Pulling the Rug Out From Under People Is Unfair, Except When It Isn't

by Neil H. Buchanan

Early last month, I wrote a post here on Dorf on Law critiquing the misnamed "sharing economy" concept, in particular the business model behind the taxi-substitute company Uber.  My bottom line in that post was that Uber and companies like it are not at all examples of a new way of doing business, but are instead rather blunt methods of evading the law and shifting costs onto workers.

If we continue to allow Uber and similar firms to evade laws regarding insurance, consumer protection, employee protection, and so on, then the companies that try to compete under the laws that actually exist will, of course, be at a severe disadvantage, and could be destroyed.  That will not, however, be a triumph of the internet-based economy, any more than bringing a gun to a knife fight proves that one is a superior fighter.

That post elicited two especially interesting ideas on the comments board.  One commenter noted a further way in which the Uber business model does what I said it does, except more insidiously and dangerously.  That is, Uber is telling its potential new employees -- Oh wait, they're not employees at all, they're independent operators who just happen to benefit from Uber's ride-sharing app technology!! -- that they can make a bunch of money by providing Uber rides during the down time when their cars would not otherwise be used (and when the driver would, if the stereotypes are true, otherwise be hanging out in a hipster coffeehouse, writing a terrible movie script).

As the commenter noted, however, this is one of the biggest hidden costs of being an Uber driver, because the normal lifespan of a personal car, and the normal maintenance requirements of such a car, are in fact based on the car being driven rarely.  If the average car is driven 10,000 - 15,000 miles per year (as car warranties usually specify), with a combination of city and long-distance driving, then signing up for Uber is a pretty rude surprise.  Taxis need to be maintained continuously, and inspected much more frequently than personal cars do, but in the words of the commenter: "Ask any taxi fleet operator about maintenance cycles for its vehicles... and then ask if Uber has disclosed any of that to its drivers."

To which I can only say, "Thank you."  This is yet another example of how costs can be well hidden, and why it is important to be skeptical of knee-jerk complaints about "needless regulations."  As bad as regulated transportation can be, unregulated transportation is worse.  As I noted above, allowing a faux-innovative business to operate outside the law has the potential to kill the regulated businesses, yet we might end up allowing this to happen simply because the hyper-aggressive owners of Uber have convinced people that the internet magically erases the laws of business.

I want to devote the rest of my comments, however, to the other interesting idea that was batted about on the comments board.  Professor Dorf noted that cities generally limit the number of taxi medallions (which are essentially operating licenses), to the point where the medallions become highly prized pieces of property.  The city sells medallions to taxi companies that meet certain requirements, and the owner of an existing medallion can (with the city's approval) sell the medallion to a new operator.  As Professor Dorf then noted: "But of course, if the city is artificially constraining the number of taxis below the equlibrium level, then the answer should be to create more medallions or eliminate the need for a medallion entirely ..."

One of our regular commenters agreed with Professor Dorf, but added that "he ignores the horrific economic penalty that existing holders of quasi-monopoly positions have when that quasi-monopoly is eliminated and free competition is introduced. In the situation with medallions, as he points out, some individuals (and investors) have paid $1million or more."  This argument led me to think about the nature of reliance interests, especially in the realm of proposals to change public policy.

Note that the taxi medallion is a particularly vivid example of something that can change in value due to government policy.  As the commenter noted, "The value of these medallions is created solely by government policy."  That is true, but recall that the presence or absence of a premium on taxi medallions depends on the interaction between government policy and the market for rides in the city.  But that is going to be true of every piece of property, the value of all of which is "created solely by government policy" in the same sense.  For example, my ability to buy and sell zoned residential property with a house on it is predicated on all of the rules that make such ownership beneficial.  Change those rules, and the value of the property changes.  (This all but begs for a discussion of "regulatory takings," but that will have to wait for another day.)

This, in turn, is merely a useful example to show that every law ever passed both creates and destroys economic value, and that the person who owns a particular piece of property under any set of laws is always at risk of losing his shirt, if the laws change.  I will return to the "equilibrium" aspect of this in a moment, but first, consider another example, which will allow us to ask who gains and loses when the government changes policies.

Rent control laws are very unpopular among orthodox economists.  In fact, in Introductory Economics textbooks, including those written by even the most liberal orthodox economists, rent control is frequently used (sometimes in the first chapter) as a prime example of an ill-advised government policy.  After showing that the law will surely create a shortage (quantity demanded will exceed quantity supplied), the analysis turns to the effect on landlords.  The basic idea is that the government's imposition of rent control laws reduces the value of the property, and the landlords have thus been robbed.  (Sometimes, there is an attempt to show that the next step is for landlords to become slumlords, because the property has lost value and supposedly is not worth maintaining.)  Surely, the reader learns, the right policy is to eliminate the rent control law.  Surely.

