Colour-dependent dishonesty in a game of chance

We report an odd result of a coin-flip experiment which incentivises dishonest behaviour. Participants of two treatments were asked to flip a coin in private, of which one side was WHITE and the other side BLACK, and report the colour shown by the coin. Payoff depended on the reported colour: in one treatment WHITE was the more profitable outcome whereas in the other treatment BLACK was more profitable. Surprisingly, the magnitude of cheating, as reflected by the difference between the frequency of reporting the more profitable colour and its statistical expectation (50%) was not, more or less, the same in both treatments. Rather, significantly more participants cheated when BLACK was the profitable outcome. This result reappeared in two variants of the coin-flip task. We suggest that a sense of entitlement triggered by a WHITE outcome may explain this behaviour.