Taxi Medallions, Congestion, and Coase

The economic case against occupational licensing is, roughly, that it creates, implicitly or explicitly, an artificial cap on supply, increasing price for the licensed good. The political economy case against occupational licensing is, roughly, that, since it is done by the government, it is definitionally inefficient. (The technical term, in the discipline, is 'baaaaad.' Related note: If saying 'Government is baaaaad' makes you sound like a sheep, there may be two reasons for that.)

Typically, the artificial supply constraint is inefficient because there is nothing to be gained by limiting, for example, the number of barbers or nail-ladies or African-hair-braiders in a given market. If there suddenly appeared a barber shop on every corner, the market would drive down the price of haircuts until some number of barbers couldn't or wouldn't continue doing business.

The idea of externalities, or spillovers, is that the costs and benefits of some transactions are not captured directly in the transaction. A positive externality occurs when a social benefit accrues to people not party to a transaction, such as the construction of a wind farm to provide electricity. Even those who do not buy the electricity generated by the wind farm will likely suffer from less pollution if there is a move away from coal power. A negative externality occurs when a social cost is inflicted upon people not party to a transaction. A common example is cigarette smoking. Those near to the smoker are exposed to the smoke through no choice of their own.

The intersection of the licensing problem and the spillover problem is the taxi medallion problem. Taxis in New York are capped through the issuance of medallions that the car must display on its hood. There cannot be more taxis than there are medallions. Thus, the cost of medallions can be rather high. At an auction last November, 200 medallions were sold for 'record prices of up to $1.3 million each.' This is a fairly efficient market, so observers can conclude that the buyers of the medallions plan to recoup their purchase price within the bounds of their discounted time-consistent indifference curve between spending now and spending later.

The question is whether taxi medallions should be $1.3 million. Does this not inflate the price of taxi services? Is this not the big, baaaad state expropriating from its sovereign citizens? That depends, of course, on whether or not you remember our little chat about externalities from before.

If Jasper contracts with Theophilus for taxi services to get Jasper from, say, the Cedar Tavern to La Lanterna, and does so at a miraculous time when there are zero cars on the streets in between, it will take ~4 minutes. If there is another car somehere in between, it will still likely take about four minutes. If there are two cars in between, it will still likely take about four minutes. Therefore, the cost of each additional car on the road in terms of Jasper's time is zero, right?

Right?

If there are three hundred cars on the streets between the Cedar and La Lanterna, the cost to Jasper is no longer trivial. Thus, each additional car exacts a cost in each other car. Since taxis make up a large proportion of vehicle traffic in New York, it makes sense to limit their number. The theory is akin to that behind congestion taxes; the revenue raised through traffic-mitigation taxes can be put toward less-congestive infrastructure. 

The challenge, of course, is in identifying an optimal method and magnitude of application. Certainly, in many less congested areas, there is simply no need for such intervention. In Durham, North Carolina, where I live, the only traffic I ever really encounter is around schools at opening and closing times, and at certain times in the very small downtown area. In communities like this, the cost of identifying specific problem areas and implementation of congestion taxes for those areas is likely greater than the social benefit of the taxes. In a congested area, detailed studies of the precise manner of congestion is necessary to determine the exact nature of the problem. Once that is accomplished, a reliable and efficient implementation method must be devised. In a city with the geography of New York could, for example, collect a fee for cars to enter the city and collect a surtax on car registration at addresses in Manhattan.

Many people would call such a plan "heavy-handed." Whether that's because they are drivers who would not appreciate the personal cost, or because such people are inherently opposed to any sort of government intervention is beside the point. Prices determined by an unimpeded market simply do not always incorporate the full social cost and the full social benefit. Ronald Coase believed that such situations were solvable when property rights are comprehensively defined. The congestion problem confounds this formulation, and is only solvable with carefully constructed intervention.