Kindernomics: The Developmental Origins of Other-Regarding Preferences in Children

Abstract

People systematically allow others’ outcomes to affect their decisions. These tendencies, known as other-regarding preferences, are irrational according to traditional models of economics, and yet their existence is increasingly well-documented. This picture, however, is unbalanced. More attention has been devoted to examining positive other-regarding preferences, behaviours which help others, than is the case with negative other-regarding preferences, behaviours which harm them. This thesis aimed to help rectify this imbalance by using economic experiments to study the emergence and development of negative other-regarding preferences, and the motivations which lay behind them, in childhood, in a sample aged from 4-13 years of age. Experiments 1 and 2 focused upon costly punishment in a variant of the ultimatum game. Only children aged 6-7 years and upwards were observed to consistently show negative other-regarding preferences, which generally increased with age in both experiments. Experiment 3 used the moonlighting game to compare children’s positive and negative other-regarding preferences, in the form of their willingness to make reciprocal responses to pro- and anti-social behaviours. Negative reciprocity exceeded positive reciprocity in children of all ages, and the two traits were not observed to be correlated within-subjects. Experiments 4 and 5 examined whether negative other-regarding preferences would undermine cooperation in two mutualistic contexts, the battle of the sex game and the stag hunt, and also in the chicken game. In all contexts, pairs of children failed to achieve cooperative outcomes. The implications of these findings are discussed. There was strong evidence of basic fairness concerns such as disadvantageous inequity aversion and relative comparisons affecting these results, but less evidence of higher fairness concerns or of internalised standards of normative behaviour. Negative other-regarding preferences were ubiquitous throughout pre-adolescence and outstripped more cooperative inclinations in virtually all experimental contexts. Previous work may have over-estimated children’s pro-social tendencies