Stochastic efficiency analysis with risk aversion bounds: a comment

A recent contribution by Meyer et al. (2009, p. 521) corrected an error of fact by Hardaker et al. (2004b, p. 253) about the comparison between stochastic dominance with respect to a function (SDRF) and stochastic efficiency with respect to a function (SERF). While both methods compare risky prospects for a bounded range of degrees of risk aversion, SERF, unlike SDRF, also demands an assumption that a chosen measure of risk aversion is constant over all levels of outcomes being evaluated. It is argued that it is generally reasonable to make such an assumption, especially when the form of the utility function and the bounds on the degree of risk aversion are carefully chosen. Then SERF has the advantage that it can lead to a smaller efficient set than that identified by SDRF. SERF also has advantages of ease and transparency in use.