Treating Car-Hire Services as Public Goods
I love Emily Badger’s Wonkblog piece about Uber’s effect on the taxi industry. Badger addresses many different sides of the issue, discussing the likely impact of Uber on medallion owners, on drivers, on medallion management companies, and on consumers. She is non-dogmatic throughout. She moots a few different potential outcomes, recognizing that much will depend not just on the “market” itself but also on the strategic behavior of the different institutional players. I wish all economic journalism were this good.
I do want to expand on Badger’s treatment of taxi deregulation, and to say a little more about my idea, in my last post, that regulators “treat car-hire services as public goods.” In retrospect, that post jumped too quickly from that idea into a broader argument about distributive policy and politics. That was unfortunate, I think, because the possible regulatory approaches here warrant deeper consideration.
On the first point, Badger notes that various Sunbelt cities deregulated their taxi industries in the 1970s and 1980s, with decidedly mixed effects. Supply went up, but fares also went up, and service seems to have declined. Drivers refused short trips and trips to poorer parts of town more often. Airports became a sort of bazaar.
Why did this happen? Some economic models of the traditional taxi industry suggest that it has pervasive information asymmetries. In the absence of set prices, consumers don’t know how much a standard ride will cost, and may not want to haggle with drivers. Riders and taxis also face high search costs, especially in relatively low-traffic areas. Another peculiarity of the taxi market, related to search costs, is that excess supply (in the form of vacant cabs) can affect demand. Intuitively, if consumers have difficulty finding cabs because cabs are scarce, they may tend not to search for cabs.
Uber has solved both problems, at least for the wealthier users who can utilize it. It dramatically reduces search costs, and its standardized rates prevent price gouging. It also solves a collective action problem among telephone-dispatched drivers. A potential client may take another vacant cab that passes by, and a cab headed toward a call may pick up another fare on the way and never arrive. Under Uber’s watch, though, riders trust that their cars are going to arrive, and drivers can’t take one another’s fares. +1 for private regulation.
None of this means, however, that regulators should allow Uber to continue competing directly with traditional taxis. Regulators should also keep in mind what Uber means for drivers, and what it means for the poor.
First, on the poor. Badger notes that the basic tradeoff within the cab sector has long been restricted entry in exchange for equity-oriented regulations. Medallions grant an exclusive license to operate, but drivers must follow a fare schedule, accept short trips, and not discriminate on the basis of race or neighborhood. Badger also notes that the current structure of the industry—in which drivers lease cars from medallion owners or medallions management companies—actually exacerbates some of the problems faced by the poor in getting service. Drivers have little incentive to answer calls in poorer areas of town, and regulators don’t discipline them often.
Uber might further exacerbate that dynamic. The poor are less likely to have iPhones or other devices needed to e-hail and Uber. More importantly, if UberX drivers continue to work selectively during the most lucrative shifts—when wealthier consumers are coming home on weekend nights—then cab drivers will need to make up that lost income. One natural response will be for cab companies to lobby for higher fares to offset their lower utilization rate. On net, then, Uber would enhance wealthier consumers’ welfare but reduce poorer consumers’ welfare.
Now, I think there are ways to ensure decent service for the poor, but first let me address how to ensure equity for cab drivers. In particular, since entry and price controls seem necessary for a street-hail market to operate, regulators might tweak licensing requirements and property rights to help drivers to capture more of the rents created by those controls. Here are some potential ways of doing that.
First, regulators might make it impossible for third parties to own medallions. Toronto recently took that step. Drivers could then directly capture monopoly rents. Second, cities could permit third-party medallion ownership, but require that owners directly employ any third-party drivers rather than leasing them cars. Such drivers could also unionize— as current drivers cannot—and then press both cab companies and cities to ensure decent rates for drivers.
Third, regulators could literally turn cab service into a public good. A public authority could own or lease cabs and employ drivers to operate them. Those drivers might in turn unionize and force the authority to share any monopoly rents with them.
Now, none of these options are optimal. Drivers who hold medallions could become a classic cartel. A public authority may become corrupt or bureaucratic. A cab drivers’ union might become a partner in ensuring mediocre service. Moreover, any regulations that increase consumer costs will drive consumers toward Uber, threatening a downward spiral in the traditional street-hail sector.
But the alternative may be a cab sector that goes under or fails to serve poor communities. This is why, as I noted in my prior post, I suspect cab drivers are going to need transfers to keep their heads above water, particularly if they depend more and more on poorer consumers.
The real advantage of these three options over third- party medallion systems is that they would enable such transfers without thereby subsidizing medallion owners. The public option might be best, since it would also enable public authorities to more easily direct drivers toward poorer areas of town, supplementing public transit systems more effectively—as is common in Europe. Of course, whether it is possible to create such a public system given vested interests’ power in the industry is another question entirely.