About the Interview
When you flip a coin, you expect heads and tails to show up with a 50% chance each. But what if all you knew was that heads and tails each have a chance of at least 25%? That's how Scott Condie captures Knightian uncertainty in asset markets. He models investors who act on basis of ambiguous knowledge, with the result that asset prices fail to reflect all private information. This is financial market modeling beyond the efficient market hypothesis - this is new economic thinking.
About Scott Condie
Scott Condie is an assistant professor of economics at Brigham Young University. His research focuses on asset pricing theory, in particular the general equilibrium pricing effects of investors with non-standard preferences. Full profile
Comments
Professor Condie's work is facinating. But If I may respectfully suggest, I believe this theoretical approach needs to be extended to incorporate the concept of engineered uncertainty. By definition, engineered uncertainity is a purposfull attempt on the part of a subset of investors to generate mis-information and market volatility.
Naturally, those who would generate the uncertainity and the market volatility have private "real information" as opposed to noise. Once this private information becomes public, investors discount this information and provide the engineers with a window of liquidity to profitably re-allocate the portfolio, or use derivatives.
Agree we need to learn how to model risk and uncertainty. Agent-based simulation models are crucial because one needs heterogeneous agents to discover any interactive emergent properties. However, engineered uncertainty and private information in asset markets would seem to be more of a game-theoretical problem.
I believe the key points made here are Condie's reference to worst case scenarios, which is confirmed by loss averse behavior ala prospect theory by Kahnemann and Tversky. People seem to act on acceptable worse case scenarios and then choose the highest expected payoff within that subset. This has implications for portfolio theory.
The second point applies to the unique behavior of financial asset markets, where the worst case scenario is to be last one standing in musical chairs when the music stops. This leads to "irrational" herd behavior in those markets.
To veer off on a tangent, I would venture the thought that the policy implications of risk and uncertainty may be most productive in solving the immediate political problems of social insurance entitlement reform. But that's a subject for another day.
Was Daniel Ellsberg's doctoral dissertation (which was finally published in 2001, almost forty years from when he first completed it) cited in Scott Condie's research?
The relevant parts of Ellsberg's work appeared in some earlier articles which were cited. Points that Ellsberg made in the thesis which only appeared in the book have also appeared elsewhere, by others, in the 40 years between writing and publication. Condie cites this work too.
Post new comment