Testing for efficiency in the New Zealand horse racetrack betting market a thesis in partial fulfilment of the requirements for the degree of Master of Applied Economics at Massey University

Abstract

Using a large sample of New Zealand pari-mutuel horse race betting data, this study tests for market efficiency. This involves testing for weak form efficiency and an anomaly known as the favourite longshot bias. Additionally, a test developed by Henery (1985) is used to examine the extent to which bettors discount their losses. Also, two utility' estimations are calculated using the first three moments surrounding the distribution. Each test is performed twice, firstly with the odds at the close of the tote and secondly with the odds quoted 30 minutes before the tote closes. A number of previous studies are reviewed. The data set is discussed along with its limitations. An extensive description of the research methodology' is presented, followed by the results, interpretations and discussion. Many former studies have found that racetrack betting is not weak form efficient, but instead there exists a negative risk-return trade-off in the market. This study, exhibiting the negative risk-return trade-off and the favourite longshot bias, is consistent with previous studies. Using opening odds, there is much evidence to suggest that the favourite longshot bias exists 30 minutes before the tote closes but is essentially eliminated in the final 30 minutes of betting. The estimation of Henery's test is consistent with his results that bettors discount approximately 2% of their losses as 'not typical'. The implications of this are also discussed. The estimation of bettor's utility' functions shows that bettors are risk lovers and, contrary' to one study, the inclusion of the third moment is insufficient to prove bettors are in fact risk averse.