What happens, however, when the law has been on the books for decades, and the ownership of the affected properties has changed hands a number of times?  Each time a rent-controlled property is sold, its price reflects the current legal reality that the units are subject to rent controls.  (A pricing model based on "rational actors" would take into account the probability that the law will be changed, but we can set that aside for now.)  That means that the repeal of the law would be a pure windfall for the current owners, who bought the properties encumbered by rent control laws, and thus at a steep discount.  Repealing the laws does increase the value of the affected properties, but the people who reap those gains have no claim that they are the rightful beneficiaries of the policy change.

This problem of transitional windfalls and penalties comes up all the time.  For example, in a series of posts a few years ago, in which I decried the policy regime in the U.S. whereby people are encouraged to buy homes (via policies such as the mortgage interest deduction, property tax deductions, government-guaranteed mortgages, subsidies for suburban development, and on and on), the most difficult question that I confronted was what to do about reliance interests.  That is, for people (like me) who bought their homes in the existing policy regime, a change in government policy to discourage individual home ownership is almost certainly going to radically reduce the value of their most valuable asset.

Reliance interests are central to many areas of law, not just in property law but in contract law as well.  Indeed, some scholars have suggested that "expectation damages" are the wrong default in breach of contract cases, and that "reliance damages" should be promoted from the fall-back rule to the default rule.  The idea is simple: If a person did something in reliance on certain facts, and another party changes the facts of the world on which the person relies, then that other party should make the person whole.  Pulling the rug out from under someone is presumptively unacceptable, requiring compensation.

In the case of taxi medallions, both of the comments on my Uber post that addressed this issue presumed that cities are setting the number of medallions below "equilibrium," by which we usually mean that the very existence of a price for the medallions is proof that the government created an artificial shortage.  But the second comment identified the correct inquiry, which is whether the policies that create the property value in the first place are "ill adivised, economically inefficient and not in the public interest."  Although both that commenter and Professor Dorf apparently consider it obvious that this is the case in terms of taxi medallions in NYC, it is important to think through the alternatives.

To go back to the rent control example for a moment, however, the problem with the standard textbook analysis is that it severely limits what it takes into account as the costs and benefits of rent control.  In one of my posts about home ownership, I pointed out that the supposed benefit of "neighborhood stability," which is often invoked by those who favor policies to encourage people to buy houses, is actually tied not to owning versus renting, but instead to how long people live in their neighborhoods.  A reader then commented that rent control laws, even if they are problematic in all of the ways that the textbooks say, at least make it possible for people to stay put, and to know in advance that they will continue to be able to stay put.  This has the advantage of connecting them to their neighborhoods, which further allows them to put time and effort into their local schools, parks, and so on.

Where does this leave us?  If we want to get rid of an existing rent control law, we know that some property owners will receive a windfall, while other property owners (if they still exist) will see their previous loss restored.  In the meantime, we will eliminate housing shortages, but only by forcing some people out of their homes (relocating who knows where), and we will break up the social cohesion of neighborhoods.  In some cases, that is a good thing, if the neighborhoods are dysfunctional, but we have seen how that excuse has been used by urban "visionaries" as an excuse to displace poor people.

Similarly, NYC in particular has completely defensible reasons for limiting taxi traffic in the city.  Unfortunately, although it succeeds in doing that, it is simultaneously prevented by the New York state government from doing other things that would improve life in the city (like limiting all traffic flow in high-density areas).  Whether one thinks that it would be acceptable to change the underlying laws that indirectly support the price of taxi medallions -- and thus whether one thinks that it would be OK to pull the rug out from under current taxi medallion owners -- depends on whether one thinks that the current mix of all related laws is reasonably achieving some desirable set of public policy goals.  If it is, then suddenly opening the taxi market to new entrants is a bad idea.  If one believes that it is not, then presumably one also believes that the extra benefits of deregulation are greater than the losses from pulling the rug out from under some people.

On taxi medallions, I remain unconvinced that there are too few of them, mostly because I do not see why the "free-market equilibrium" number of taxis is immune from the over-grazing problem of public goods (aka, the tragedy of the commons).  But beyond that particular example, I do think that it is important to re-emphasize that transition costs are an inevitable part of any policy change.  Nearly all standard economic approaches to analyzing public policy changes ignore reliance interests, so much so that it becomes necessary to remind oneself to take them into account.  Unless we do that, however, we are ignoring potentially dispositive facts about the balance of costs and benefits of policy changes.