“Equilibrium play in voluntary ultimatum games: Beneficence cannot be extorted” - Comment

Smith and Wilson (2018) argue that behavior in the ultimatum game may be due to the typical implementation, in which players are not given the opportunity to opt out of the game. Using insights from Smith (1759), they suggest that making play voluntary would increase rates of equilibrium play. They conducted an augmented ultimatum game where the responder first decides whether to participate, and compare their experimental data to stylized facts from the literature, reporting “far higher rates of equilibrium play...than heretofore reported”. However, they do not run standard versions of the ultimatum game as a control. To ensure their interpretation is warranted, we conducted experiments of both their augmented game and an analogous standard ultimatum game. In our data, rates of equilibrium play were not higher in the augmented game. Thus, we find no support for the primary conclusion of Smith and Wilson (2018).