Keith Rankin - February 7, 2005

Most Economics 200 students will know of the "Prisoners' Dilemma" as the standard example of how individual self-interest can lead to the worst possible collective outcome for the affected individuals.

Two prisoners together are guilty of a single crime, and are interrogated separately. (It is not the interogators who know that they are both guilty; only us, the audience of students or whoever.) Each prisoner is offered a small sentence (eg 2 years) if both confess, avoiding the need for a trial. In the event of a trial, while the combined sentence would be 10 years, there would be freedom for one who successfully testified against the other. Each prisoner, seeking freedom for himself and not trusting the other to confess, agrees to testify against the other. Both are found guilty on the testimony of each other, so they get 5 years each.

It's a strange example though, that is not really meant to suggest that collaborative action generally has greater societal benefits than action based on self-interest. First, the subjects of the dilemma are 'baddies' - guilty criminals who collaborated in crime, and whose interest is simply to minimise their prison sentences. Second, the idea of a public interest beyond the collective interest of the collaborating criminals is completely undeveloped, at least in the standard textbook telling of the story.

The moral of the story is that, if the two prisoners only trusted each other, they could have got a combined sentence of just 4 years, much better for them than the 10 they did get. The silent subtext, though, is that the public is 6 man-years years better off as a result of the prisoners being tricked into what for them became a lose-lose outcome. The undeveloped (though hinted at) paradox is that selfish private action can (maybe quite often) provide significant public benefits. And isn't that what we are really asked to believe about capitalism, that private greed collectively leads to economic efficiency, a societal benefit.

By focusing on guilty prisoners, we forego the chance to use much better illustrations of an underlying problem whereby individuals each seeking to 'win' create an outcome which ultimately makes everyone worse off. I call my illustration the "big car dilemma".

Safety isn't the only reason why many city people opt for 4-wheel-drive SUV vehicles that are really only appropriate for off-road use. Both 'bigness' and 'ruggedness' are seen by any (and not just men) as a symbol they would like to be identified with. But we will ignore that motive, and just focus on the "I want to keep my family safe" motive.

People who think that SUVs are safe are generally not thinking that they are less likely to have a crash, although some do say they feel safer because their vision is less impeded by other vehicles. Generally, by safe, they are thinking that, in a crash, it's the other guy who will die. And if just a few people think like that, then they are probably right. Not only is the other guy more likely to die, but their improved road vision comes at a cost to the other guy. SUVs obscure more of the roads.

As an individual choice, a big SUV makes sense. One more big car hardly makes a nation's whole roading system more dangerous. Yet it does make that one more family feel more safe.

But what if half of all vehicle owners make that same individual choice to buy an SUV in order to make their individual families feel safer. Without doubt, the other half of the population are substantially less safe. They are, in effect, being forced to also buy a big car in order to avoid the usually fatal damage that occurs when small strikes big. The logical outcome is that, if it is rational for just one general user to acquire such a vehicle, then it is equally rational for others to do the same. And once enough others have done the same, then all others feel compelled to do the same. "Motorists might have bought bigger vehicles to reduce their risks" (NZ Herald 5 Feb 2005).

So we must now consider if the roads are a safer place if we all drive big SUVs, or if we all choose cars that are no bigger than necessary to meet our transportation needs. Basically, small-on-small is a less lethal crash than big-on-big. So the logical outcome of individuals rationally choosing big SUVs for their own safety is a world in which everyone feels obliged to acquire such a vehicle, and in which all drivers (ie not just the few remaining small car drivers) are at greater risk than they were before the urban SUV craze took off.

Rational individual action can create a dystopic society in which everyone is worse off.

Fortunately, the logical conclusion rarely plays itself out.

Too many people are "socialised" and resist the choice that is rational for one individual, because they implicitly understand the dilemma. They see the societal cost of the proliferation of such vehicles. Some even see that, for each individual, their share of the societal cost could eventually outweigh the individual benefit (a benefit that diminishes the more individuals acquire it).

---------------------------

Incarceration of guilty prisoners has certain societal benefits. Highway carnage has none. The big car dilemma needs to be solved; by public action, and by a socialisation process that teachers 'players' (ie vehicle purchasers and truck drivers) to emphathise with other road users. (In the prisoners' dilemma, that socialisation process of inter-prisoner trust is actually detrimental for society, because the guilty criminals are released from prison much sooner.)

One form of public action will make penalties for reckless driving greater in proportion to the third-party impact of their vehicles. Big car drivers need a bigger disincentive to drive recklessly. They may actually have the opposite incentive, in part because airbags and other new safety features tend to be introduced first into big cars. The more 'bulletproof' drivers personally feel, the less the disincentive to drive recklessly.

(Seatbelts would appear to be largely exempt from this problem. Reckless and careless people don't even use them unless forced to by law. Seatbelts will continue to be the most effective car safety device ever implemented.)

Indeed the roads might even become safer if the driver airbags of large vehicles were replaced with daggers pointed at drivers' seatbelts. Few persons would buy vehicles larger than they need to meet their transport requirements. And those remaining drivers with large vehicles would drive very very carefully. With their seatbelts on.