Neiman 2014 estimates that the global decline in the relative price of investment goods explains about half the cross-country decline in labor share. McKinsey identifies depreciation as the second-largest contributor — accounting for roughly 26% of the rise in capital share, driven by faster-depreciating intangibles like software and IP. See our FAQ on intangible capital.

Synthesis by regime. Before 1980, the labor share fluctuated with the cycle around a stable mean of ~64%, with no discernible structural trend. From 1980 to 2000, modest downward drift coincided with the start of skill-biased technical change and early offshoring, but the labor share remained within historical bounds. From 2000 to 2016, the China shock, the maturation of digital automation and the rise of superstar firms combined to produce the steepest decline since BLS records began — falling from 63.3% to 56.7% in a single decade and a half. Since 2016, the labor share has stabilized at the lower level but shown no recovery.

The labor share was supposed to be a constant. It now appears to have been a regime — and regimes change.

→ Analytical framework: Macro-financial regimes

What it means for different economic actors

Equity investors. The flip side of a falling labor share is a rising capital share — which has historically translated into higher corporate profit margins relative to GDP. US corporate profits as a share of GDP rose from ~5% in the 1980s to over 10% by the late 2010s. This expansion is one structural driver of the bull market in equities since the early 1990s.

Workers. The decline is concentrated in routine middle-skill occupations and in industries that have consolidated into superstar firms. High-skill workers in tech and finance have continued to capture wage gains; low-skill service workers have been more sheltered than middle-skill manufacturing workers.

Policymakers. A lower labor share narrows the wage tax base relative to capital income, with implications for fiscal sustainability of pension systems funded by payroll taxes.

A common error is to attribute the entire decline to a single cause. The empirical literature consistently finds that no single channel can explain more than half of the observed fall — meaning the structural shift is a confluence of forces, not a master variable.

Practical observation

Go deeper

Related questions

• Why do monopoly concentrations distort macro statistics?
• How does automation affect wages and employment?
• What is total factor productivity and why has it slowed?
• Why do salary rises feel poorer than inflation suggests?

Frequently asked questions

How is the labor share actually measured?

The BLS computes the labor share as total compensation of employees in the nonfarm business sector divided by sector value-added. The technical complication is the treatment of self-employed income, since proprietors earn both labor and capital returns simultaneously. Different methods can produce labor share estimates that differ by 1.5 to 2 percentage points. Elsby-Hobijn-Şahin (2013) estimate that about one-third of the published US decline reflects how proprietors’ income is allocated, not actual change in worker compensation patterns.

Why has the decline been concentrated in certain industries?

Autor-Dorn-Hanson-Katz-Van Reenen (2020) documents that the labor share decline is concentrated in industries that have become more concentrated in superstar firms — manufacturing, retail, finance and information services. In these sectors, a small number of high-productivity firms with very low labor share have captured market share from competitors with higher labor share. The aggregate fall is driven by reallocation across firms within industries, not a uniform shift across all firms.

Is this trend reversing in the post-COVID labor market?

The US labor share has stabilized but not meaningfully recovered. The 2021-2024 wage surge for low-skill workers narrowed wage inequality at the bottom but did not reverse the aggregate decline. France and Germany have seen modest recoveries since 2008, partly reflecting policy choices on minimum wages and collective bargaining. The structural decline appears persistent in the US absent significant policy or technological change.

Last updated — 12 July 2026