Hey,
First, my apologies. There’s a whole bunch of news I’d actually like to write about today, but I can’t because I had a bunch of commitments. That’s why I wrote 90 percent of this particularly un-newsy “news”letter yesterday before much of the news I’d like to write about happened. So I completely understand if you don’t want to read a whole pile of words about, of all things, prices. Indeed, this is one of those G-Files where I think out loud—or technically, as I type—as I try to figure out what I think about something. If that’s not your cup of tea, I completely understand.
I had Ryan Bourne of the Cato Institute on The Remnant last week. I was disappointed to learn he’s not related to either Jason or Randolph, but I was glad to have him on. He’s the editor of a great new book, The War on Prices: How Popular Misconceptions About Inflation, Prices, and Value Create Bad Policy. It’s a collection of top-flight essays on, you guessed it, how popular misconceptions about inflation, prices, and value create bad policy.
One of the things I like best is the title. Because, from a certain perspective—i.e. mine—the war on capitalism or the free market can be understood as a war on prices.
Prices are simultaneously simple—everyone knows what prices are, even if definitions vary a bit—and wildly complex. Socialism—I mean real socialism, not the herring-scented welfare-state capitalism of Scandinavia—doesn’t work because socialist systems cannot accurately price stuff. I could spend the next few paragraphs explaining the “economic calculation problem” and Ludwig Von Mises’ great contribution to economics. But I gotta keep moving. Also, they promised me I wouldn’t have to do math. I’m a words guy.
So, let’s talk about words. One of the problems with understanding the role of prices is that we have one word—“prices”—for two very different things. “Prices” in a socialist system aren’t the same thing as “prices” in a market system. It’s sort of like how totalitarian countries use words like “democratic” and “republic” while meaning something very, very different. China is not a “people’s republic” in any way that theorists of republicanism would recognize (certain French revolutionaries excluded). East Germany wasn’t a “democratic republic,” nor is the People’s Democratic Republic of (North) Korea.
One of my favorite bits in the movie Barcelona is about hamburgers. Set in late 1980s Spain, when anti-Americanism ran really hot, Spaniards liked to make fun of Americans as unsophisticated rubes. Ted, an American expat living in Spain explains to his visiting friend that hamburguesas—Spanish for hamburger—are emblematic of the problem. “Take hamburgers,” Ted explains. “Here, hamburguesas are really bad. It’s known that Americans like hamburgers, so again, we’re idiots. But they have no idea how delicious hamburgers can be. But it’s this ideal burger of memory we crave, not the disgusting imitations you get abroad.”
Prices in socialist countries are hamburguesas, prices in market systems are hamburgers. Just as the hamburguesas look like hamburgers, but don’t do the same job as hamburgers, the price tags in socialist systems—on cars, bread, whatever—look like prices, but they don’t do the same job as real prices.
Alex Tabarrok famously described prices in a market system as “a signal wrapped up in an incentive.” Market prices take into account vast amounts of information and boil it down to a single number. The price of a loaf of bread at your supermarket reflects an amazing number of variables: droughts in one place, rain in another, crop yields in both, fertilizer costs, anti-carb fads, the price of competing products, inflation, retail rent, war in Ukraine, agricultural subsidies, advertising, gas prices, location, the value of specific shelf-space, and on and on. Each of those factors is dependent on other factors with equally long lists of dependent and constantly changing variables. There are so many variables that no expert could grasp them all. But prices convey information without demanding knowledge of all those variables. Prices in a market economy reflect the “is” of a constantly changing economic reality.
In a command economy, on the other hand, prices reflect the “ought” as determined by a handful of experts. In Venezuela, socialist rulers—and those ruled by socialists—believed that a country with roughly 20 percent of the world’s oil shouldn’t have to pay a lot for gas. So, they subsidized the hell out of gas. As Kevin Williamson wrote in 2020, this ended up wrecking Venezuela’s oil industry. Fake prices at the pump, and the corruption and mismanagement they represent, distorted signals and prices up and down the sector, and the whole economy, to the point where the whole thing went kablooey (I’m simplifying things somewhat).
