Draft Two(a), Section 5

5. Robustly Regulating Casino Gambling

The robustness principle, like Mill’s harm principle, suggests that gambling should be legal. As it would be difficult for people to gamble in a safe manner without gambling providers, “sales” of gambling services also must be legal. Casinos are the standard method of providing such services, though lotteries (the most popular form of gambling) generally operate in a separate system outside of casinos.

Robustness requires that the regulatory regime lead to acceptable outcomes even when many gamblers are addicted or misinformed or irrational with respect to their wagering behavior. These rationality shortfalls are not just theoretical in the case of gambling; rather, they are pervasive. Something on the order of one to two percent of the adult population typically falls within the clinical understanding of disordered gambling – though the percentage of casino patrons who are problem gamblers is much higher. These individuals suffer enormous costs stemming from their gambling fixation, including family breakdown, bankruptcy, and job loss. But even non-problem gamblers often misunderstand probabilities of winning and losing, and selectively forget past losses relative to wins, so that their views of their own gambling outcomes are quite skewed. Gamblers hold unreasonable beliefs about past play influencing future outcomes, such as the notion that a particular slot machine is “hot”: they believe (incorrectly) that the machine temporarily offers better winning prospects.

Countering incorrect beliefs is an important component of any robust gambling policy. Clear and prominent identification of the odds of winning (or perhaps of losing) could be mandated. One of the more startling facts about electronic gambling machines is that players are very poorly informed about the average “price” of a play. Regulations generally require slot machines to pay out some minimum percentage of their takings, such as 80%. A machine that costs $1 per play, and just met such a regulatory standard, therefore would have an average net price of 20 cents per play. But the machine might be set for an average return of 98% of the wagers, leaving an average net price of 2 cents per play. Gamblers typically are not informed the average net price of a play, however, even though it can vary by an order of magnitude across different machines (or on the same machine over time). Imagine going to the store and buying a loaf of bread, and not knowing until you bought it whether it cost $1 or $10. And in the case of the slot machines, unless you play for many, many hours, you can’t be confident of the average net price even after you make your purchases. A mandate to provide clear pricing information would seem to be an all but necessary element of a robust casino regulatory regime.

Price information might be provided in easy to understand ways, including the “average net price per play,” as described above. In addition, for instance, the expected loss per $100 wagered could be given. (Today’s networked, progressive slots imply that the expectation changes over time, but second-by-second updating is easy to electronically compute and display. The potential to play multiple lines on a slot machine also muddies the picture, but again, not in such a way that prices of different gambler options can’t be determined and displayed.) Other information that could continuously be on offer might be the amount of money wagered in a given session, the current net loss position, and the amount of time spent gambling. (Problem gamblers not only suffer from money woes, they also end up spending very significant amounts of time gambling, to the detriment of other dimensions of their lives.) The electronic cards that most regular gamblers use as part of casino loyalty programs also can be used to gather and disseminate this information.

Self-limit and self-exclusion programs should also be part of a robust gambling regulatory regime. A self-limit program allows gamblers to pre-commit as to how much money they are willing to lose prior to a gambling session. (The relevant timeframe could be longer, too, such as limiting losses per month or per year, or limits could be set across multiple timeframes.) The limit is irrevocable, so that in-the-moment passions cannot override judgments made earlier and (hopefully) in a more considered fashion. It would also be compatible with a robust regime to require every patron to choose a binding loss limit upon entering a casino – a policy Australia has considered implementing – as such a policy can do much to limit the harms of gambling, while imposing very little on rational bettors, who, like everyone, would be allowed to choose a very high limit if that is their preference.

Self-exclusion programs are an extreme form of self-limit, in that the limit is set to zero. Generally gamblers who choose to self-exclude are not allowed admission to a casino, and can even be charged with trespassing should they attempt to breach their agreement. Administration is less than perfect, but enforcement can be aided through technological means such as facial recognition software, or by requiring that all casino patrons present identification upon entering. (The imposition of an ID requirement on non-excluded gamblers is small enough that it would not violate robustness, and in any event, is often necessary to enforce age restrictions.) Many jurisdictions offer options concerning the duration of a self-exclusion agreement, some as brief as 6 months or a year, with longer durations including bans of 5 years or even for a lifetime.

A self-exclusion program allows gamblers to opt out, while offering more commitment to that choice than comes from most declarations of abstinence. (Individuals might be able to sign private commitment contracts that also raise the stakes of breaching a no-gambling pledge, and hence potentially provide more motivation to abstain, too; see stickk.com.) Further, the self-exclusion system operates in a fashion that is all but invisible to those satisfied gamblers who choose not to opt out. A more forceful version of an exclusion regime is to alter the default setting so that exclusion is the starting point: adults who want to gamble must opt-in, must take some positive steps to acquire a “license” to gamble. Those positive steps might be quite minor, such as a few days delay between the moment that a license (or “membership” in a casino, perhaps) is applied for and when the license is granted – this was a longtime British casino regulation – or something more involved, including passing an exam indicating that the applicant possesses a basic understanding of probabilities and the likelihood of monetary loss from casino gambling. People who choose not to take the requisite positive steps, then, people who do nothing, are thereby self-excluded.