Between 1979 and 2025, net productivity in the United States grew 90.2 percent while pay for the typical worker grew 33.0 percent. Those are figures from the Economic Policy Institute, which has tracked the series for years. Productivity grew nearly three times as much as pay over that period. The gap is the subject, and the interesting question is not whether it exists but what opened it.
What the two lines measure
Precision matters here because the series is frequently misquoted.
Net productivity is output per hour worked, net of depreciation. When it rises, each hour of work produces more.
The pay line tracks hourly compensation, wages plus benefits, for production and nonsupervisory workers. That category covers roughly 80 percent of private-sector employment and deliberately excludes managers and executives. It is a measure of the typical worker rather than the average worker, and the distinction is the entire point. An average that includes executive compensation rises when executive compensation rises, which tells you nothing about the middle of the distribution.
EPI’s productivity-pay gap series documents the methodology.
The period before 1979
The two lines used to move together. From the late 1940s through the 1970s, productivity and typical worker compensation rose at broadly similar rates. That coupling is why the divergence is dated to 1979: it marks where a long-standing relationship broke rather than where a new one started.
Treating the postwar coupling as a natural law is a mistake in the other direction. It was a historically specific arrangement resting on high unionization, a rising real minimum wage, limited import competition in manufacturing, and a particular regulatory environment. Each of those has changed.
Where the output went
If workers produce 90 percent more per hour and receive 33 percent more in pay, the remainder is distributed elsewhere. Economists identify several channels, and they are not mutually exclusive.
Part went to higher earners rather than out of labor entirely. Compensation growth concentrated at the top of the wage distribution, so total labor compensation tracks productivity better than typical worker pay does. That is a distributional shift inside labor’s share, not a transfer from labor to capital.
Part went to capital. Labor’s share of national income declined over the period, with profits and returns to capital absorbing more.
Part is a measurement artifact. Productivity is typically deflated by an output price index and pay by a consumer price index, and the two indexes diverge. EPI and other researchers have examined how much of the gap this accounts for, and the general finding is that it explains some but not most of it. Anyone claiming it explains all of it, or none of it, is overstating.
The scale, in dollars
EPI’s estimate is that if pay had kept pace with productivity, the typical worker would earn $16.40 more per hour today, $13.53 of that in wages rather than benefits.
Read that carefully. It is a counterfactual describing what an unbroken historical relationship would have produced, not a claim that any particular policy would deliver that amount. It is useful as a measure of scale: the gap is not a rounding error, and it compounds over a working lifetime.
What sits at the bottom of the distribution
The federal minimum wage is $7.25 an hour and has not changed since 2009, according to the U.S. Department of Labor. A full-time year at that rate is roughly $15,000 before taxes.
Because the federal floor is not indexed to inflation, its real value declines every year no action is taken. That is a continuous downward movement produced by holding a number still, and it operates on the bottom of the wage distribution while the productivity line keeps rising.
At the other end, EPI puts the CEO-to-worker pay ratio at roughly 290 to 340 to one at large firms. The two facts describe the same period.
Why this changes the framing of household finance
The U.S. Census Bureau put median household income at about $80,000 in 2023. Against that, KFF put the total annual family health premium at roughly $25,000 in 2024 with the worker share above $6,000, and Child Care Aware reports center-based childcare commonly running $10,000 to $17,000 or more per child per year.
A household facing those obligations on that income has a structural problem rather than a behavioral one. The standard response, that people should budget more carefully, assumes the resources are adequate and the allocation is wrong. The productivity-pay series suggests the prior question is whether the resources tracked what workers produced, and for the typical worker since 1979 the answer is that they did not.
This is the argument made by organizations working on cost of living, including the nonpartisan grassroots 501(c)(3) Fight For A Living Wage, which holds that affordability across housing, healthcare, childcare, food, transport and education is the core issue and the wage floor one part of it. The productivity data is one input to that case, not proof of it.
What the gap does not prove
Three cautions are worth stating plainly, because this chart gets used to carry more weight than it can hold.
It does not identify a cause. It is an accounting relationship between two series. Declining union density, trade exposure, technological change, corporate governance shifts, and monetary policy have all been proposed, and economists who agree the gap is real still dispute their relative weights.
It does not prescribe a remedy. Knowing that pay lagged productivity does not tell you whether the effective response is wage floors, sectoral bargaining, labor law reform, tax policy, or something else. Those are separate empirical and political questions.
And the measurement critiques deserve engagement rather than dismissal. The deflator issue is real, benefits are hard to value consistently, and reasonable economists dispute the size of the gap. What has held up across those disputes is the direction: pay for the typical worker grew substantially slower than productivity over a period of more than forty years, and the disagreement is about magnitude.