An NBER working paper by Jonathan Guryan, Melissa Schettini Kearney (2009) gives strong evidence that gambling is addictive using a creative identification technique:
“We use the sale of a winning ticket in the zip code, the location of which is random conditional on sales, as an instrument for present consumption and test for a causal relationship between present and future consumption…Our data from the Texas State Lottery suggests that after 6 months, roughly half of the initial increase in lottery consumption is maintained. After 18 months, roughly 40 percent of the initial shock persists, though estimates become less precise. These estimates provide an upper bound on the degree of addictiveness in lottery gambling. They also highlight the potential effectiveness of innovations and advertising campaigns designed to increase lottery gambling.”
The authors wisely note that the welfare implications of their findings are ambiguous. Is the increase in lottery sales due to increased addiction or learning-by-doing (i.e., it’s fun to play the lottery and I do it in moderation)? If individuals become “…rational addicts in a Becker-Murphy sense, optimal provision and pricing would depend only on the external harm imposed, for example, on family members from the displaced consumption of other household goods.”
Further, government sponsored gambling is a way to raise money for public expenditures. One must compare the deadweight loss from standard revenue raising measures (e.g., sales and income taxes) compared to lotteries. I presume that the deadweight loss from raising revenue via the lottery is smaller than sales and income taxes in the absence of addictive gambling, but may be larger in the case where compulsive gambling becomes a widespread problem.
- Jonathan Guryan, Melissa Schettini Kearney (2009) “Is Lottery Gambling Addictive?” NBER WP # 14742.