Under prospect theory a truth-telling mechanism for assessing personal beliefs, the quadratic scoring rule, is exploited to develop a new measurement tool for loss aversion. As the latter is a local measure of behavior at the reference point, exhibited as a concave kink in the utility, and utility can be assumed linear for small scale gains or losses, we can directly measure loss aversion without invoking the elicitation of entire utility functions. Under these assumptions and testable properties of probability weighting functions, an experimenter can identify the degree of loss aversion of a subject with only two quadratic scoring rule questions. The additional benefits of the new tool are inherited from characteristic properties of scoring rules: simplicity of implementation for experiments, easy to understand for experimental subjects, incentive compatibility, and lack of error propagation.
Lecture Hall, IfW
Prof. Horst Zank (Manchester)