Belief Elicitation When More than Money Matters: Controlling for "Control"

Elicitation mechanisms typically presume only money enters utility functions. However, nonmonetary objectives are confounders. In particular, psychologists argue people favor bets where ability is involved over equivalent random bets—a preference for control. Our new elicitation method mitigates control objectives and determines that under the widely used matching probabilities method, subjects report beliefs 18 percent higher than their true beliefs to increase control. Nonmonetary objectives account for 68 percent of what would normally be measured as overconfidence. We also find that control is only a desire to bet on doing well; betting on doing badly is perceived as a negative.