I was thinking of doing a piece about rent control or a piece about the idea that “corporate greed” is responsible for inflation, but then I read Adam Iscoe’s story in the New Yorker about restaurant reservations.
The Internet has made the mechanics of booking a restaurant reservation much easier, but this is a bit of a tragedy of the commons. Iscoe writes about a New York restaurant called Roscioli which is an outpost of a restaurant in Rome. I can’t speak to the quality of the New York restaurant, but the one in Rome is amazing, get the carbonara. At any rate, because it’s easier than ever to book reservations, the tables all get booked within minutes of opening up:
On a recent Thursday morning, I stopped by Roscioli. Like many hot restaurants, Roscioli usually has tables for famous people, investors, other chefs, and regulars. But most are snapped up on Resy as soon as they become available. Amelia Giordano, Roscioli’s reservationist, invited me to sit with her in the empty restaurant, the walls lined with bottles of wine, and watch her iPad’s screen as the tables filled up for fourteen days hence.
At 10 a.m. sharp, someone booked a four-top, for 5:45 p.m. By 10:01, there were seven reservations, including two for birthdays. Names started madly flashing on the screen. 10:03: “Everything but the later tables have booked,” Giordano said. 10:06: fully committed.
Annoying! But it’s worse than that.
Not only does the internet make it easier than ever to book reservations, it makes it much more logistically feasible to trade reservations. In the past, even if you’d phoned up and booked a table at Roscioli that you were willing to swap to someone else for money, how would they ever find out about it? Today, though, you can bid on reservations at AppointmentTrader.com, and a person who is so inclined can book up hot restaurant tables and earn a living selling them for money.
Which backs us out to the point.
The food and drink at Roscioli has value, and Roscioli can charge money for it. But the opportunity to get a table and eat there also has value. Yet in this case, Roscioli (and other restaurants) don’t charge money for it, and that allows the economic value of the Roscioli table to be captured by people who are only providing brokerage services. It would be much better for popular restaurants, working with companies like Resy and OpenTable, to sell reservations separately from the sale of food. Tables would be more widely available. Restauranteurs would make more money. And you’d be driving economic value to the people who are actually creating value, ultimately supporting the creation of more great restaurants. Explicit prices are a great way of allocating scarce commodities, and trying to allocate via some other method generally creates problems.
Why does Taylor Swift sell out?
A related issue is that when a popular musician like Taylor Swift tours, everyone expects her to sell tickets for less than a market-clearing price, leading to situations where the shows are sold out. But just because a show is sold out doesn’t mean that it’s not possible to get tickets. You just need to buy them on a secondary market:
Still, because Ticketmaster and Ticket One are the official sellers of The Eras Tour, purchasing via the sites is the most secure way to snag a ticket. That said, if you can’t find one, a number of resale sites, including StubHub and Vivid Seats, are also listing tickets. However, in the past, there have been reports of people listing fake tickets, so it’s not a surefire solution.
This seems decidedly worse than selling the tickets for higher prices, having the inventory move a little bit slower, and letting buyers be certain that they aren’t purchasing fake tickets on the secondary market.
People offer a lot of rationalizations for why popular artists don’t charge higher, market-clearing prices, but most of it doesn’t really bear much scrutiny. You’ll hear that a band doesn’t want only rich people to be able to come to the shows or that it’s an act of generosity to true fans or some such thing. But underpricing tickets doesn’t address the underlying scarcity, which is what necessarily deprives some true fans of the opportunity to see the show. In Swift’s particular case, she and her team put a lot of effort into working with Ticketmaster to prevent ticket brokers from amassing inventory, but making tickets totally non-transferrable would be inconvenient to fans, so ultimately a lot of resellers reap windfalls anyway.
These takes also just reflect a somewhat simplistic view on pricing in general.
It’s true, obviously, that a rich person is more likely to be willing to pay $1,000 for a concert ticket than a non-rich person. But the United States is a very affluent country and typical middle class Americans dedicate non-trivial resources to discretionary purchases. The BLS’ Consumer Expenditure Survey says that families with incomes between $50,000-$69,999 spent $2,768 on “entertainment” last year. If you really, really love Taylor Swift, you could make that your only entertainment of the year. Conversely, if you can’t get tickets in the city you live in but you’re rich, you can plan a whole trip to Buenos Aires where tickets were more available thanks to Argentina being in a state of economic meltdown. It’s just not possible to achieve egalitarian economic outcomes by manipulating the price of one specific good — you need to have a more equal underlying distribution of economic resources.
Note that by contrast, there is no objective scarcity of Taylor Swift recordings. An artist could, were she so inclined, make her actual music super-cheap to buy or put the records in the public domain. Nobody does that, which is fine, but I think it underscores that the only real issue here is PR: Artists think it would be bad for their public image to charge dramatically higher prices. And they probably know their business! I don’t blame anyone for being brand conscious. But I do always wonder — if you charged market prices and donated the windfall profits to highly effective global public health charities, thus saving the lives of lots of desperately poor people, would that be so bad for the brand?
At any rate, Swift doesn’t need my marketing advice. The point is that restauranteurs are operating in the same normative environment where they worry, rationally, that it would be bad branding and bad marketing to charge market-clearing prices, and they do not have the kind of publicity power that major musicians have to change perceptions. That leaves them trapped in a bad equilibrium.
Unbundling food and tables
The key thing about restaurants is that food service establishments are really selling two different commodities.
