Dictator game
From Wikipedia, the free encyclopedia
The dictator game is a very simple game in experimental economics, similar to the ultimatum game.
The first player "the proposer" determines an allocation (split) of some endowment (such as a cash prize). The "responder" in this case simply receives the remainder of the endowment not allocated by the proposer to himself. The responder's role is entirely passive (he has no strategic input into the outcome of the game).
This game has been used to test the homo economicus model of individual behavior: If individuals were only concerned with their own economic well being, proposers would allocate the entire good to themselves and give nothing to the responder. However, Henrich et al (2004) discovered in a wide cross cultural study that proposers do allocate a non-zero share of the endowment to the responder. (This 2004 study was an extension of earlier developments in the dictator and impunity games).
This result appears to demonstrate that either:
- Proposers fail to maximize their own expected utility, or
- Proposer's utility functions include benefits received by others.
However, other explanations have been offered, such as the anonymity hypothesis, which claim that the experiment is not correctly designed to test for "altruistic" behaviour, and that the presence of the experimenter causes the proposer to avoid the appearance of "greed".
[edit] Nick Bardsley's -and John List's- Experiments
University of Chicago economics professors John List and Steven Levitt discussed the Dictator Game, and Levitt said something like "It's great that people in the test want to give 30%-40% of their money away, but how come I've never once been on the subway and had someone come up to me and say 'I just got $10! Have some of it!'"
So List designed a new experiment - one where the "dictator" can either give $0-$10 away, or take a dollar from the other person. The results: Many, many people did nothing, a few still gave some money and a few took the dollar. Basically, the bell curve shifted to have most people do nothing.
This suggests that most people don't want to give money away, but they also want to seem like nice people because people, the scientists, are watching them. When the option of not stealing anything becomes "a nice thing to do" (whereas before it was the most selfish) most people would rather have more money.
Another experiment had people choosing between giving $0-$10 away and taking $0-$10 from the other person. The results were that people usually took around $5, which made them not seem terrible, at least in their minds, but still got them a lot of money.
The last experiment he told us about was where people first had to stuff envelopes for an hour and then they got $10. After that, they could choose whether they wanted to give any money away or take some from another envelope stuffer. Most people choose to keep their own money and not take anyone else's. This result satisfied Levitt because it does mirror reality - most people don't randomly give away money if they feel like they earned it, and they won't take other people's money if they think the other people earned theirs.
However, the result that dictators will take given the opportunity to do so had already been discovered by experimentalist Nick Bardsley in the UK, when he was working at the University of Nottingham. The experiments were presented at the June 2005 meeting of the Economic Science Association in Montreal.
Bardsley had previously run three independent experiments along essentially the same lines as the above, with similar results. In one experiment, subjects in one treatment had to choose between doing nothing and giving up to £3 to another subject. Whilst subjects in a second treatment chose between doing nothing and taking up to the same amount. A bootstrap (statistics) procedure returned a 95% confidence interval of 34% - 93% for the proportion of givers who would have taken if they were in the other game, with a point estimate of 69%. In other words, it looks like most givers would take in the other game and there's strong evidence that more than 1/3 would switch.
The results can be interpreted as an effect of demand characteristics of the experimental situation, a problem well-known in the methodological literature on experimental social psychology.
[edit] Notes
- ↑ Henrich, Joseph, Robert Boyd, Samuel Bowles, Colin Camerer, Ernst Fehr, and Herbert Gintis (2004) Foundations of Human Sociality: Economic Experiments and Ethnographic Evidence from Fifteen Small-Scale Societies. Oxford University Press.
- ↑ For example, Bolton, Katok, Zwick 1998, Dictator game giving: Rules of fairness versus acts of kindness in International Journal of Game Theory, Volume 27, Number 2 (Article Abstract). This paper includes a review of dictator games going back to 1994 (Forsythe R, Horowitz JL, Savin NE, Sefton M, 1994 Fairness in simple bargaining experiments. in Games and Economic Behavior).
- ↑ For example, the model of "inequity aversion" proposed by Ernst Fehr of the Henrich et al. study above: This model assigns personal disutility to an uneven split for equal work, sometimes called "guilt".
- ↑ See Bolton et al. page 270.
- ↑ Demand characteristics are discussed in Bardsley (2005), Experimental Economics and the Artificiality of Alteration in Journal of Economic Methodology, Vol 12, Number 2. Open Access
[edit] Related research
Haley, K. and Fessler, D. (2005). Nobody’s watching? Subtle cues affect generosity in an anonymous economic game, Evolution and Human Behaviour, 26. 245-256.
(The research above concludes that people tend to be more generous if there is a picture of a pair of eyes watching them)
