One of the perks of moving to Ann Arbor last month is the ready availability of public transportation. A few weeks ago, I drove twentysomething miles one-way to work, costing me perhaps eight hours a week in unproductive driving time and $40 a week in gas money. Now, I “commute” about three and a half miles one-way on a bus, requiring about twenty minutes each way (call it four hours a week) and not requiring weekly trips to the gas station or pricey central campus parking passes. Pretty sweet deal, even if the buses are a bit crowded sometimes. Maybe a lot crowded. Okay, I’ve been in mosh pits with more personal space and fewer elbows. Seriously, another dozen undergrads want on?!?
. . . Right. On the other hand, nearly everyone on the bus behaves really well. They take up a reasonable amount of space given the ubiquitous backpacks, book bags, and purses; they keep the conversation volume down to a dull roar; practically all of the physical contact is involuntary and inevitable given the standing-room-only conditions in a moving bus. I’m pretty sure at least eight in ten are wearing deodorant and/or bathed recently. The least decorous behavior I’ve seen in the week since classes started involved a student shouting to let the driver know that more people wanted to get off the bus at the current stop, which seemed pretty reasonable given that the driver was closing the doors and letting off the brakes. In turn, the driver thought he could safely close the doors and move on because the people who wanted to get off the bus voluntarily squeezed back to make space for new passengers instead of assertively pushing their way out the door.
This brings me to the actual point of this post. Social scientists often talk about collective action problems in which individuals’ narrowly-conceived self interest leads them to behave in ways that leave everyone worse off in the end. For example, when a fish population is on the verge of collapse due to overfishing, individual boat owners often try even harder to overfish the collapsing population under the assumption that self-restraint just leaves more fish for the competition. Of course, this depletes fish populations even faster and guarantees full collapse (and the resulting destruction of the local fishing industry). In the context of crowded transportation, passengers fighting over scarce legroom demonstrate a similar problem: when resources get more scarce, grabbing all you can and fighting over the rest can seem the only alternative to “everyone else” crushing your kneecaps and/or pinning your seat into its full upright position*. Economists from the center-right leftward** typically tell collective action stories in roughly the same pessimistic way: poor myopic homo economicus can’t properly evaluate long-term costs, so we need at least some government regulation to help the economy recognize long-term costs in the short term.
This isn’t the whole story, though. First, people don’t always behave in the way that a model of narrow self-interest would predict. Experiments show that human players in a one-off prisoner’s dilemma game choose “cooperate” roughly 40-50% of the time***. If this sounds low, consider that the “rational” strategy in a non-iterated prisoner’s dilemma is to always defect (and thus, to cooperate 0% of the time). The vast majority of airline passengers don’t fight over seating, and my fellow sardines on the bus go so far in being “nice” that they fail to clearly signal their intent. So in the best case, people don’t tend to behave as unpleasantly as a rational-choice model with a narrow interpretation of self-interest would imply****. Second, the anecdote of the bus passengers suggests that sometimes we can get a lack of clarity and a failure to coordinate collective action not from short-term self-interest (amounting to mutual hostility) but from what seems to be at least politeness and possibly misplaced altruism.
It should, but can’t, go without saying that the possible solutions for this problem do not include asserting that people should just be more selfish/self-interested. For every case of someone being overly or unthoughtfully polite and inadvertently creating inconvenience (we could add the example of drivers at a four-way stop on a country road repeatedly waving the other on while neither actually drives through the stop), we can find people being obnoxiously assertive and forcing their fellow airplane passengers to detour for a sudden removal of the problem passengers. The inconvenience - or indeed the disastrous environmental harm, when we get to collective problems involving overfishing or climate change - caused by the kind of narrow self-interest that some game theoretic models describe as “rational” is far worse****. Instead, I would suggest that the missing link in the case of the self-abnegating bus passengers is some kind of deliberation, that is, an effort on the part of individual passengers to consider the impact of their actions on everyone involved instead of defaulting to a polite but counterproductive behavior.
--------------------
* But domestic and international flights serving the U.S. carried 826 million scheduled passengers in 2013. So if we assume that fights over seating typically involve two people and that one in a thousand fights escalates to a flight diversion or otherwise makes the news, the three recently-publicized incidents would translate to roughly one scheduled passenger out of every 137,000 getting into a fight over seating.
** I initially learned about the empirical pattern behind “collective action problems” from libertarians advocating transferable fishing quotas and learned the phrase from Greg Mankiw’s undergraduate micro textbook.
*** Sally (1995) in Rationality and Society. You’ll probably need JSTOR access to get the full text.
**** Of course, more complex game theoretic models will give results favoring relatively altruistic behavior. For example, Bob Axelrod (a longtime political science professor here at Michigan, by the way) pioneered the “iterated prisoner’s dilemma” back in the early 1980s, in which the game is repeated and players therefore have to worry about the so-called “shadow of the future” (i.e. the costs of having a reputation as a defector). Axelrod’s book The Evolution of Cooperation (1984) is excellent; I can also recommend a game theory chapter in the 1989 edition (and later) of Richard Dawkins’ The Selfish Gene for a shorter introduction to these concepts.
