Preexisting Inequality
In every study we’ve read, the stakes are presumed to be roughly identical for each subject—a potential $10 payoff (or $5 fine) shapes incentives similarly for all involved. Since the money in play is relatively insignificant, and since all subjects are typically college students, this assumption is probably a fair one. After all, even $10 is unlikely to mean much more to a poor college student than a wealthy one, whereas $10 to the average citizen of Malawi is like earning an additional three weeks’ salary.
Still, when I’ve tried to start thinking about applying the various findings of the studies we’ve read to legal doctrines, I’ve found myself struck by the problem of inequality. The typical calculus for a driver in David’s example is likely to change based on the driver’s wealth. To the extent that the cooperation research is applied to internal policing as in Kahan’s paper, I think inequality has little significance. But if we are designing legal sanctions from the government, the effect of preexisting inequality may complicate things.
One obvious effect of preexisting inequality on the studies: in the “one-way identification with information” DG, I suspect that, if the information were to suggest significant inequality of resources between dictator and recipient, the division would be affected.
vhettinger
February 12, 2008 @ 3:44 pm
I’ve been bothered by this as well. I would imagine that a person who has a higher marginal utility for each additional dollar would display a unique pattern in both the public goods and the trust game of investing more than the so-called ‘self-interested’ players, but less than the average* ‘cooperating’ or ‘reciprocating’ player (*average meaning those for whom the marginal utility of each additional dollar is lower).
I would also imagine that these people would be less likely to engage in so-called ‘altruistic punishment’, and would be more likely to accept lower offers in the ultimatum game.
The studies’ attempts to account for different strategies are incomplete in so far as they fail to separate out people who are genuinely motivated by the ‘low stakes’ monetary reward from those who view the game in a more abstract sense. (The first reading from last week hand-wavingly mentioned that the trends tended to translate whether the game was for high or low stakes, but this was so incompletely explained that I remain unconvinced.)
A footnote in this week’s reading mentioned the effects of gender (women tended to give more in the dictator game). The footnote didn’t elaborate, so I don’t know what the researchers who made that discovery concluded, but I suspect it might have something to do with gender differences in the inculcation of cultural messages about ‘winning’. I’d be very interested to see a gender breakdown of behavior in the Prisoner’s Dilemma game, especially the variant with the punishment component.