Dynamic pricing charges for the decision, not the seat

The most expensive week in this magazine's Ledger is the week ending December 28, 2025, when the average Broadway ticket cost $177.88. Nothing about the product changed that week. The chairs were the same chairs, the shows were the same shows, the orchestras had the same number of chairs in the pit. It was Christmas week, and the calendar reached into the average and lifted it by sixty-six dollars over the file's cheapest week, the middle of September 2026, when the average printed $111.74. Sit with those two rows for a moment, because they contain the whole argument of this page. Broadway has learned to price three things at once: the seat, which is the part everybody discusses, the week, which is the part nobody can do anything about, and the buyer, specifically the moment the buyer makes up their mind. The seat is the least interesting of the three. It may be the least expensive, too.

The modern system has a birthday, and birthdays can be checked. In June 2016 the producers of Hamilton raised the price of 200 premium seats at the Richard Rodgers to $849, a Broadway record, while the rest of the house, some 1,075 seats, moved from $139 to $177 up to $179 to $199. Resale sites were asking between $1,660 and $2,535 for comparable rows, and the New York Times estimated that scalpers were taking as much as $60 million a year out of this single show. The lead producer, Jeffrey Seller, explained the arithmetic without flinching: his team had monitored the secondary market, found where the average resale price sat, and set the premium at a level meant to take the motivation out of the scalpers' hands. In the same announcement the production doubled its $10 lottery seats, from 21 a night to 46. Seller again, in words that have followed him for a decade: "In some ways, we're taking from the rich to give to the poor, because there's no question those premiums are subsidizing those $10 tickets." It was the most honest sentence a Broadway producer had said about money in years. It was also, read closely, a description of a transfer. Transfers have two ends.

Give the man his due first, because the defense is mostly true and this page isn't interested in pretending otherwise. A seat the market prices at $2,000 will be sold at $2,000 by somebody, and the only question is whether the somebody is the production that took the risk or a resale operation running bots. Before 2016 the answer, at scale, was the bots. After 2016 it was the show, and the show used part of the money exactly as advertised, on cheap seats at the front of the house. Within a few seasons, premium categories and demand-based pricing were standard practice across Broadway. (The airlines call this revenue management, and the airlines are nobody's model of a loved industry, but the mechanism is the mechanism.) The scalper wasn't abolished. He was outbid by the box office, which is the polite, legal, and considerably more efficient version of the same transaction.

Now the ends of the transfer, because Seller's sentence named only one of them. The lottery he doubled is 46 seats in a house of roughly 1,321, three and a half percent of the room, and the same announcement that funded it also raised the price of the ordinary middle of the house, the 1,075 seats where the merely committed sit. The system's real dividing line is not rich and poor. It is early and late. Commit a year ahead and your price is fixed before any algorithm meets you: DPAC in Durham reports more than 18,000 Broadway season members, and Toronto's Mirvish organization reports some 40,000 subscribers, buying certainty at prices set in advance. Decide on Thursday for Saturday night in New York and you aren't buying a seat so much as volunteering to be measured. The price that greets you is the market's opinion of your hesitation, rendered to the dollar.

We expected our own numbers to embarrass that argument, and we should say so, because we went into the Ledger looking for a surge and found something subtler. The file holds 74 weeks of League grosses, from June 2025 to the week ending October 25, 2026. The average ticket in the first week of the file is $134.77. In the most recent week it is $129.18. Seventeen months of dynamic pricing, and the mean went down. If the practice were a general tax on attendance, the average would show it, and the average shows nothing, because the system's action isn't in the average. It is in the spread: in the top of the market, where Hamilton still sits in most weeks (it took $2,524,951 in the file's most recent week), and in the calendar, where Christmas week costs sixty-six dollars more than September for the identical chair. Dynamic pricing didn't raise the price of theater. It raised the price of particular theaters on particular nights for particular buyers, and it lowered the price, quietly and without press releases, of being the sort of person who plans.

Our own pages contain the consumer's field guide to this arrangement, and rereading it is an uncomfortable experience, because our cheapest-way-in guide turns out to be a catalogue of exceptions: lotteries, rush lines, standing room, the $10 door and the $49 door, every one of them a small raffle or a queue. We published the exceptions and called the result a system, which is the kind of sentence a magazine writes when it hasn't yet done the arithmetic on its own advice. The arithmetic is that the legitimate discounts survive as garnish. They are real, they matter enormously to the people who win them, and they are sized like garnish. The meal is priced by the float.

None of this makes dynamic pricing a scandal, and an editorial that called it one would be selling you a feeling rather than a finding. It makes it a sorting machine, and sorting machines should be judged by what they sort for. This one sorts for commitment, for planning, for the subscriber's temperament, and against the casual, the spontaneous, and the young, the three categories from which future subscribers are traditionally recruited. A pricing regime can be perfectly rational and still eat its own seed corn, one undecided Thursday at a time. People don't fall in love with surge products. They budget for them, they game them, and on the nights the number comes back insulting they stay home. The industry's word for the undecided buyer is demand, as if she were weather. The older word is audience, and audiences are made of evenings exactly like the ones being taxed.

The week ending December 28 will print near $178 again this year, and nobody at any box office will have raised a price by hand to make it happen. The calendar will do it, the way the calendar does everything in this system now, automatically and without a face to argue with. That facelessness is the tell. The $849 chair had a producer attached to it, a man who stood in public and defended the number and doubled the lottery in the same breath, and you could agree with him or call him a villain in print, as plenty of people did. The floating price has no author. It has weather. Broadway spent a century teaching its audience that the ticket was a promise between strangers, and a decade teaching them it is a quote, good until the market moves. Quotes expire. Promises get renewed.

Figures from The Ledger, this magazine's file of Broadway League weekly grosses (74 weeks, June 2025 to October 2026): average ticket $134.77 in the week ending June 1, 2025 and $129.18 in the week ending October 25, 2026; highest weekly average $177.88 (week ending December 28, 2025), lowest $111.74 (week ending September 13, 2026); Hamilton the top-grossing production in the most recent week at $2,524,951. The 2016 Hamilton prices, resale ranges and producer statements: The New York Times, June 2016, as carried by the Associated Press. Subscriber counts: DPAC and Mirvish, as reported by those organizations.