A Quasi-Experimental Test of the Marginal Trader Hypothesis
Posted: 11 Jan 2011
Date Written: January 10, 2011
The Marginal Trader Hypothesis (Forsythe, Robert, Forrest Nelson, George R. Neumann and Jack Wright. (1992). Anatomy of an Experimental Political Stock Market, American Economic Review 82(5): 1142-61) posits that a small group of well-informed traders keep an asset’s market price equal to its fundamental value. Forsythe et al. base this claim on evidence from the Iowa Electronic Markets. I test the Marginal Trader Hypothesis by comparing a prediction generated by the Iowa Electronic Markets to that of a prediction contest, a decision task that precludes marginal traders. Specifically, participants are asked to predict the outcome of the 2008 U.S. presidential election. I find that the forecasts generated by the prediction market and the prediction contest are equally accurate. This result is inconsistent with the Marginal Trader Hypothesis, but consistent with the Surowiecki Hypothesis, which states that under certain conditions, any decentralized decision-making mechanism will produce an accurate prediction.
Keywords: Marginal Trader Hypothesis, prediction market, electronic market, experiment, election prediction, experimental asset market, prediction contest
JEL Classification: C9, G4, D4
Suggested Citation: Suggested Citation