Monday, March 25, 2013

Games and Tribes

By Kevin D. Williamson
Saturday, March 23, 2013
 
There’s a joke among psychology researchers: We know almost nothing about human psychology, but we know everything about the psychology of Harvard psychology undergrads — because they’re the ones who serve as the subjects of influential studies. It is likely that the same thing holds true of experimental economics.
 
Economists like to study human behavior through games, two of the most common being the Dictator Game and the Ultimatum Game. Dictator Game is simple: Player A is given a sum of money to divide any way he likes with Player B. Player B has no say in the transaction. (Dictator Game is not formally a game, since outcomes are not determined by any interaction among players.) Ultimatum Game is like Dictator Game, but with an important variable: Player B has veto power. Player A proposes a split, and Player B has the power to accept it, in which case both parties get something, or reject it, in which case neither party gets anything.
 
Ultimatum Game is often used to show that people in the real world are not representatives of the purely utility-maximizing species Homo economicus. If they were, the game would play out like this: Given $100 to divide, Player A would propose a split of $99 for himself and $1 for Player B, and Player B would accept it every time — and, perhaps more important, Player A would know that Player B would accept it every time. Player B is better off with $1 than with nothing, so Player A is confident that he can keep most of the prize for himself in exchange for offering the minimum benefit to Player B. In the real world, it does not work that way. Most people in the B position will reject a split of less than 60/30, and people in the Player A position tend to offer splits around 50/50. When asked why, players remark that very uneven splits just seem “unfair.”
 
At least, certain kinds of players do that — namely, players from developed Western societies. Those with very different cultural backgrounds behave in ways that would surprise most Americans, as the social scientist Duncan J. Watts documents in his very enjoyable book Everything Is Obvious Once You Know the Answer. For example, A players from cultures that have strong traditions of reciprocal gift-giving may make hyper-generous offers, presumably because such an open-handed gesture would oblige the recipient in some meaningful way. Interestingly, B players with similar backgrounds will reject hyper-generous offers more often than Westerners typically would, presumably because they do not wish to take on the unspecified obligation that accepting such a gift would entail. Conversely, players from cultures with very little tradition of cooperation between people who are not related tend to behave more like Homo economicus: Offering and expecting the minimum.
 
The ability to negotiate the two kinds of games in a way that maximizes one’s own interests — parsimonious in the Dictator Game, just generous enough in the Ultimatum Game — is sometimes used as a measure of what researchers call “Machiavellian intelligence,” the ability to navigate social groups profitably.
 
As Watts points out, players in these kinds of games, regardless of their cultural background, have a hard time explaining why they have the particular sets of preferences they have: They have internalized their rules of the game without ever quite being conscious of them.
 
Liberals and conservatives do not agree on much, but we often agree on this: The people in the opposite camp often seem like they come from another country, from another society and culture altogether. Conservatives in New York City or on the campus of Bryn Mawr College may feel like they have been parachuted into an alien and possibly hostile environment, and no doubt liberals feel like that when their magazine editors send them on National Review cruises. It may be the case that Red America and Blue America really are separate societies, at least on some micro-cultural level.
 
Such studies as have been done on the relationship between the Ultimatum Game and political preferences have been inconclusive, which is disappointing. There are a few tantalizing little bits: High levels of Machiavellian intelligence have been associated with “right-wing” economic preferences (which of course are more accurately known as “liberal” economic preferences outside our own perversely up-is-down political discourse), while progressive types have been found to be relatively generous in the Dictator Game but not in the Ultimatum Game, which comports nicely with the conservative hunch that liberals are most generous when the other guy doesn’t really have any say in the matter. (That’s only the case in one study I’ve come across; it would be interesting to see this explored more deeply.)
 
It could be the case that progressives simply have a different threshold for what they consider “unfair” in the Ultimatum Game. (Again, this has not been established; I’m just hypothesizing here.) Even a small difference — perhaps one so small that it has not shown up robustly in studies — might incline one favorably toward more redistributive political schemes. And conservatives, whose beliefs are more firmly aligned with classical economic assumptions (or an oversimplified version of them, a progressive might protest), could very well be more open to less even splits in the Ultimatum Game. That, too, opens up some interesting lines of inquiry: If conservatives are more open to uneven splits in the Ultimatum Game, is that because they think more like Homo economicus — never mind “fairness,” both parties are better off than they were before — or do they find that line of argument appealing because it comports with a preexisting micro-cultural tendency?
 
A more likely hypothesis is this: Liberals elevate the abstraction of “fairness” over the concrete benefits that accrue to both parties under even the most uneven split in the Ultimatum Game because they believe that life is in fact a lot like something somewhere between Dictator Game and Ultimatum Game. The fundamental liberal economic fallacy is that the economy is a kind of pie and that somebody, somewhere, is deciding how to divide it up. Economies do not work that way, of course, which is why the phrase “distribution of wealth” is in fact meaningless in most modern contexts — wealth is the result of a dynamic process of creation and exchange, not something that sits in a bucket waiting to get handed out by The Man. (My guess is that this is a holdover from the liberal land-redistribution project that goes back at least to the Lex Sempronia Agraria, the conflict over which some scholars blame for the destruction of the Roman republic. Unlike modern wealth, land really does exist in fixed amounts in any given political territory, and was in fact handed out by the prince according to his whims. Our liberals: Still bravely fighting the social injustices of 134 b.c.)
 
Ask a liberal to describe the relationship between employer and employee, and you will very often hear something like Dictator Game or Ultimatum Game: The lordly employer simply proposes to share some of his revenue with his serf, who, as in Ultimatum Game, may have the wherewithal to walk away or may face immediate material privation unless he accepts whatever terms are offered him, even if they are as meager as those that would be dictated to him in the Dictator Game. In the real world, things are of course much more complex — there are lots of employers, lots of employees, lots of options, and lots of negotiation — but the case for (e.g.) labor unions is generally presented as a response to an implied Dictator Game. The nice thing about free markets: You may have some Ultimatum Games, but you never have a Dictator Game.
 
It may be that we have highly rational, intellectually consistent, and empirically provable reasons for our political differences. It may be that we are just members of different tribes. But on some very important metrics — material prosperity is not the least of them — some tribes do better than others, and one suspects that there must be a reason for that.

No comments: