Monday, June 29, 2015

Barberis (2013) on Tail Events

Nicholas Barberis, “The Psychology of Tail Events: Progress and Challenges.” American Economic Review 103(3): 611-16, 2013.

• “Tail events” are low probability, but high impact outcomes, like a huge stock market crash. 

• Analysis of human reactions to tail events focuses on the perceived probability of the event, and, with that probability given, how the outcome itself is judged. 

• People tend to overestimate the probability of rare, but monumental events, whether the outcomes are superb (like winning the lottery) or disastrous (like being victimized by terrorists). 

• Even when people assess probabilities objectively, however, they tend to put “excessive” weight on the outcomes tied to tail events. Hence they might be willing to pay to avoid a very unlikely but large loss, at the same time they are willing to pay to buy a very unlikely chance at a windfall: the rare outcome of the big loss is overweighted, as is the rare outcome of the big win. This overweighting is a feature of preferences and (probably?) cannot be said to be a mistake. 

• The excessive weight on low-probability outcomes indicates that positive skewness (a lottery-like low probability of a big win) is valuable to people, so that shares of stocks of individual companies that offer such skewness do not have to have as high of an expected return to attract buyers. 

• But the excessive weight placed on low probability events also implies that negative skewness (a small probability of a large loss) is aversive, so that assets that are negatively skewed (such as the overall stock market, which might crash) require a premium in terms of expected returns to attract buyers; this approach offers an explanation for the equity premium puzzle.

No comments:

Post a Comment