, Volume 9, Issue 2, pp 121-151

Gambling as a rational addiction

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


This paper applies the Becker-Murphy model of rational addiction to gambling and tests the hypotheses of the model empirically using data on pari-mutuel betting at horse tracks from 1950 through 1987. Gambling demand equations which explicitly account for the fact that gambling is an addictive behavior are derived from the Becker-Murphy theoretical model of rational addictive behavior. The effectiveness of changing the takeout rate, the price variable, on gambling behavior, is examined within the addictive framework. Using instrumental variables techniques, gambling demand equations are estimated, with the results supporting the hypothesis of model of rational addictive behavior. In particular, significant inter-temporal linkages are found in gambling consumption, confirming the assumption that gambling is addictive. Future events are found to have a significant impact on current consumption, implying that individuals are not behaving myopically. The long run price elasticities of demand implied by the estimates obtained for the addictive demand equation for handle per attendee is approximately −0.68, significantly larger (by approximately a third) than those obtained from demand equations estimated under the hypothesis of nonaddictive behavior (with an elasticity of −0.454). This reaffirms that the takeout rate is an effective policy instrument for state governments as they set the price of gambling.

The author wishes to acknowledge the assistance of Michael Grossman and Frank Chaloupka. Comments and insights on early drafts of this paper from Gary Becker, Kevin Murphy, Robert Cherry, Alice Hughey, and William Eadington are appreciated and have improved the final product. Peggy Hendershot (Thoroughbred Racing Commission), and Terri LaFleur (Gaming and Wagering Business Magazine) made the job of collecting data for this study a manageable task. Also, thanks goes to Eugene Martin Christiansen for spending hours explaining terms and giving me direction, in an untiring and supportive manner. The research assistance of Christopher Mobilia and Allan Markowitz is appreciated. The views expressed, along with any errors are those of the author.