Fake prices are a form of corruption. Because the people setting the prices are putting politics and graft—but I repeat myself—in front of economic reality. The corruption could be theft—gotta move those stolen Incan matrimonial masks quickly—or fraud, or, well socialist (or nationalist!) economics.
Another way to think about it: Prices are only knowable through competition. It’s sort of like if you have only one football or baseball team, it’s impossible to get a real sense of how good the team is, because that can be revealed only by contests with other teams. You can know a lot about the inputs—how fast some players are, their ages, etc.—but until they’re tested against another team you won’t know the score, literally and figuratively. Indeed, scores are a little like prices. They reduce countless variables—conditions of the athletes, morale, strategy, teamwork, etc.—into a single number (for bettors, the odds or Vegas line serve a similar function). That number may not reflect the sum total of the parts, but it’s the number we work with—until the next game.
As von Mises put it, “Where there is no free market, there is no pricing mechanism; without a pricing mechanism, there is no economic calculation.”
It’s personal, and it’s business.
So what’s the point of all this? Bourne and I got into a discussion about “personalized pricing,” which is a form of price discrimination. I am not opposed to all forms of price discrimination. I have no problem with “surge pricing.” When traffic is very heavy during peak hours or when, say, there’s a sporting event or concert, Uber or Lyft will raise prices. Critics call this “price gouging.” But this ignores that whole incentive-wrapped-in-a-signal thing. By raising the price, Uber and Lyft send a signal and an incentive to additional drivers to turn on their apps and make themselves available in the same way that high oil prices encourage oil exploration. The classic example of this is umbrella salesmen in big cities. When it’s sunny out, lots of umbrella salesmen don’t bother to bring their inventory out on the streets. Demand is low because need is low. The only way to sell an umbrella on a sunny day is to price the product so low that people will be willing to buy an unneeded umbrella because it’s such a bargain. When it rains, people need umbrellas, so the hawkers can charge a price that’s worth the effort. That will bring out more umbrella salesmen and umbrellas. It’s an efficient system of resource allocation, which is better for consumers in the aggregate.
With personalized pricing, the umbrella salesman charges rich people more than poor people. On one level that sounds fine, right? Sticking it to rich people and giving a (relative) bargain to poor people sounds kind of virtuous. This is basically how negotiations work. If you want to buy a Persian rug in some Turkish bazaar, the dealer takes measure of you and tries to guess how much he can get out of you. If he thinks you’re rich, his “absolute lowest price” is likely to be higher than if he thinks you’re not so well-off. No one’s holding a gun to anyone’s head, and if both get the carpet they want at a price they can live with, so what? Personalized pricing takes this and, with the magic of computers and animal sacrifice, tries to do the same thing at scale.
Well, I have two problems with this. One is conceptual, the other is … I don’t know the right word. Aesthetic? Emotional? Pre-rational?
So let’s deal with the conceptual one. What happens to the signal? The incentive is still there. The seller has the same incentive he always had—to make a profit. But doesn’t the signal get muddied? Tom Sowell seems to think so:
Price discrimination is both a symptom of a noncompetitive market and a further distortion of economic knowledge, as it conveys different information about the relative scarcity of the same product to different users—causing them to economize differently, and thus at least one of them wrongly.
Brian Albrecht—from whom I got that Sowell quote—disagrees. He thinks price discrimination of this sort conveys—or can sometimes convey—more information, not less. He also argues, persuasively, that price discrimination can be a net benefit to the general welfare. He uses an extended example of a burger joint to illustrate the point. But I’ll offer a different one. When the first mobile phones came out, they were wildly expensive. They cost well more than $10,000 in today’s dollars. “Gouging” the rich—aka “early adopters”—made it possible to cover the massive upfront costs of developing the industry and the technology. Over time, the price went down to the point where, today, you can get a mobile phone—with a camera!—for less than the price of two Happy Meals at McDonald’s. The rich subsidized that. You can tell similar stories about miracle drugs, cars, computers, etc.
