Cheating at the End to Avoid Regret

How do people behave when they face a finite series of opportunities to cheat with little or no risk of detection? In 4 experiments and a small meta-analysis, we analyzed over 25,000 cheating opportunities faced by over 2,500 people. The results suggested that the odds of cheating are almost 3 times higher at the end of a series than earlier. Participants could cheat in 1 of 2 ways: They could lie about the outcome of a private coin flip to get a payoff that they would otherwise not receive (Studies 1-3) or they could overbill for their work (Study 4). We manipulated the number of cheating opportunities they expected but held the actual number of opportunities constant. The data showed that the likelihood of cheating and the extent of dishonesty were both greater when people believed that they were facing a last choice. Mediation analyses suggested that anticipatory regret about passing up a chance to enrich oneself drove this cheat-at-the-end effect. We found no support for alternative explanations based on the possibility that multiple cheating opportunities depleted people's self-control, eroded their moral standards, or made them feel that they had earned the right to cheat. The data also suggested that the cheat-at-the-end effect may be limited to relatively short series of cheating opportunities (i.e., n < 20). Our discussion addresses the psychological and behavioral dynamics of repeated ethical choices.