The Price Tag Was a Machine

First principles ·

The Price Tag Was a Machine

The fixed price feels like a law of nature. It is younger than the light bulb, it was built to solve a staffing problem, and the problem just went away.

Tonight, 11:48 p.m.

You price a flight on your laptop, then again on your phone to be sure. Two numbers. Same seat, same day, eleven dollars apart. You feel cheated, and you cannot say by which number — the high one, the low one, or the idea that there was ever one number at all. You buy the cheap one and close the lid harder than you need to.

For most of the history of trade, a price was a conversation. The bazaar, the market square, the general store: the merchant sized you up, you sized up the merchant, and the number you settled on belonged to the two of you alone. Haggling was not a quirk of faraway places. It was the default of the species, on every continent, for thousands of years.

The exceptions were religious before they were commercial. Quaker shopkeepers posted one price for everyone because they held it dishonest to charge two people differently for the same good — a store where a child could be sent with a coin and come home with exactly what her mother would have gotten. For two centuries that stayed a testimony, not a technology. It did not spread, because it did not need to.

Then the testimony became machinery

The department store changed that. A floor with a hundred thousand items and hundreds of clerks hired last Tuesday cannot haggle. The owner can trust his own judgment at the counter; he cannot trust an army of strangers to exercise it a thousand times a day without losing the margin, the inventory, or the customer. So in the 1870s the price moved out of the clerk and onto a slip of paper tied to the goods. Macy did it in New York — himself a Quaker. Wanamaker did it in Philadelphia, one price for everyone and your money back if you regretted it.

What they built was not honesty. Honesty was two hundred years old. What they built was throughput. Clerk training collapsed from an apprenticeship to a week. A price could be printed in a newspaper, which made modern advertising possible. A return could be checked against a record. And by 1916 the self-service grocery had removed the clerk entirely, leaving the tag to do all of the talking.

Economists later wrote the result into theory: in an open market, one good converges toward one price. Jevons called it the law of indifference. But read the history in order and the law shows its true shape. One price never held because markets were perfect. It held wherever asking was expensive. A posted price is negotiation, amortized — the store bargains once, with everyone at the same time, and both sides surrender the tails. The merchant gives up the buyer who would have paid more. You give up the deal you might have talked your way into. In exchange, nobody spends their life negotiating toothpaste.

Figure 01

Three ages of the number

  1. The testimony. Quaker shops post one price because two prices are a lie. A moral position, adopted by almost nobody.
  2. The machine. Department stores adopt the tag because a thousand hired clerks cannot be trusted to haggle. The moral position turns out to scale.
  3. The expiry. Software negotiates for both sides at no cost. The tag keeps its place on the page and loses its reason to exist.
The first two ages are history. The third is my argument, not a record.

The reason just expired

The tag’s reason was the cost of conversation, and that cost has fallen to nothing. Your AI can ask. It can ask every merchant, every day, in parallel, politely, without embarrassment and without fatigue — the two taxes that actually kept you from asking. The merchant’s AI can answer, and its answer can be tuned to the moment: the inventory, the hour, the season, you. Negotiation stops being a skill you have or lack and becomes a background process, like spam filtering. And when haggling is free, the machine that existed to amortize it has no work left to do.

Here is the part most people miss. The posted price will not die loudly. It will die semantically. The number stays on the page; it just stops being the price. It becomes the opening position — what hotels call the rack rate and car lots call the sticker — an anchor for the humans and a first move for the machines. The real price retreats into the pairwise exchange: it exists between your agent and theirs, for a moment, for you. One price per good becomes one price per relationship. Which is to say: the bazaar, rebuilt at planetary scale, running at machine speed, with nobody raising their voice.

The number stays on the page. It just stops being the price.

Notice that this is not a market failing. It is a market finishing. The old law assumed asking was expensive for everyone; it never imagined asking becoming free for everyone. Complete information plus free bargaining settles somewhere the textbook never priced: a different number for every buyer.

And the tag was carrying a second job nobody ever assigned it. One price protected the shy, the hurried, the polite, the busy — everyone who would rather overpay than argue. The tag was a fairness device by accident. It taxed no one for being bad at a game they never agreed to play. Take it off and the game returns, except the skill now belongs to software. Your price level stops being a fact about the market and becomes a property of your representation.

