The AI Boom Is Making America Richer—and Its Workers Broke

(SeaPRwire) –   By: Ethan Gallagher

Two men in charge of America’s economy say the AI productivity boom is coming, it’ll be deflationary, and we can forget about the $40 trillion debt. Treasury Secretary Scott Bessent and Federal Reserve Chairman Kevin Warsh are telling this story publicly. Meanwhile, the workers who supposedly benefit are watching their income share collapse to a record low.

Gregory Daco, chief economist at EY-Parthenon, put it plainly. He said “productivity growth protects margins, not income.” The data backs him up. In the second quarter, economic output grew just 1.7%, driven by only 0.3% more hours worked. Compensation rose 2.6%, but set against a spring and summer of oil-driven inflation, that amounts to flat to slight contraction in real terms. Daco told the audience that this isn’t even the AI boom yet.

Let me lay out what is actually happening. Corporate profit margins hit a record 14.9% of GDP last quarter. The labor share of income fell to 52.8%, the lowest level since the government started counting in 1947. Daco said 50% is not a floor. As long as concentrated gains continue flowing to the capital side within a narrow set of firms, labor’s share can keep plunging further.

Now look at the other side of the ledger. The productivity behind those margins mostly comes from a decade of old-school automation, post-pandemic hiring discipline, and capital spending—not AI. All AI has delivered so far is further concentration. Daco noted that every major technological revolution follows this pattern initially. The railroad boom of the late 19th century. The dot-com revolution of the 90s. Large vertically integrated firms capture the early gains. Smaller firms face persistent cost pressures, policy uncertainty, and higher interest rates.

There was a difference back then. In the 90s, a handful of companies front-loaded capital investment in software, reaped the gains, but then productivity spread quickly through the economy and wage growth followed. The question hanging over today is whether AI follows the same timetable. No one knows.

What we do know is that this boom is uniquely and historically capital intensive. Data center investment is projected to reach $31 trillion by 2050, nearly the size of current U.S. GDP, according to PricewaterhouseCoopers. The Federal Reserve’s Beige Book revealed something stark this week. Construction and manufacturing are roaring right now because of data centers. Without that demand, the industry would be in recession. The engine sputters everywhere else.

That sounds like growth. Companies spending hundreds of billions on equipment that should eventually let the economy produce more with less. But much of that equipment is not made in America. Imports of large computers used in AI servers exploded over the past year. Net imports of GPU servers—the Census category for large computers—hit a $450 billion annualized pace last month. Through 2023, the number was roughly $50 billion a year. Economist Joseph Politano compiled the Census data.

Here is the accounting reality. An imported server adds to investment and subtracts as an import by the same amount. The net contribution to GDP is zero. So the capital spending surge looks massive on paper but translates into almost nothing for domestic output.

Jon Hilsenrath, former Wall Street Journal Fed reporter now advising hedge funds at Serpa Pinto Advisory, wrote: “While U.S. investment is booming, growth in gross domestic product has been modest.” The numbers line up. Margins are enormous. Productivity is ticking up. Hiring is weak. Housing is struggling under tight rates. The labor share keeps shrinking.

The tension for Warsh and Bessent is stark. Do they let the AI buildout run wild, or intervene? Growth is not the same as broadly distributed income. If every dollar of output increasingly accrues to data center owners collecting passive checks, or shareholders who own the data center owners, then the fiscal math gets complicated. The tax base and political constituency associated with a traditional boom grow much more slowly. Fanning suspicion about AI’s benefits is inevitable.

The investment itself carries costs and risk. Hundreds of billions compete for capital in an economy where borrowing is getting more expensive. Higher long-term rates make mortgages expensive and suppress homebuilding. The WSJ published a chart on this this week that was startling.

The AI productivity boom is not failing. That is not the point. The point is that it is not immediately obvious how it boosts labor’s share of income. Daco said it again, and it bears repeating. “I don’t think there’s a floor.”

Author bio: Ethan Gallagher is a Silicon Valley hardware architect and infrastructure strategist with two decades of experience in semiconductor supply chains and enterprise technology deployment.