Wonderful research assistance was provided by Jessica San-San Tien, Dov and Amanda Rosenberg, and Hongjai Rhee. Excellent secretarial work was done by Gail Nash. Helpful comments were received from the MacArthur Foundation Preferences Group, the Russell Sage Foundation Behavioral Economics Summer Camp (Berkeley, 9/96), and discussions with Teck Ho, Hongjai Rhee, Charles Plott and Simon Wilkie. Preparation for this research was provided by Rich Palmer and Burt Camerer (who taught me to ask "why is that?"). Published as Camerer, Colin F. (1998) Can Asset Markets Be Manipulated? A Field Experiment With Racetrack Betting. Journal of Political Economy, 106 (3). pp. 457-482.

To test whether naturally-occurring markets can be strategically manipulated, $500 bets were made at a large racetrack, then cancelled. The net effects of these costless bets gives clues about whether market participants react to information potentially contained in large bets. While the bets moved odds on "attack" horses visibly (compared to matched-pair control horses with similar pre-bet odds), the net effect on betting was close to zero. A second study with $1000 bets at a smaller track replicated the result. These markets could not be successfully manipulated, indicating that bettors did not mistakenly infer information from the experimental bets.

Submitted - sswp983.pdf