Even if the United States is entering a productivity upswing—possibly accelerated by artificial intelligence—workers should not expect it to reverse a long-running decline in labor’s share of national income. The piece highlights a stark benchmark: labor’s share has fallen to 53.8%, described as the lowest since records began, underscoring how much of the country’s income growth has increasingly accrued to capital rather than wages and benefits.
The author’s central claim is about who captures the gains from productivity. In principle, rising productivity can expand the economic pie and create room for higher real wages. But, absent stronger institutions and bargaining power for workers, the cost savings and efficiency gains from AI are more likely to be retained by firms and shareholders—meaning the labor share could keep shrinking even while the economy becomes more productive.
That dynamic matters because household income is the main fuel for consumer demand. Consumer spending accounts for around 70% of US GDP, so weaker income growth for broad swaths of households could become a macroeconomic drag over time. It adds that the impact may be delayed because household balance sheets have been supported by rising asset prices, which boost net worth and help sustain spending even as hiring slows and unemployment rises. Still, households appear to be dipping into savings: the personal saving rate is cited at 3.5%, the lowest in more than three years.
Today’s “productivity-pay gap” is a decades-long structural problem rather than a short-term cycle. It points to analysis from Economic Policy Institute showing that between 1980 and 2025, productivity rose more than 90% while hourly pay rose only 33%, implying productivity grew about 2.7 times faster than wages—evidence that efficiency gains have not translated proportionally into typical worker compensation.
Why has the split shifted?: weaker worker bargaining power (including diminished union influence), ongoing technological change, and globalization. Even if globalization is no longer accelerating as it once did, the author argues the broader trend is unlikely to reverse soon, especially as work becomes more fragmented through remote, contract, and gig arrangements that can reduce collective leverage.
AI is presented as the next test. PIMCO’s Tiffany Wilding is cited as noting AI could lower prices in areas like healthcare and business services, lifting real incomes even if nominal wage growth stays modest. The piece closes on a broader implication: if the economy produces more with fewer workers sharing in the gains, social and political tensions around affordability and inequality may intensify—and, economically, the question becomes who will ultimately buy the output of an AI-driven economy.