But this isn’t quite the same thing as personalized pricing. It seems to me that sticking it to the well-off when developing a new product or technology is different from sticking it to the rich just because you can get away with it. I’m open to Albrecht’s point that both approaches can be a net benefit for society (a Pareto improvement) since pretty much every consensual economic transaction is non-zero sum. The rich person who voluntarily buys an overpriced umbrella still paid only what he thought it was worth to him. Lots of rich people waited for the price of mobile phones to go down before getting one.
But what happens to the signal and the incentive? Instead of finding a price that the market can bear, isn’t the game now to chase the marginal consumer you can overcharge? What does this do to the allocation of resources? Apparently, most economists think this is awesome. In a really wonderful essay for The Atlantic, Christopher Beam writes:
When you think about it, though, dynamic pricing is a pretty crude way to match supply and demand. What you really want is to know exactly how much each customer is willing to pay, and then charge them that—which is why personalized pricing is the holy grail of modern revenue management. To an economist, “perfect price discrimination,” which means charging everyone exactly what they’re willing to pay, maximizes total surplus, the economist’s measure of goodness. In a world of perfect price discrimination, everyone is spending the most money, and selling the most stuff, of all possible worlds. It just so happens that under those conditions, the entirety of the surplus goes to the company.
I’m open to this, really. I’m certainly opposed to heavy-handed regulation that would deny companies the ability to experiment with such things. With one caveat: You need transparency. When you’re haggling with a carpet merchant in the souk, thanks in part to some less-than-flattering stereotypes about carpet merchants, you know the score. But what if you don’t know that some algorithm has decided that you can dig deeper than the next guy and charges you accordingly without telling you?
I’ll give you an example. When I was in Prague last fall, my daughter and I both took out our phones to call an Uber. She was quoted prices—for the same trip, from the same location, at the same time, with the same choice of vehicles—considerably lower than the prices I was quoted. Presumably both fares included profit, but if I called for the Uber, the profit would be greater (I have no idea if the driver would share in the greater profit, but I suspect not). But at no point did Uber tell me “this is the Jonah Goldberg price.”
Imagine shopping in a supermarket where the prices are all displayed digitally. Using AI and some panopticon-style tech, the store knows—maybe from reading my phone—that I’m well-off. The person pushing their cart a few steps in front of me is more down on his luck. As he walks down the aisle the prices for soup, cereal, or Pop-Tarts display one number, but as I walk by the numbers go up. That feels like actual, personal, price gouging, to me.
Now imagine the same scenario, except you’re shopping on Amazon and you can’t see the lower prices—for the same products—offered to someone logging on from a poorer part of town. The price doesn’t convey information about the scarcity of the product anymore, at least not as clearly.
(Hold on a second while I read an ad for Express VPN.)
So far, what I’ve been describing should sound like music to the ears of Elizabeth Warren types. Charging the rich more than the poor? Sign her up! Sure, she’d rather the gouged proceeds all go to the taxman, not the “greedy” corporations. But even here presumably the government gets bigger revenues from sales taxes and corporate profits. Part of me can’t help but see personalized pricing as private sector wealth tax. And Warren loves wealth taxes.
But of course, Warren hates all innovation in pricing because she thinks it will harm poor people as corporations cater everything toward customers with greater lifetime value. I don’t think that’s necessarily an unreasonable concern for reasons Beam gets into in his essay. After all, personalized pricing seeks to limit the bargains for everybody, rich and poor alike, by getting the highest price possible out of everybody. But there’s a greater incentive to keep richer consumers happy.
And that brings me to my second objection. Look, I love the free market. I literally wrote a book about how the ideas that created the free market were a “miracle” that pulled humanity out of crushing poverty and big chunks of humanity out of the clutches of tyranny. But I’ve always conceded that one of the problems with liberal democratic capitalism is that it feels wrong. We weren’t designed to live wholly in the world of contracts and commerce. We are not homo economicus—and contrary to a lot of anti-capitalist polemicists, virtually no champions of the free market ever claimed otherwise.
