Engelmann and Strobel (2002) examine the relative importance of efficiency, maximin preferences, and inequality aversion in simple distribution experiments. They compare two fairness theories: Bolton and Ockenfels' "Theory of Equity, Reciprocity and Competition" (ERC) and Fehr and Schmidt's "Theory of Fairness, Competition, and Cooperation" (F&S). While F&S performs better in direct comparisons, this is attributed to its alignment with maximin preferences. The study finds that efficiency and maximin preferences have a stronger influence than inequality aversion. The authors argue that efficiency concerns and maximin preferences are significant in simple distribution experiments. However, these motives may confound interpretations of other games. For example, choices that seem to reflect inequality aversion may actually be driven by efficiency or maximin preferences. The study also shows that F&S outperforms ERC in taxation games, where it predicts the maximin allocation, which is consistent with maximin preferences. In envy games, F&S performs poorly when faced with Pareto-dominant allocations, while ERC performs better but not well. The basic C&R model performs well in these games. The study also finds that efficiency concerns and maximin preferences are important in the rich and poor games. In the rich game, where the decision maker receives the highest payoff, both ERC and F&S predict the efficient allocation, but only a small percentage of subjects choose it. In the poor game, where the decision maker receives the lowest payoff, both ERC and F&S predict the least efficient allocation, but a larger percentage of subjects choose it. This suggests that maximin preferences are important in these games. The study also examines control treatments to isolate the effects of role uncertainty. It finds that role uncertainty does not significantly affect the results, suggesting that the findings are not driven by this factor. The study concludes that efficiency and maximin preferences are important in simple distribution experiments, and that F&S outperforms ERC in these experiments. The results suggest that efficiency concerns and maximin preferences are important in the interpretation of other experiments.Engelmann and Strobel (2002) examine the relative importance of efficiency, maximin preferences, and inequality aversion in simple distribution experiments. They compare two fairness theories: Bolton and Ockenfels' "Theory of Equity, Reciprocity and Competition" (ERC) and Fehr and Schmidt's "Theory of Fairness, Competition, and Cooperation" (F&S). While F&S performs better in direct comparisons, this is attributed to its alignment with maximin preferences. The study finds that efficiency and maximin preferences have a stronger influence than inequality aversion. The authors argue that efficiency concerns and maximin preferences are significant in simple distribution experiments. However, these motives may confound interpretations of other games. For example, choices that seem to reflect inequality aversion may actually be driven by efficiency or maximin preferences. The study also shows that F&S outperforms ERC in taxation games, where it predicts the maximin allocation, which is consistent with maximin preferences. In envy games, F&S performs poorly when faced with Pareto-dominant allocations, while ERC performs better but not well. The basic C&R model performs well in these games. The study also finds that efficiency concerns and maximin preferences are important in the rich and poor games. In the rich game, where the decision maker receives the highest payoff, both ERC and F&S predict the efficient allocation, but only a small percentage of subjects choose it. In the poor game, where the decision maker receives the lowest payoff, both ERC and F&S predict the least efficient allocation, but a larger percentage of subjects choose it. This suggests that maximin preferences are important in these games. The study also examines control treatments to isolate the effects of role uncertainty. It finds that role uncertainty does not significantly affect the results, suggesting that the findings are not driven by this factor. The study concludes that efficiency and maximin preferences are important in simple distribution experiments, and that F&S outperforms ERC in these experiments. The results suggest that efficiency concerns and maximin preferences are important in the interpretation of other experiments.