Modern Prospect Theory: The Missing Link Between Modern Portfolio Theory and Prospect Theory

Kahneman-Tversky’s (KT) seminal paper on Prospect Theory (PT) found that individuals placed value on changes in wealth. Individuals appeared to display different reactions to gains and losses and, displayed loss aversion, which depended on the distance from a “critical value”. These two findings, and the notion that MPT was missing a “reference point”, appeared to ring the death knell for MPT. They suggested replacing utility curves with a value curve and Expected Utility Theory with PT. However, KT and others also struggled with practical notions of “reference wealth”. We argue that the true reference point, for both MPT and PT, should be the liability of an individual/institution (and not an asset position or the level of the stock market), as all investment decisions are made to meet some liability. The original KT paper touches on this point but does not appear to explore this further. If we assume that investors derive utility from funded status, borne out in survey results of institutional investors, then we have a clearly articulated and legitimate reference point, liabilities, that KT and Markowitz highlighted, but did not nail down. It is possible to derive a version of the value curve from this adjusted utility curve that reflects KT’s notion of a “critical value” – namely, the point of full funding. When we re-do KT’s questions to individuals with an explicit statement of wealth and liabilities, we find that some results may be reversed and that loss aversion and risk aversion may be linked to funded status. This behavior is very clearly exhibited in the actual portfolio choices of institutional investors and in their survey responses. Hence, if a liability-centric view is taken of both utility theory and PT, as opposed to a simple wealth-centric view, one may find the missing link between these seemingly contradictory approaches.