The world’s cities are dying. The diagnosis is heart disease, or, as it’s also known, traffic congestion. The cause of the problem is Uber, Lyft, and other ride-hailing services. The solution to the problem is taxes.
Congestion isn’t new; gridlock predates Uber. Still, before Uber came along, it wasn’t particularly easy to drive around a city in a private vehicle. Either you had to spend thousands of dollars buying or leasing your own car—on top of parking and insurance and gas—or else you needed to navigate an unfriendly and expensive system of cabs and taxis, whose numbers were carefully managed by taxi regulators, medallion systems, and the like. The harder it was to drive, either in money or effort, the fewer people did it and the less traffic congestion there was.
Uber and Lyft changed the entire system by ushering in an explosive rise of cars available for hire. In just a couple of minutes, at the tap of a button on your phone, you can find a car that is both cheaper and more convenient than anything that existed before. The downside of that innovation was an unprecedented jump in traffic. Show me a city where Uber has taken off, and I’ll show you a city where congestion has risen in tandem. That’s true even in cities like London, which were already trying to manage traffic flows by imposing a hefty congestion charge of about $16 per day. There seems to be no limit to how high supply and demand can go—unless, that is, local governments start stepping in.
Felix Salmon (@felixsalmon) is an Ideas contributor for WIRED. He hosts the Slate Money podcast and writes the Cause & Effect blog. Previously he was a finance blogger at Reuters and at Condé Nast Portfolio. His WIRED cover story on the Gaussian copula function was later turned into a tattoo.
The problem is that cities’ standard tools won’t work on the likes of Uber. Up until now, economists’ usual response to traffic has been to implement a congestion charge: set a zone where congestion is a problem, and then charge drivers a fee for driving there. Since most drivers have other ways of getting into town (buses, trains, that kind of thing), the fee nudges many drivers onto public transit, thereby reducing the number of cars sitting in traffic. Congestion charges also raise new money, which invariably gets used to improve public transportation.
Uber, however, breaks that model. Uber drivers aren’t using their car as a means of getting from A to B; they’re using it as a means of earning money. If they took a bus or a train into town that would defeat the purpose: They wouldn’t earn any money at all. Increasingly, they are the alternative to driving into town—only instead of driving in and then parking, taking themselves off the roadway, they drive in and then just continue driving, for hours and hours, making congestion even worse even as they effectively amortize the cost of any congestion fee.
In other words, Uber drivers aren’t like the drivers historically targeted by a congestion charge. While charging them to drive into a crowded zone can certainly raise tax revenues, it’s not going to reduce congestion, because drivers-for-hire are almost entirely price-inelastic. They’re effectively forced to pay whatever the fee is.
What’s more, if you charge Uber drivers, you’re charging some of the lowest earners in the city, people who really need the money they’re making. The tax might be effective, but it would also be regressive. If Uber drivers needed to pay a congestion fee on top of the cost of leasing their vehicle, paying for gas and insurance, and all the other costs associated with driving for Uber, their take-home pay would drop further towards or even below minimum wage. These are not the people you want to hurt, especially when Uber and Lyft give drivers almost no discretion in terms of where they’re expected to drive. If you call an Uber and ask to be driven into the congestion zone, your driver has to drive you into the congestion zone, whether she wants to pay the charge or not.
Would Uber throw a fit if this kind of tax were proposed, and threaten to pull out of any such city entirely? Well, the old Uber, the Travis Kalanick Uber, certainly would. Possibly the kinder, gentler Dara Khosrowshahi Uber would not.
What we need instead, then, is a real incentive for the puppetmasters—Uber and Lyft—to free up road space and get cities moving again.
Such an incentive would not need to touch regular car owners at all, and it wouldn’t even require local governments to define congestion zones or times. All those political decisions about who’s in the zone and who’s out, whether bridges are included, what happens at weekends—all of them could be rendered moot. After all, cities no longer need to work out ex ante where the congestion is going to be: Uber and Lyft have that information, in real time.
And so a tax naturally emerges. Every day, or month, or quarter, whatever makes sense, Uber and Lyft would need to make a tax payment to the city government, based on the number of hours its cars spent stuck in traffic. The tax could be quite simple: 10 cents per minute, say, for any time that any car spent traveling below 10 mph on surface streets or 40 mph on highways. Or it could be more complex, involving a sliding scale of higher payments for slower traffic speeds. Importantly, the tax would be paid by the companies—Uber and Lyft—rather than by the drivers. The companies only take about 20 percent of the total fare paid, so if they wanted to raise fares in order to be able to cover the new tax, that would have the effect of increasing, rather than decreasing, drivers’ incomes.
Such a tax would create all the best incentives. Uber and Lyft would start charging more for journeys in high-congestion areas or at high-congestion times, reducing demand and therefore reducing traffic. (Those increased fares would not be the same as the surge pricing which currently exists, since surge pricing is designed to act as a signal to drivers that demand is high, thereby attracting new drivers to high-demand areas. Congestion-related price hikes, by contrast, would not be linked to passenger demand, and indeed would be a sign that demand was likely to fall in response to higher fares. So drivers would stay away.)
One of the big lessons that Uber has learned—and one reason why the company continues to lose money—is that its passengers are very price-sensitive. When fares go up, they travel less; when fares come down, they travel more. Now that the final fare is clearly visible in the app before you book, it’s easy to change your mind and decide not to take an Uber after all, or to share with others.
If fares were significantly higher for people wanting to journey through a congested area and lower for everybody else, then two things would happen. Firstly, demand for cars would naturally shift in the desired direction. Then, inevitably, supply would too: drivers without passengers would gravitate away from the congested core, towards low-congestion areas which offered a higher likelihood of picking up a fare.
On top of that, the routing algorithms would change as well. Uber and Lyft would have a financial incentive to route cars around high-congestion areas, even if the journeys took a little longer. Meanwhile, people in congested areas who were thinking of ordering an Uber would have a choice: Either pay a bit more and wait a bit longer to get your car, or find alternative means of transportation. That might be unwelcome news to today’s Uber passengers, but it’s exactly the kind of incentive that cities want to provide to their inhabitants.
Would Uber throw a fit if this kind of tax were proposed, and threaten to pull out of any such city entirely? Well, the old Uber, the Travis Kalanick Uber, certainly would. Possibly the kinder, gentler Dara Khosrowshahi Uber would not. (Indeed, Uber and Lyft have not objected to tax proposals currently being considered in cities and states.)”But there’s a way around even this problem: Simply instigate this tax in one of the handful of cities which don’t need Uber as much as Uber needs them—New York, London, Los Angeles. (Based on Uber’s $72 billion valuation, it’s conceivable that New York could be worth $10 billion to the company.) Once it was perfected there, it could more easily be rolled out in Austin, or Melbourne, or Istanbul. It could transform cities in China. And it would have broad public support, since taxing multibillion-dollar global corporations tends to go down pretty well in just about any city.
If you’re a big city, then, and if it’s politically difficult for you to tax all drivers, remember that you don’t need to do that. Just tax Uber and Lyft instead. You can have much the same result, with a fraction of the political downside.
Photograph by WIRED/Getty Images