Introduction: economics of betting markets

Click to increase image sizeClick to decrease image size

Notes

¹ For instance, Bruce and Johnson (1992) report average stakes of £22.63 on race horse favourites. Strumpf (2003) reports that average bet size averaged in excess of $1000 in a study of six illegal bookmakers.

² Friedman and Savage (1948) and Markowitz (1952) in their seminal articles provide a critique of the entertainment rationale. For instance, Markowitz noted that if the utility of a gamble is the expected utility of the outcomes plus the utility of playing the game then, for given fair odds, the smaller the amount bet, the higher the expected utility. This implies millionaires should play poker for pennies and no one should purchase more than one lottery ticket.

³ When surveys ask gamblers why they gamble, a majority (approximately 42–70%) cite financial reasons: ‘to make money’ (see, for example, The Wager, 2000).

⁴ Many experimental results conflict with the standard expected utility model. An excellent discussion of some of this experimental evidence can be found in Starmer (2000).

⁵ See Brockett and Garven (1998) and Cain and Peel (2004). For instance, for a 1 unit bet with odds o win probability p the relationship between the moments is given by

skewness, s, is negatively related to expected return but there is no behaviour implied.