The Morality of Surge Pricing — II


In part I of this morality tale I looked at some of the effects of Uber’s surge pricing. Uber raises prices when demand is high — Friday and Saturday nights, say, when a lot of people want to drink without worrying about driving or in the particular case I looked at, after the explosion in the Chelsea neighborhood of New York City (which did turn out to be a bomb, evidently) when a bunch of people in that neighborhood suddenly decided they very much wanted to be somewhere else.

When the price of a ride increases during a time of unusually high demand, there are two effects. Some drivers who might otherwise not drive will find it worthwhile to drive. And some riders who might otherwise have requested an Uber will choose not to. They will either postpone their trips or skip them altogether.

The beauty of the price system is that the price steers me toward doing something I might want to do if I had full information. If I am a driver, the higher price rewards me for responding to an increase in demand. I don’t have to know why more people want rides. And if I am a potential rider but my demand for a ride is not very valuable to me, then I will forego or postpone the ride allowing other people to essentially jump the queue.

Suppose you’re thinking of going to a movie on Saturday night in New York City. You don’t know about the explosion. In a world without surge pricing or in a world of taxis, you might request an Uber or call a taxi oblivious to the desperate need of the people in Chelsea. If a frightened person there could get your attention, they would ask a favor of you — look, I know you’re about to get in that cab to enjoy a movie, but I think my life might be in danger. Could you please let me have that cab you’re about to get in and send it my way?

Most of us would gladly step aside. But in the real world, there is no way for the frightened person to get our attention. You get in the cab to go to a movie unaware that you can do someone a desperately desired good turn by delaying going to the movie until the next showing or staying home and watching a movie at home instead. The frightened person cannot flag me down and ask me give up my ride in the cab. There’s no way to reach me.

But when the price of an Uber ride doubles during surge pricing I might conclude that the extra pleasure from the big screen isn’t worth it and I’ll stay home. I’ve stepped aside; I’ve given up my place without knowing why. I might read about it in the paper the next morning.

Part of the outrage over surge pricing comes from the distaste many feel over the idea that Uber itself is profiting from people’s desperation. You look down at your phone, notice surge pricing is in effect and curse out Uber for its greediness. What is harder to notice, because you literally cannot see it, is that the higher prices are drawing more drivers into the area. More people are going to be taken care of which is precisely what is desirable at a time of increased demand. You may or may not be one of them, depending on your willingness to pay for where you are thinking of going.

You literally do not see that if prices did not rise, if Uber did not exist, there is no mechanism for letting people know that more people than usual want rides right now. If Uber did not exist, many people would simply not be able to ride at all because the number of taxis does not adjust to demand. If Uber did not exist, some subset of potential rides would be lucky to have a ride. The rest would be out of luck with no way to signal their desire to party (in the case of a typical Saturday night) or their desire to flee danger (in the case of the Chelsea explosion.)

But there is another level of discomfort with surge pricing. Sure, more people get rides. But money is being used to decide who gets to ride and who doesn’t. So Uber discriminates against poor people, say the critics: the rich get rides during surge pricing while the poor stay home or face danger in the Chelsea case.

One answer to this point is that more people get rides than they otherwise would. The poor people are no worse off. Surely that is better than the taxi status quo. In fact, if it is true that the rich get the surge-priced Ubers while the poor cannot afford them, what that actually means is that there are now more taxis to go around for anyone not interested in an Uber. So surge pricing actually helps the poor indirectly.

But I think the truth is more complicated. Are the rich the only people who use Uber when there is surge pricing? Yes, how rich you are affects your willingness to pay 20% more or 2x more when demand for rides is high. But it’s not the only thing that matters. Surely on the night of the Chelsea explosion, some rich person on the Upper East Side of Manhattan decided to wait a bit before going out while a much poorer person gladly paid an extra $10 or $20 or even $50 to be away from danger in Chelsea.

There is a temptation to think that in a market system where goods are allocated by price, that the rich effectively corner the market on everything while the poor get nothing. But that’s not the way it works. The rich do get all the Maseratis. But there are Hondas and Kias and Hyundais for the rest of us. Supply of stuff isn’t fixed. There isn’t a fixed number of cars that the rich outbid everyone else for. The rich get the fanciest stuff. But even a poor person in American can eat steak now and then. And plenty of hamburgers the rest of the time. Poor people don’t starve in America because the rich hog all the food. A lot of poor people get smart phones in America that let them order Ubers if they want. (Though they may often not have a credit card, but that is another story.)

So on the night of the explosion in Chelsea I doubt the richest New Yorkers took all the Ubers and left the poorer people to fend for themselves. Some of the rich who wanted to leave Chelsea already had private drivers. Some of them chose the fancier kind of Ubers — UberBlack and UberSelect leaving the cheapest Ubers for poorer riders. And some of the richest New Yorkers, far away from Chelsea decided not to Uber at all when they saw the higher prices, unaware they were doing a fellow New Yorker a good turn as the extra drivers incentivized by surge pricing sped to Chelsea.

A final issue to consider are those who are profiting from the high demand — Uber and its drivers. Is it right for them to profit from the desperation of others that night in Chelsea? And of course just as the person far from Chelsea avoids Uber because of surge pricing, not knowing the cause of the higher prices and not having to know, many drivers headed toward danger in search of profit not knowing they were putting themselves in harm’s way. And couldn’t Uber have left prices low that night while paying drivers extra to encourage the increase in available rides? Is it right to profit from desperation and fear? I hope to address these issues in Part III of this series.

A note in closing. I do not work for Uber. I am not an investor in Uber. I’m just interested in how prices work. (You can see my novel on the power of prices, The Price of Everything, here. Lots there on so-called price-gouging there. And here is my conversation with Mike Munger on price gouging from EconTalk. And more on the “stepping-aside” phenomenon and the motivation of people responding to demand during a natural disaster, is here.)

And yes, Uber has apologized for surge pricing in times of crisis and yes, they have agreed under threats from government officials to limit how much price rises in certain cities during surge pricing. I understand why they do that. Whether that is a good idea for the rest of us is what I’m exploring in these essays and in The Price of Everything. I also should add that the reality of surge pricing is almost certainly more complicated than I have made it out to be here. I don’t know Uber’s algorithm and how much they try to match supply with demand. I also don’t know if Uber drivers can help create surge pricing conditions by going off-line. My real interest here is the fundamental of using prices to allocate scarce goods and what is good and bad about that.

Thanks to all the readers who responded to Part I and helped inspire my response here in Part II.

Leave a Reply