Don't Shut Them. Send Them.

One Wednesday in 1999, at Buick City in Flint, a team leader named Juan came back to my line after a long absence. His attendance was really bad. He started doing jobs at seat install, and car after car began showing up with electrical connectors under the carpet not plugged in correctly. We were building Bonnevilles and LeSabres then.

Left unchecked past the trim gate, those connectors meant cars at the end of the line went to repair, where they waited hours, sometimes days, for someone to get to them. Dispatch slipped, more secondary defects piled on, and some of them went out to customers unchecked.

I walked up to Juan to tell him. He cussed me out and sent me reeling back to my stand-up desk, where I called Rick Stork. Rick gave me a royal tongue-lashing, handed me a couple of courage pills, and ended with a line I still hear.

“You’re a f—ing PhD! Do your job!”

I walked back to Juan and sent him home on discipline. It was a big moment in my growth.

Rick was a supervisor in Trim 1, and someone upstairs had handed him a problem: a new man with a doctorate in industrial engineering who had never run a line. Form him.

Under all of it sat one lesson. You answer for the product and for the people who build it, at the same time, and you are never permitted to trade one against the other.

Christopher Marquis, a business school professor, has argued in The New York Times that business schools should be shut down. I would rather they sent their students to the floor first, for years, not a summer, to be formed the way Rick formed me, by someone who answers for the product and the people at the same time. Don’t shut them. Send them. That’s the argument.

The case Rick would have handed back

Marquis has taught in business schools for more than twenty years, at Harvard, Cornell and Cambridge. His case is that they cannot be repaired.

His sharpest exhibit is a case he has taught many times: Rob Parson at Morgan Stanley. Parson is a star producer whose abrasive, go-it-alone style has alienated people across the firm. Students sit in his manager’s chair and vote: promote him, delay it, or push him out. The dilemma underneath is whether results should be rewarded when the person producing them damages the people around him.

Rick would have handed that case back unread. The dilemma is the lesson. It teaches a future manager that results sit on one pan of a scale and people sit on the other, and that his job is to choose. On Rick’s floor, that trade was the one thing you were never allowed to make. Juan’s connectors were what it looks like when you make it.

Marquis sees part of this. He writes that the case narrows the student’s imagination to one senior manager’s decision and almost never asks whether Parson is himself a product of his industry. That is true. But look at who is still missing from his list of absent voices. He names workers, regulators, communities and ecosystems, and then moves on. Nobody in that classroom is asked to sit in the chair of the person on the line.

Where he is right

The strongest evidence in his essay is ours.

He cites research by Daron Acemoglu and colleagues. In the United States and in Denmark, firms that put managers with business degrees in charge cut wages and shrank labor’s share of income. They did not become more productive for it. They did not grow sales, investment or employment any faster. What looked like managerial skill was value moved from the people who did the work to the people who owned it.

In Built to Extract I call that the Siphon. Marquis has found the office where the permission to run it is issued, and it comes with a tuition bill.

Here is the exhibit he did not use, and it is the most damning one I know. Lincoln Electric in Cleveland is one of the best-selling cases in the history of Harvard Business School. It was first written in 1948 and has been taught for more than seventy-five years. Guaranteed employment, piecework, profit sharing and a worker Advisory Board, all running for generations while the company outperformed its industry. Nearly every M.B.A. in America has read it. Almost nobody has copied it. Fortune ran a feature on Lincoln in 2023 whose headline amounted to: good luck replicating this.

Why? Because the school teaches Lincoln as a decision to admire, not a formation to undergo. You read twenty pages, you debate whether it would work at your company, you vote, and you go to lunch. What makes Lincoln work is three-quarters of a century of supervisors forming people who trust they will not be fired for being productive. That does not fit in twenty pages.

I learned this the hard way, and before any business school could teach it to me.

In 1989, I was working on a doctorate in expert systems, the AI of its day. At Flint Metal Fab, when the big transfer presses went down and the electricians were stumped, GM flew in a troubleshooter named Jim from Chicago. I built a system I called Trades to bottle what Jim knew. It died a natural death. It may have shaved some diagnosis time for a novice, but they still hunted the basement for spare boards, and Jim still flew in. His knowledge was too vast for my boxes and arrows.

A business case is an expert system for executives. It works about as well as mine did.

Where he is wrong

His cure never leaves campus.

Abolish the school, he says. Fold business into sociology, law and public policy. Build what the theorist Martin Parker calls schools for organizing, where students study co-ops, unions and social movements alongside the shareholder corporation. Some of that is good. All of it is still a seminar. In a long essay about what business education gets wrong, nobody is sent to a plant.

He also repeats the founders’ old apology. Medicine had biology and law had case law, he writes