In a mobbing game, subjects frequently coordinate on selecting a victim, even for modest gains; higher gains make mobbing more likely; no evidence that fear of becoming the victim explains mobbing; ingroup members are less likely to be victims

How To Choose Your Victim. Klaus Abbink, Gönül Doğan. Games and Economic Behavior, https://doi.org/10.1016/j.geb.2018.10.006

Abstract: We introduce the experimental mobbing game. Each player in a group has the option to nominate one of the other players or to nominate no one. If the same person is nominated by all other players, he loses his payoff and the mob gains. We conduct three sets of experiments to study the effects of monetary gains, fear of being mobbed, and different types of focality. In the repeated mobbing game, we find that subjects frequently coordinate on selecting a victim, even for modest gains. Higher gains make mobbing more likely. We find no evidence that fear of becoming the victim explains mobbing. Richer and poorer players are equally focal. Pity plays no role in mobbing decisions. Ingroup members – introduced by colours – are less likely to be victims, and both payoff difference and colour difference serve as strong coordination devices. Commonly employed social preference theories do not explain our findings.


---
The labels [M,T, G, P] are also a hidden homage to the inmates Mather, Travers, Greenhill and Pearce, who escaped from a Tasmanian prison camp in a group of eight in 1822, only to get lost in the forest. When food ran out, the four conspired to apply the Custom of the Sea to the others. When no-one else was left, they turned to killing and eating one another, until only Pearce survived. All victims were chosen in decidedly non-random ways. This story is one of the great Australian foundation myths, and it was an inspiration for this study (for a dramatic reconstruction, see Van Diemen’s Land (2009)). We are confident that none of our Northern European subjects made that connection.