Uber and the cognitive zone of discomfort
If you spend a fair amount of time among privileged dot-com types, you’ll probably be familiar with Uber, a kind of luxury car service for the smartphone era. The idea is that you pull out your iPhone, punch a couple of buttons, and in a few minutes a swanky black car pulls up to drive you to your next destination. You get out, no tipping, and the cost of the fare is automatically charged to the credit card you have on file. Elegant!
You do pay for that convenience. An Uber cab costs significantly more than you’d pay for a taxicab, and I’ve met a lot of people who suffer from Uber sticker shock. I’m one of them, truth be told: after getting charged $43 for my first Uber cab ride, last month, I haven’t used it since. I probably will at some point, when the trip is shorter or when it’s raining or when I’m stuck in the middle of nowhere and there’s no easy way to get a cab. But if you start getting into the habit of using these cars on a regular basis, that habit can get expensive fast.
Uber doesn’t seem to have worked out how it wants to deal with the central question of cost. On the one hand, it’s positioning itself as “everyone’s private driver”: it basically stands in relation to the chauffeur-driven car as NetJets does to the private jet. And compared with the cost of hiring a full-time car and driver, Uber is certainly dirt cheap.
On the other hand, Uber doesn’t like being told that it’s out of reach for people without a lot of disposable income. When Marlooz, a soi-disant “poor freelancer”, said that Uber was “too expensive” for her, the company responded with a 1,750-word data-filled blog post explaining how, even though Uber costs twice as much as a cab, it’s still a good deal. Especially if you’re calling for a cab in San Francisco on a weekend evening, when most of the time the cab you called won’t even turn up.
But the fact is that Uber is too expensive for most people. Hell, taxis are too expensive for most people. Uber is a luxury service, and they charge accordingly. Cab rates aren’t entirely apples-to-apples, but they generally have three components: a fixed base fare, and then a rate for time and a rate for mileage. The meter works out whichever is the higher, and charges you that. And if you compare Uber’s rates to the taxi rates in San Francisco, New York, Boston, Chicago, and Washington, you can see that Uber is a lot more.
|San Francisco||Base fare||Per hour||Per mile|
|New York||Base fare||Per hour||Per mile|
|Boston||Base fare||Per hour||Per mile|
|Chicago||Base fare||Per hour||Per mile|
|Washington||Base fare||Per hour||Per mile|
*includes $0.50 in surcharges
The other thing which becomes clear when you look at these prices is that Uber raises its prices pretty much in lockstep with local taxi rates. The cheapest Uber cabs — the ones in Washington — are still significantly more expensive than the most expensive yellow cabs — the ones in San Francisco. But on an absolute basis, it’s easy to see why people in Washington feel happier grabbing an Uber to get home than people in San Francisco do. If you get stuck in traffic and it takes 30 minutes to get home, that’s $29.50 in Washington; in San Francisco, it would be $45.50.
On top of that, Uber has dynamic GPS-based pricing which automatically charges you on a per-mile basis whenever the car is going faster than 11mph, even if it’s only for a brief period of time. And then of course there’s the fancy surge pricing that we saw on New Year’s Eve, when people started getting charged exorbitant rates — Brenden Mulligan, for example, got charged $75 for a ride which took just 136 seconds, and Dan Whaley got charged $135 to go 12 blocks.
Uber loves to explain its surge pricing with fancy supply-and-demand curves, but you could call it a “rip off drunk people” strategy too. Mulligan has ideas about how Uber’s software could be improved: at the very least, it should display the current minimum fare prominently, rather than just the current multiplier.
But this gets back to my disagreement with Jacob Goldstein and Matt Yglesias about the wisdom of deregulating taxi rates. They reckon that deregulated fares are a great idea, while I think they would be chaotically disastrous. And I think that the experience of Uber on New Year’s Eve — which has resulted in a significant number of refunds for unhappy customers — is important here. Uber’s customers are as savvy and sophisticated as cab passengers get, and Uber was genuinely trying hard to be transparent about pricing. After all, if surge pricing doesn’t reduce demand at peak times, it doesn’t really work. And it only reduces demand if people understand what they’re going to pay.
But Uber got a fair amount of bad press from its New Year’s experiment, and of course its fares 99.9% of the time — just like the fares of other deregulated companies like those in Stockholm — are set and fixed. That’s good business: Uber provides a valuable service by allowing people to know, when they’re out and about, that if they want to call an Uber cab, it will take them home for roughly twice the cost of a taxi.
If Uber pricing was continuously dynamic, prices might well come down during periods of light demand, especially in the early mornings. But our brains hate having to do dynamic cost/benefit calculations. Instead, we rely on simple heuristics: “I should always take the subway if I can”, “cabs in New York are cheap enough that I can take them when I want to”, “cabs in London always cost more than you think they will”, “I can afford Uber if it’s just a short ride”, that sort of thing. When we think about the costs and benefits of various different types of transportation, we don’t actually think in dollars, most of the time, just in terms of a vaguer cheap/expensive spectrum. Dynamic pricing like Uber’s New Year’s experiment takes us out of that comfort zone, and people hate being forced to re-think these things.
It’s nearly always a bad idea, then, for companies like Uber to implement variable pricing: it forces customers to think too much, and it invariably happens on nights when demand is high precisely because lots of customers are inebriated and therefore in no position to drive. Or overthink things.
But from the point of view of the passenger, Uber itself adds a whole new level of complexity to what used to be a relatively simple heuristic. Most of us understand pretty intuitively what the differences are, in terms of costs and benefits, between walking; taking public transport; and taking a cab. But then if you move to Stockholm, or if you start using Uber, things become much more complicated, since now you need to work out the tradeoffs between various different cab options. And those tradeoffs, as Bradley Voytek’s Uber blog post explains, get very complex very quickly, and involve things like whether you’re calling a cab or hailing it, your expected wait time, and the probability of a called cab turning up at all.
It takes a long time to turn all of that into an unthinking heuristic, and in the meantime Uber’s customers will always feel as though they’re ticking the “none of the above” box, rather than simply expanding the menu which is currently hard-wired into their decision-making apparatus. And that I think helps explain why many people remain uncomfortable with Uber, even if they’re exactly the kind of people who should love it.
Uber is a great idea in theory, and the mechanics of it tend to work well in practice. But Alex Rolfe has an important point: if Uber’s prices came down to the point at which they were vaguely the same as a taxi, then we could just lump Uber in with cabs as far as our mental heuristics are concerned. Because Uber’s prices are as high as they are, however, and because when they change they go up rather than down, customers react with snarky hostility.
Uber, in other words, is a car service for computers, who always do their sums every time they have to make a calculation. Humans don’t work that way. And the way that Uber is currently priced, it’s always going to find itself in a cognitive zone of discomfort as far as its passengers are concerned.
Update: Rocky Agrawal adds some very smart comments. A taster:
When people feel ripped off, they don’t want to hear about economic theory or the team of Ph.Ds you have developing optimal supply and demand mechanisms.
Most people have a sense of what is “fair”. Study after study has shown that people will make suboptimal economic decisions in the name of fairness. Product and pricing decisions have to take that into account…
I’m disappointed that Uber didn’t turn New Year’s Eve into a positive marketing opportunity. I would have strongly advocated subsidizing rides with some of the company’s $32 million in new funding (from Amazon CEO Jeff Bezos, Menlo Ventures and Goldman Sachs) to create a delightful customer experience. The company is young enough that it could benefit from positive customer feedback.