I have no problem with changing prices. Again, prices need to change to reflect economic reality. But I think there’s something important to the idea that when the prices change, they change for everybody. This was never a universal rule. Friends and family discounts have always been a thing, and always will be. Having “a guy” in this or that industry has always been a good thing. Famous and powerful people have always been able to skirt the laws of supply and demand for things like restaurant reservations. If I call a trendy restaurant and ask for a table for six for this Saturday, they’ll laugh. If I say I’m George Clooney or Barack Obama, they’ll probably ask what time is best.
But personalized pricing feels like it’s taking this sort of thing and implementing it at scale, or at least trying to.
It also feels—note, I’m using the word “feels” because I’m not sure I’m right—like a step backward. Or really two steps backward. The first step backward is that the old special prices for friends and family or for important people is the way pricing used to work before the advent of the modern economy. Price discrimination was just a subset of discrimination, pure and simple. A Jew who tried to buy a horse in czarist Russia might not be able to get one at all, but you can be sure he’d pay a premium if he could. This is why it would have paid to have a horse guy.
The second step backward is that haggling is bad enough, but being haggled without knowing it feels even worse. That’s why I think transparency is so important. I’m not cheap. I’m not even particularly frugal. But man, do I hate feeling like someone is getting over on me. When I travel, I will often stop at a store and buy snacks for the simple reason I feel like a sucker for being charged $14 for the small jar of cashews in my hotel room. I’ll do that even when I can pass the cost onto the people paying for my room. I just don’t want to give Big Hotel the satisfaction. But at least I know the hotel is gouging everybody on the cashews. (Yes, I know, it’s not gouging, it’s simply paying a premium for convenience, but you get me.) I’d be even more pissed if I knew the guy in the next room was getting a deal on the cashews.
There’s just something kind of undemocratic about bespoke pricing—particularly hidden bespoke pricing—based on what some algorithm thinks about each and every one of us. Beam tells the story of John Wanamaker, the Philadelphia businessman who people say invented the price tag in the 1800. He “was a devout Christian whose advertisements promised ‘no favoritism.’ According to a hagiographic history of the Wanamaker empire from 1911, ‘One price to all was neither more nor less than the application to merchandising of the immortal note of equality sounded in the second sentence of the Declaration on Independence.’” Beam also notes that, “The price tag had practical benefits, too: You didn’t have to train employees to haggle.”
Thanks to the technology behind personalized prices, firms are able to get the “benefits” of haggling without the haggle. But it also reintroduces favoritism in new ways. Or at least it feels like that to me.
The free marketer in me says I shouldn’t worry. If this approach doesn’t “work” in some way—because consumers hate it, or find work-arounds (which they will have every incentive to find)—it won’t last. If it does last, it will because it satisfies needs and creates net improvements and efficiencies that are self-justifying.
But, as a thought experiment, imagine if government did the same thing. Yes, we have progressive taxation in the country. The more income you have, the more you pay in taxes, not just in absolute terms but as a share of your income (put aside partisan fluff about Warren Buffet’s secretary for the time being). What if, for “efficiency’s” sake the government implemented “personalized taxation.” It would be pretty obvious pretty quickly that people would get righteously pissed off. California’s experiment with charging higher-income people more for electricity than lower-income people has not gone swimmingly. How much angrier would you be if the taxes were tailored to you personally?
I have no confidence that government won’t try to get into this game. And if it does, corruption will be inevitable. Certain groups will get fake prices. We’d be a long way off from a Chinese style “social credit score” that includes bargains for favored groups, but it feels like we’d be moving in that direction. I could see a future where planners calculate the political price of everything while knowing next to nothing about the economic value.
Or maybe not. I’m still trying to figure it all out.
Please note that we at The Dispatch hold ourselves, our work, and our commenters to a higher standard than other places on the internet. We welcome comments that foster genuine debate or discussion—including comments critical of us or our work—but responses that include ad hominem attacks on fellow Dispatch members or are intended to stoke fear and anger may be moderated.