On the one hand, there’s what the customers actually order. But on the other hand, there’s the physical space — the seat. By selling the reservations separately from the food, restaurants could leverage the fact that even if a meal is just as tasty at 10:00pm on a Monday as it is at 8:00pm on a Saturday, the demand for the meals is not the same. Restaurants normally try not to be open at times when they’ll have no customers (that would be a waste of money), but if you open only during moments of peak demand, you’ll be missing out on sales and wasting money on rent, which means they generally have to be open during some of the lower-demand times. If you’re eager to try a place out and you’re flexible, it’s often relatively easy to get in at a low-demand time. But if it’s already Thursday afternoon and you’re trying to plan a weekend date, you’re out of luck.
Selling reservations would help balance this out.
It would almost always be possible to get a table at the time you want for some price, at even the most in-demand restaurants.
Restaurants that are popular would get a revenue windfall from peak-demand reservation sales without raising food prices.
Thrifty diners and bargain hunters could simply go at lower-demand times.
Even more so than in the concert case, the idea that this would only allow for rich people to enjoy the restaurant is pretty clearly false. Younger people (or older people) with more flexible schedules, but more limited budgets could work around the peak demand times. Busy parents who don’t actually get to go out to eat that much would be able to get tables at restaurants they want to visit at the times that work for them. And the financial returns to operating a successful, popular restaurant go up — at least until those returns get swallowed by higher rents, but then we’re just having another discussion about zoning.
I think restaurants, for good reason, are reluctant to rock the boat and try something new here. But precisely because reservations are now primarily being made through big online platforms with national brands, there’s an opportunity for the platforms themselves to introduce pricing rather than allowing the reservation-broker industry to scale up.
Shortages are bad
I mean this all 100 percent seriously: Restaurants should charge for reservations, Taylor Swift shows should be more expensive, and in general, the sellers of scarce items should be given more social license license to charge market-clearing prices rather than to ration. It is good and correct to stigmatize greed (the Michael Douglas speech in “Wall Street” about how “greed is good” is incorrect), but underpricing scarce items is just not a very good way to be generous. You should charge the price that balances supply and demand, and then if you want to be generous with the revenue that follows from that, you should give money away.
But, of course, this is really a post about rent control and the idea that “corporate greed” is responsible for inflation.
I absolutely get where the intuition here come from. We like it when the price of something has a clear relationship to the input prices. If you go to a restaurant, steak will cost more than chicken because suppliers charge more for beef than they do for chicken, and that’s because raising cattle is genuinely more resource-intensive than raising chickens. If we treated chickens more humanely (which we should), that would be more resource-intensive and chicken would become more expensive. If higher prices were fully passed-through in the form of higher wages, that might be acceptable, but if higher prices generate higher profit margins, that seems like an alternative (and more damning) explanation for inflation.
But the question here is less about what is “driving” what, than it is about our policy options. If 10 percent of the country’s oil refineries were temporarily out of service for some reason, gasoline prices would spike and so would refinery profits. You could say profits were “driving” the price increases — statistically speaking, soaring margins would account for the rise in prices — and impose price controls to bring down costs to motorists. But that wouldn’t change the fact that 10 percent of refining capacity was off-line, and instead of soaring prices, you’d see shortages and gas lines. You could imagine a situation in which shortages and explicit rationing are a better idea than letting prices float, but the shortage itself would be a significant problem. In the case of post-Covid inflation, the situation was that the government handed out pandemic assistance very freely, but most people reduced their spending because they were trying to stay home and be safe. That meant most Americans came out of the pandemic sitting on a nice pile of cash they wanted to spend, but when everyone tried to spend simultaneously, it wasn’t possible for production to expand fast enough and prices soared. It’s true that profits also soared, but under the circumstances, the only way to avoid rising prices would have been to create shortages and make things even worse.
The ideal solution, when possible, is to try to identify supply constraints and remove them so that we don’t have a shortage. But when there is an objective scarcity of something, it is almost always going to be better to use prices to allocate resources. In areas like restaurant reservations where strong norms discourage that, you tend to see opportunistic middlemen exploit the opportunity. Having the government create many more similar situations would only make things worse.
I remember trying to explain to my Swiftie cousin at Thanksgiving that you can't really complain that Taylor Swift concert tickets were simultaneously too expensive and also too hard to come by.
It didn't go well.
Matt hits on some interesting points, but I think he’s unaware of economists’ actual theory to explain why some restaurants/concerts/events/etc are persistently underpriced (sell out immediately).
These are what Gary Becker called “social goods”. Plays and restaurants are more fun to go to when they’re perceived as trendy. A big line at a restaurant, or an immediate sell out of tickets, is a signal of trendiness. So sellers of social goods deliberately underprice. If they try to increase prices, they risk falling out of fashion. E.g., if Hamilton increases prices so that markets for tickets will clear, it no longer sells out. Then maybe Wicked becomes the popular play instead, and no one wants to go to Hamilton, leading to empty seats.
(It’s been recognized for a long time that the arguments Matt points out for underpricing don’t make sense. However, upon further inspection, Matt’s take that it’s bad norms/irrationality doesn’t make too much sense either. For example, why didn’t Hamilton increase prices very gradually over time? That probably wouldn’t have been noticeable enough to upset people. This phenomenon also happens for watches/luxury cars /etc. for which it’s definitely feasible to increase prices over time. Overall, I’m really skeptical that Matt just solved a problem that would increase profits for sellers around the world!)