Workers' share of US income hits record low ahead of AI boom
Labor's share of American income has fallen to 52.8%, the lowest since 1947, while corporate margins hit record highs. Economists warn the AI-driven productivity boom may further concentrate gains among capital owners rather than workers.
Workers' share of U.S. income has fallen to its lowest level on record, even before the anticipated artificial intelligence productivity boom takes full effect, raising questions about who will benefit from the coming economic expansion. Labor's share of income dropped to 52.8% in the second quarter, the lowest since the government began tracking the figure in 1947, while corporate profit margins reached a record 14.9% of GDP.
Gregory Daco, chief economist at EY-Parthenon, said the divergence between strong productivity growth and stagnant worker compensation largely predates the AI boom. Productivity gains are protecting corporate margins, not worker incomes, Daco wrote in a note. Economic output grew 1.7% in the second quarter on just 0.3% more hours worked. Compensation rose 2.6%, but when measured against oil-driven inflation from spring and summer, real wages saw flat to slight contraction.
Daco said there is no clear floor for how low labor's share could go. As long as gains remain concentrated on the capital side and within a limited number of firms, labor's share could keep falling, he said. The productivity gains behind these numbers come from a decade of automation, cost discipline following post-pandemic overhiring, and capital spending, rather than from AI itself.
What AI has delivered so far is further market concentration. Technological advances tend to create winner-takes-all environments, Daco said, pointing to historical parallels. During the railroad boom of the late 19th century and the dot-com revolution of the 1990s, large vertically integrated firms initially captured the gains while smaller companies faced persistent cost pressures, policy uncertainty, and higher interest rates.
In the 1990s, a handful of frontier technology companies front-loaded capital investment and reaped the capital gains, but productivity growth from cheaper software spread quickly through the economy, and wage growth eventually followed. There is no guarantee AI follows the same timetable, Daco cautioned, because the current boom is uniquely capital intensive.
Data center investment is projected to reach $31 trillion by 2050, nearly the size of current U.S. GDP, according to PricewaterhouseCoopers. Construction and manufacturing are currently booming because of data center development, with one Chicago manager telling the Federal Reserve's Beige Book that those sectors would otherwise be in recession.
However, much of the equipment powering this boom is not made in America. Imports of large computers used in AI servers have exploded over the past year, hitting a $450 billion annualized pace last month, up from roughly $50 billion a year through 2023, according to Census data compiled by economist Joseph Politano. The GDP accounting treats imported servers as adding to investment while subtracting the same amount as imports, resulting in zero net contribution to GDP.
This helps explain a puzzling element of the AI economy so far: capital spending is booming, productivity is improving, and corporate margins are enormous, yet hiring remains weak, housing struggles under tight interest rates, and workers' share of income keeps shrinking. Federal Reserve Chairman Kevin Warsh has described conditions as loose, while Treasury Secretary Scott Bessent has suggested AI-driven growth could even be deflationary and help address concerns about the $40 trillion national debt.
Jon Hilsenrath, the former Wall Street Journal Fed reporter who now advises hedge funds at Serpa Pinto Advisory, noted that while U.S. investment is booming, GDP growth has been modest. The disconnect raises a difficult question for policymakers: whether to let the boom run or intervene to ensure broader distribution of gains.
Growth is not the same thing as broadly distributed income. If every dollar of output increasingly accrues to data center owners passively collecting checks or to shareholders who own those companies, the fiscal math becomes complicated. The economy may grow richer while the tax base and political constituency typically associated with a boom grow much more slowly, potentially fueling suspicion about the AI buildout.
The investment itself carries costs and risks. Hundreds of billions of dollars in AI spending competes for capital in an economy where borrowing is getting more expensive. Higher long-term rates make mortgages costly and suppress homebuilding. None of this means the AI productivity boom will fail, Daco said, but it does mean it is not immediately obvious how it will boost labor's share of income.
