I am always surprised in Spring how suddenly there are cars parked on the residential streets in my town where just a month ago the streets were empty. These are narrow streets so a row of cars turns it into a one-lane street that supposed to handle two-way traffic.  And that is when we have to solve the problem of who enters the narrowed section first when two cars are coming in opposite directions on the street.

On my street cars are only allowed to park on the North side.  So if I am headed West I have to move to the oncoming traffic side to pass the row of parked cars.  If I do that and the car coming in the opposite direction has to stop for just a second or two, the driver will be understanding (a quick royal wave on the way by helps!) But if she has to wait much longer than that she is not going to be happy.  And indeed the convention on my street would have me stop and wait even if I arrive at the bottleneck first.

But of course, from an efficiency point of view it shouldn’t matter which side the cars are parked on.  Total waiting time is minimized by a first-come first-served convention.  And note that there aren’t even distributional conseqeuences because what goes West must go East eventually.

Still the payoff-irrelevant asymmetry seems to matter.  For example, a driver headed West would never complain if he arrives second and is made to wait.  And because of the strict efficiency gains this is not the same as New York on the left, London on the right. The perceived property right makes all the difference. And even I, who understands the efficiency argument, adhere to the convention.

Of course there is the matter of the gap. If the Westbound driver arrives just moments before the Eastbound driver then in fact he is forced to stop because at the other end he will be bottled in.  There won’t be enough room for the Westbound driver to get through if the Eastbound driver has not stopped with enough of a gap.

And once you notice this you see that in fact the efficient convention is very difficult to maintain, especially when it’s a long row of cars.  The efficient convention requires the Westbound driver to be able to judge the speed of the oncoming car as well as the current gap.   And the reaction time of the Eastbound driver is an unobservable variable that will have to be factored in.

That ambiguity means that there is no scope for agreement on just how much of headstart the Eastbound driver should be afforded.  Especially because if he is forced to back up, he will be annoyed with good reason. So for sure the second best will give some baseline headstart to the Eastbound driver.

Then there’s the moral hazard problem.  You can close the gap faster by speeding up a bit on the approach.  And even if you don’t speed up, any misjudgement of the gap raises the suspicion that you did speed up, bolstering the Eastbound driver’s gripe.  Note that the moral hazard problem is not mitigated by a convention which gives a longer headstart to the Eastbound driver.  No matter what the headstart is, in those cases where the headstart is binding the incentive to speed up is there.

All things considered, the property rights convention, while inefficient from a first-best point of view, may in fact be the efficient one when the informational asymmetry and moral hazard problems are taken into account.