Last summer, Momentum, a Spanish marketing agency, ran an ingenious campaign: It installed 18 smart vending machines that lowered the price of cold drinks on hot days. A drink bought in 77 degrees Fahrenheit cost 2 euro. If the temperature went above 86 degrees, you had to pay just 1.40 euro.
The experiment in sensor-based pricing was clearly a marketing stunt: What real business would be dumb enough to lower drink prices on a hot day? A business that wants to stay afloat would deploy sensors to do the very opposite. And, short of outright vandalism, consumers wouldn’t be able to do much in protest: The machine can easily tolerate any grumpy complaints.
Momentum got one thing right, though: The proliferation of cheap sensors has made dynamic pricing—whereby the cost can be adjusted in real-time without intervention by the human operator—a tempting option. And while some sensors try to work out environmental factors like the temperature outside, others could concentrate on learning more about the buyers themselves: Are they young? Are they dressed fancy? Are they on Facebook?
The first two can be answered today. In 2011, Intel and Kraft teamed up to launch iSample kiosks that rely on an optical sensor to determine the age and sex of the shopper and then suggest products to serve him or her. The machine was initially used to market Temptations—a jelly-based dessert advertised as “the first Jell-O that’s just for adults.” So, on detecting a child, the machine would ask them to step away. A similar vending machine in Japan relies on facial recognition technology to recommend drinks to different consumers: Men younger than 50 are recommended canned coffee drinks, while women in their 20s are offered tea.
Right now, sensors could help automate simple, binary decisions—don’t let youngsters borrow adult DVDs!—but it won’t take long before they enable interventions of the more elaborate variety: Once our faces can be tied our social networking profiles, all sorts of other manipulations enter the picture. Discounts, yes—but there may also be situations in which our willingness to pay for something is clearly greater than the price we are charged by a dumb, sensorless machine. If the machine can predict those situations—by analyzing our social networking profile or querying the self-tracking app on our phone to find out just how thirsty we are—it can charge us exactly what we are willing to pay.
In theory, at least, there’s much to celebrate here: Sensors will make resource use more efficient—and a new generation of startups will gladly exploit these new efficiencies. Max Levchin, the former CTO of Paypal and a prominent technology investor, said as much in a keynote talk he gave in January at the high-profile Digital-Life-Design conference (a German equivalent of TED). For Levchin, sensors can finally allow us to use “cars, houses, humans, etc.,” to their full potential. “The world of real things is very inefficient: Slack resources are abundant, so are the companies trying to rationalize their use,” he says. But today, thanks to “the digitalization of analog data, and its management in a centralized queue,” one can “create amazing new efficiencies” simply by using sensors to better allocate resources.
Take transportation startups like Uber or Hailo. When you called a taxi service in the past, the dispatcher was supposed to treat everyone the same—first called, first served. If you hung up in anger, you’d have to start all over again. Under this “dumb” system, notes Levchin, “even if you are willing to pay a hundred times more than everyone else waiting ahead of you in line to speak to dispatch, you never get to express that demand. The data exists in an analog-only format, and it moves at analog-only speeds.”
Digital systems like Uber are different: The data about you come in the digital format, and you know exactly when resources become available, how long you need to wait for them, and so forth. And, down the line, perhaps if you are willing to pay more than others, you could get a different, better service. (Uber already uses “surge pricing” when demand is high—like on New Year’s Eve.)
Levchin pushes this logic to the extreme, eagerly anticipating “dynamically-priced queues for confession-taking priests, and therapists”—perhaps they’d charge you based on how desperate you sound. Apps and startups that do just that might be a bonanza for venture capitalists, but the social implications of such zealous optimization are far from obvious. Why is it such a good idea to have someone who’s Facebook friends with Bill Gates be treated differently than someone who is not on Facebook at all?
In the case of dumb cabs, inefficiency wasn’t just an unfortunate consequence of a world without sensors. Rather, it was the logical outcome of regulating providers of transportation services under the “common carrier” principle (also known as “public carrier” principle in civil-law countries). Nondiscrimination is part and parcel of that idea: You are entitled—at least in theory—to the same treatment regardless of whether you are black, white, homosexual, or filthy rich.
Perhaps there are good reasons for abandoning this principle. But the mere fact that we now have better technology for squeezing inefficiencies out of the system is not such a reason: Inefficiency is precisely the price we have agreed to pay for nondiscrimination. To compare the heavily regulated taxi industry with the offshoots of the lightly regulated “sharing economy” like Uber—and to do this on efficiency alone—is to already stack the deck in favor of Uber. The taxi industry was built to be inefficient. (To be sure, some have argued that Uber eliminates taxi drivers’ ability to discriminate when it comes to picking up fares.)
Or consider Airbnb, which Levchin also invokes in passing. The case for Airbnb is well-known—it adds many more housing units on what might otherwise be a tight market. But what about its costs? In allowing tenants to turn their apartments into permanent hotels, Airbnb might be undermining the community spirit of their neighborhood and may even violate some rent-control requirements. (Not to mention the fact that neither Airbnb the company nor its participating hosts seem to be paying all the taxes traditionally levied on hotels. By one estimate, its annual tax debt to San Francisco alone might be $1.8 million.)
Rent controls might be terribly inefficient, but that is by design, not accident: They are implemented to privilege the social and political dimension of housing policy over the economic one. In other words, to say that Airbnb helps improve efficiency is to say very little about the desirability of rent controls. If we don’t like rent control, we ought to oppose it on political and social grounds—and not just by arguing that, thanks to smartphones and social networks, we can create new, more efficient markets for matching short-term renters with tenants.
What’s most disturbing about Levchin’s line of reasoning is that all the inefficiencies of the BS world—that is, “before sensors”—are presented as just natural products of the dumb technological environment. In some cases, they are deliberate efforts, even accomplishments, to promote justice, fairness, or community cohesion. Do we need a new breed of venture capitalists to support those?
This article arises from Future Tense, a collaboration among Arizona State University, the New America Foundation, and Slate. Future Tense explores the ways emerging technologies affect society, policy, and culture. To read more, visit the Future Tense blog and the Future Tense home page. You can also follow us on Twitter.