Fixed versus variable reference points in the risk-return relationship

Researchers using prospect theory in explaining Bowman's ¹ risk-return paradox have typically assumed a single fixed reference point, normally the industry median, in defining two decision contexts: gain and loss. Our findings suggest this reference point is elevated above industry median performance, and varies over time and across industries independent of industry performance. Consistent with prospect theory, firms above this reference point were risk averse and those below it were risk seeking. We found no evidence of a second survival reference point for firms in trouble.