Are There Constant Expected Rates of Return in Thoroughbred Horse Betting Markets And What Might It Imply About the Operation of Financial Markets?
Abstract
In the area of financial economics, there is a long and rich research history of looking into the behavior of waging/gambling behavior in the thoroughbred horse racing industry.   The intention has been to provide insight into the behavior of investors in more complicated financial markets with an assessment of how people “invest” in simpler settings such as a horse race, in which various “investment instruments” (horses) with uncertain returns are available to the “investor.”
In this paper, I attempt to answer two related questions.  The first relates to what Sauer, in his 1998 JEL survey on wagering markets, refers to as the “constant expected returns model.”  Is there data evidence supporting this model?  The second relates to the existence and structural consistency of a preference function associated with risk taking and expected returns.  Is there evidence of the so-called “long-shot” bias?
This paper attempts to address these two questions using a rich database with race results and historical-performance data.
The study is broken into two parts.  The first involves an assessment and evaluation of alternative specifications of the objective probability model.  The second involves the testing of two separate hypotheses: the constant-expected-returns hypothesis and long-shot bias hypothesis.
JEL classification: C25; D81; G14; L83
David A Poyer, Ph.D.
Associate Professor of Economics
Morehouse College
830 Westview Dr. SW
Atlanta, GA 30314
Phone: 404 681 2800, ext. 2553
E-mail: dpoyer@morehouse.edu
Submitted to the International Atlantic Economic Society to be considered for presentation at the
THE 68^th INTERNATIONAL ATLANTIC ECONOMIC CONFERENCE
Boston Massachusetts USA
October 8-11 2009