Mathematics self-confidence and the "prepayment effect" in riskless choices

We extend the analysis of a riskless choice experiment reported recently by Hochman et al. (2014). Participants select from among sets of standard playing cards valued by a simple formula. In some sessions, participants are given a prepayment associated with some of the cards, which need not be the earnings-maximizing ones. Hochman et al. find that participants choose an earnings-maximizing card less frequently when another card is prepaid. We replicate this
result under the original instructions, but not with instructions which explain the payment process more explicitly. Participants who state they do not consider themselves good at mathematics make earnings-maximizing choices much less frequently overall, but those who express self-confidence in mathematics drive the treatment effect. The results suggest that even when comparisons among choices require only simple quantitative reasoning steps, market designers
and regulators may need to pay close attention to how the terms of offers are expressed, explained, and implemented.