Eliciting ambiguity with mixing bets

I show how to reveal ambiguity-sensitive preferences over a single natural event. In the proposed elicitation mechanism, agents mix binarized bets on the uncertain event and its complement under varying betting odds. The mechanism identifies the interval of relevant probabilities for maxmin and maxmax preferences. For variational preferences and smooth second-order preferences, the mechanism reveals inner bounds, that are sharp under high stakes. For small stakes, mixing under second-order preferences is dominated by the variance of the second-order distribution. Additionally, the mechanism can distinguish extreme ambiguity aversion as in maxmin preferences and moderate ambiguity aversion as in variational or smooth second-order preferences. An experimental implementation suggests that participants perceive almost as much ambiguity for the stock index and actions of other participants as for the Ellsberg urn, indicating the importance of ambiguity in real-world decision-making