A Tuesday, 3:07 a.m. — soon

Your delegate reopens the internet bill, cites the competing offer it solicited an hour earlier, and accepts the retention desk’s third counter. The retention desk is also a machine. The whole exchange takes four seconds. By breakfast, three of your prices have moved, and none of them crossed your attention. The ledger reads like weather.

What I expect

  1. The posted price becomes the rack rate.

    Listed prices keep their fonts and lose their meaning: opening positions, not promises. “Nobody pays the sticker” escapes the car lot and spreads to everything with margin. The page shows a number; the receipt knows better.

  2. Subscriptions turn into flows.

    Anything billed monthly — software, insurance, storage, the gym — gets re-settled continuously by delegates, because set-and-forget pricing only ever worked because customers forgot. Agents do not forget. The renewal notice stops being a bill and becomes a bid.

  3. One price returns as a brand.

    Some merchants will take the Quaker position on purpose — one number, for every human and every agent, take it or leave it — and some will win on it, because a fixed price is the cheapest trust a business can buy. The tag’s last job is a signal, the way a handwritten letter now signals precisely because it is inefficient.

  4. The haggling tax comes back, inverted.

    The unrepresented pay it. A person without a capable agent pays quietly, persistently more — the new version of not knowing you were allowed to ask — until posted-price protections for consumers get argued in public, the way price discrimination between businesses was argued in 1936.

  5. Prices go private.

    The spread between what two neighbors pay for the same thing widens while each of them sees only their own number. “What did you pay?” becomes an interesting question again, as it was in every bazaar in history, because the answer stops being printed on the shelf.

I run a small business, and I write a number on the page like every other merchant. But watch what I have become as a buyer. I told my own AI you decide a while ago, and last week I wrote that the same work already carries two honest prices while AI reprices the economy from the inside out. This piece is where those two sentences meet. The entity that decides for me is also the entity that asks for me, and the asking now costs me nothing. So I have stopped reading posted numbers as prices. I read them as opening claims. That habit, multiplied by everyone, is the whole argument.

How this could be wrong

Already true
Fares, rides and online listings already reprice by the hour, and repricing software changes a number more often than any clerk ever could. A loyalty coupon is a personal price wearing a costume. Large buyers already let software negotiate supplier terms end to end. And the one-price world was always softer than it looked: the United States needed a law in 1936 — Robinson–Patman — to police price discrimination between business buyers, which tells you discrimination is the natural state, not the aberration.
What has to happen
Agents need the authority to close, not just to ask: standing budgets, payment mandates, an identity that lets an offer bind. Merchants need a door for negotiation that is not a human retention call — an endpoint, not a hold queue. And both sides have to be staffed. One-sided automation does not end the tag; it just breeds better coupon-clippers.
Where I am probably wrong
Trust may beat discrimination. Customers have revolted against personalized prices before, loudly, and a merchant’s cheapest answer to that anger is the posted price. If the resentment is strong enough, the equilibrium is the Quaker store at scale — one public number as the dominant strategy, with negotiation confined to the edges where it already lives. The law could freeze it there too. My failure condition is plain: if in ten years most consumer receipts still match most shelf tags, I was wrong, and the tag’s job was never only cost. It was peace.

Go back to the laptop at 11:48. Two numbers for one seat felt like a glitch, and you reached for the word cheated because you were raised inside the machine — raised to believe one price is what honesty looks like. But the tag was never the honesty. It was the armistice: a truce between a merchant who could not afford to ask and a customer who hated asking. The truce held exactly as long as asking stayed expensive, and asking is now free.

The conversation that built every market before this one is coming back.

You just won’t be the one having it.

Background

Author: John Rector

John Rector is a Charleston-based entrepreneur, author, and AI strategist. He co-founded E2open, the supply-chain software company acquired for $2.1 billion in 2025, and in 2026 opened Charleston AI, a 3,000-square-foot lab that helps people and organizations understand and use artificial intelligence. He is the creator of The Reality Equation — a lecture series, book, and curriculum exploring attention, prediction, and how reality is experienced — and the author of more than two dozen books. He writes and speaks widely on artificial intelligence, attention, and the future of human work.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.