The Robot Wage

He notices the plastic sheeting first.

Not tarps for a renovation — the kind that mean new drywall and fresh paint. These are clean industrial sheets, sealed with tape at the edges, partitioning off a section of factory floor he's walked past for nineteen years. Behind them, something hums. Not loud. A low, patient frequency, like a machine learning to breathe.

He's forty-seven. He tightens bolts on an assembly line near the Ohio–Pennsylvania border. Has done since he was twenty-eight. Same plant, same shift, same parking spot by the dumpster. When the state minimum wage went up last January, he got a modest raise — seventy cents an hour. He bought his daughter new cleats. He felt, briefly, like the system had remembered him.

Three months later, his supervisor walked him past that sealed-off section. Didn't explain much. "Upgrades," he said. The way people say "fine" when they mean the opposite.

He tells his wife about it at dinner. She asks if he's worried. He says no. But he starts taking a different route to his station — the one that doesn't pass the plastic sheeting. Not because he's afraid. Because the hum is too steady. Too calm. Like it already knows something he doesn't.

Nobody told him the raise and the robots were connected. Nobody had to.

The connection is now peer-reviewed.

A new NBER working paper — #34895, published February 2026 — tracks 240,000 single-unit U.S. manufacturing firms across nearly three decades, from 1992 to 2021.^[1] The authors, led by Stanford economist Erik Brynjolfsson, used confidential Census Bureau microdata linked to customs import records to identify which firms adopted industrial robots — by tracking imports from Japan, Germany, and Switzerland, the three dominant producers.

The central finding is clean and unsentimental: a 10% increase in the minimum wage is associated with an approximately 8% increase in the likelihood a firm will adopt robots.

The methodology matters. This isn't a think-tank estimate or a model projection. It's thirty years of actual firm behavior measured against actual policy changes, using the gold-standard approach of comparing near-identical firms separated only by a state line and a wage law.

The finding replicates internationally. Turkey's 33.5% minimum wage increase in 2016 drove medium and large firms to accelerate robot adoption. China showed the same pattern between 2008 and 2012. Germany's 2015 introduction of a national minimum wage pushed routine-task plants toward automation.^[3] Four countries. Four independent datasets. Same direction.

This is the double squeeze.

In August 2025, Brynjolfsson published a companion study using ADP payroll data showing AI caused a 13% relative employment decline for entry-level workers ages 22 to 25 in AI-exposed occupations.^[4] That was the white-collar end. Now the manufacturing study shows minimum wage hikes accelerating blue-collar automation. Two technologies. Two workforce segments. Two mechanisms. Same direction.

The timing compounds. Anthropic released a report on March 6, 2026, mapping which jobs AI could potentially replace, warning of a possible "Great Recession for white-collar workers."^[5] Moody's chief economist calls this a "Cortés moment" — companies are burning the boats on human labor. The metaphor is dramatic but the data trail is calm and methodical. The displacement isn't sudden. It's structural, cumulative, and moving from both ends of the skill spectrum toward the middle.

The question is no longer whether automation displaces workers. The question is whether policy responses can keep pace with the displacement they inadvertently accelerate. A wage floor that triggers a robot ceiling is not a failure of policy design — it's a signal that the underlying economic architecture has shifted. The seventy-cent raise was real. The robots were also real. Both things were true at the same time, on the same factory floor, for the same worker.

That's not a paradox. That's the new math.