Discussion of: "The Loss Aversion/Narrow Framing Approach to Stock Market Pricing and Participation Puzzles" by Nicholas Barberis and Ming Huang

The authors make a nice case for modifying the utility function of the representative investor in the standard model to incorporate two research findings that characterize individual decision making in experimental settings: Loss Aversion and Narrow Framing (LANF). The authors show how this can be done in a parsimonious way so that the investor's optimization problem can be solved in a rational expectations general equilibrium framework. With this modification alone, they are able to reconcile the low risk free rate with the high equity risk premium at moderate levels of risk aversion. In addition they are able to explain limited participation in the stock market. This is an achievement and the authors must be complemented. I have just one observation to make. LANF, possibly an inherited trait, might have been a desirable characteristic in primitive societies. Those who were willing to fight for avoiding a potential loss, however small when taken in a broader context, in equilibrium probably most often ended up avoiding the loss without actually fighting. However, LANF, probably because it makes individuals react instinctively to situations, can lead to potentially inferior decisions in the modern world. In those situations where LANF leads to high social costs, I expect institutions to develop endogenously to train and educate individuals to minimize the undesirable impact of LANF on their decisions. Smith, Dickhaut, MacCabe and Pardo compared risk versus ambiguity in gain and loss situations on subjects. They found that choice under risk/loss generates more use of the calculational part of the brain (neocortical dorsomedial system), i.e., the more developed part of the brain takes over. Ventromedial system (that arose phylogenetically earlier supporting decision making in animals) plays more of a role in risk/gain, ambiguity/gain and ambiguity/loss situations. Their findings suggest that individuals are likely to react instinctively when they encounter unfamiliar situations that they do not understand well, with the less developed part of the brain playing a more important role. With education individuals are likely to process alternatives in a more calculated manner instead of relying on intuition, with the more developed part of the brain taking over. Therefore the degree of LANF exhibited by an individual may not remain constant over time and is likely to decline with the level of education and experience. In order to fully understand the implications of narrow framing and risk aversion on security prices, it is necessary to allow for heterogeneity in investors' preference parameters, with some investors having standard preferences not subject to the influence of LANF.