Anxiety, overconfidence, and risk taking

The NY Fed research staff has written a provocative paper called Anxiety, Overconfidence, and Excess Risk-taking. The authors take research work on changing risk preferences through time and apply it to investor behavior. This is used to explain how investors may behave in different risk situations and show a tendency towards over-confidence. The researchers try to model the over-confidence often found with investor behavior. The paper is not easy to understand, but it does expose the reader to some interesting work in psychology and changing behavior to risk. A lot can be explained when we assume the preferences change through time.

For example, the idea of "anxiety" is found in psychological testing. When posed with taking a risky bet today versus a risky bet tomorrow, many will take the risk tomorrow. If someone has to take the same risk today, they will "chicken-out". Risk preferences change based on when the risk will be realized.

People who are willing to commit to a risk for tomorrow will change there minds if given the same opportunity the next day. Many will have inconsistency with their risky betting strategies. The people who make inconsistent bets seem to be aware of their behavior. They have to pre-commit or they change their views on the type of bets taken.

These changing risk preferences create a complex problem for unraveling the behavior of decision-makers. There is solution to these problems with systematic models. It seems that models that force or impose commitments will have a distinct advantage over discretionary behavior when facing a risky environment.