Nevada and the Win-Win Game, Compulsive Gamblers and Alcohol
By William N. Thompson and R. Keith Schwer

The casino industry has this great advantage over every other industry: Casinos offer a win-win game to customers, providing the best word-of-mouth advertising ever. Quite simply, "winners talk and losers walk." Tales of wins spread like wildfire as they leave the lips of lucky (or occasionally lucky) players, while stories of losses are muted and hidden among more vocal expressions of satisfaction with great buffets, shows, room bargains, near misses and - for Las Vegas casinos - great weather.

Other industries suffer from the fact that complaints are more noticeable. For every bad customer story coming out of a department store, auto dealership, etc., 10 good stories are needed to repair the damage.

The Win-Win Game
Casinos offer a win-win game, but they can still lose. The game is lost if players feel they were cheated or that the games were dishonest, or if the "losers" encounter bad service or if people believe a casino has exploited its players. Government regulators and internal security personnel work hard to keep all the games honest; casino managers who have any sense at all work overtime to maintain top levels of customer service.

The one remaining area where you can lose the win-win game is exploitation. It is here our casinos and gaming venues need to step up and do a better job, particularly when it comes to exploitation tied to pathological or problem gambling. Stories about troubled gamblers losing all their wealth spread quickly by word-of-mouth and sensationalized media accounts. As these stories of loss grow, they are embellished with images of embezzlement from work places and charities, families broken apart and under distress, and, of course, suicide. It is of vital interest to casinos that these stories are not only contained, but that they do not exist in the first place. Problem gambling and its effects must be reduced as much as possible.

Problem Gambling Programs Emerge in Nevada
Before 2003, Nevada had allocated exactly zero dollars for problem gambling programs. Then lawmakers had an epiphany, and $250,000 (one-thirtieth of one percent of the money the state gains from casino taxes) was earmarked for the cause. All of the money went toward a research project that determined the prevalence of pathological and problem gambling among the Nevada population. This was yet another study added to the 100 or so already published revealing that .6, .9, 1.3 or 2.2 percent of adults were current or former pathological problem gamblers. Nevadans were found to have more than twice the prevalence rates found elsewhere. It was good research, to be sure - peer reviewed, academically sound - but that was all it was. While the researchers were "the best" in the field, we were left asking, "So what?" So what should we do about problem gambling? So what should we do to reduce the problem?

After the study was published, the industry could justifiably say, "It appears that problem gamblers are a very small portion of those who enjoy the gambling opportunities we provide." On the other hand, industry critics could point out that the small percentage still adds up to a lot of people among the general population. The legislative response was a mandate that casinos provide brochures and put up signs warning players that gambling can lead to problems for some, and touting the availability of a 1-800 helpline. No real action was taken to reduce the numbers of problem gamblers or the impact of their maladies.

Nevada's "Compulsive" Gamblers
Our research led us in another direction. We wanted to know what the dollar costs of problem gambling represented to other people (non-gamblers) and to the general economy. We went to pathological gamblers for answers, or at least to those who would define their gambling as "compulsive." In 2002, we surveyed 99 members of Gamblers Anonymous (GA) groups in southern Nevada. Of course, these problem gamblers likely encountered - or caused - more severe consequences than the average problem gambler on the street. Other researchers have found that the "on-the-street" compulsive gambler (measured by the South Oaks Gambling Screen, a tool widely accepted in the research community) exhibited costs that were 51 percent of those exhibited by typical GA members.[1] Hence, one seeking to project costs to a full population in an area could discount our numbers as such.

Our survey population of 99 GA members in southern Nevada consisted mostly of Caucasians (80.6 percent), and was evenly divided on gender (48.4 percent women, 51.6 percent men). The mean age was 47.3 years. It was closely divided among those currently married (30.1 percent), those single or widowed (31.5 percent), and those divorced or separated (31.5 percent). Of the latter category, 65.5 percent indicated that the break-up was caused by gambling problems. Respondents were well educated (31.6 percent had a college degree) and had reasonably high incomes, with the mean being $54,495 per year.

The average respondent began gambling at 26.8 years old, gambled weekly (or more) at age 31.8, first borrowed to gamble at age 33.4 and said problems started at age 34.1. The average respondent had been in GA for 2.3 years. Losses, on average, were $112,400. Before joining GA, the average respondent's debts caused by gambling were $60,714.; 45.4 percent incurred personal bankruptcies and 15.1 percent had been sued over debts. A majority (63.3 percent) had stolen property because of gambling, with the average thefts (spread over all the respondents) amounting to $13,517. Nine had been arrested, with the average respondent serving .2 months in jail or prison and .1 months on probation.

The average gambler lost 8.7 hours of work time per month, while 22.9 percent quit work, losing 4.2 months of employment due to gambling; 24.2 percent were fired, losing 2.4 months of labor. Only 3.4 percent accepted welfare because of gambling, whereas 5.8 percent accepted food stamps; 14.9 percent were hospitalized because of health problems related to gambling, while 23.0 percent had outpatient treatments. Nearly two thirds (65.9 percent) planned suicide, while 27.7 percent attempted suicide.

TABLE 1 Where Problem Gamblers Play (Twice Weekly or More)
	Local Casino	82.6%
	Major Resort	50.7%
	Bars	44.4%
	Convenience stores	41.4%
	Grocery Stores	40.0%
	Other	9.2%
TABLE 2 Types of Games Presenting Problems
	Video Poker	70.5%
	Other Machines	49.4%
	Table Games	38.7%
	Sports Gambling	23.0%
	Other	15.6%

Tables 1 and 2 indicate gambling venues and games presenting the most problems for the respondents. The Las Vegas neighborhood casino was the most attractive venue, but it must be noted that all the gamblers were Las Vegas residents, not tourists. Video Poker and other machines dominated their play. Table 3 indicates where players obtained funds when their own funds were inadequate for their gambling activity.

Table 3 (Non-Self) Sources of Gambling Funds
	Passed Bad Checks	63.3%
	Sold Personal Property	60.0%
	Spouse	57.7%
	Cashed in Securities	55.5%
	Used Casino Credit	34.4%
	Children	30.0%
	Used Bookies	16.7%

This self-reported information enabled us to formulate a cost profile. While the length of the "problem gambling career" (onset of problem to initial treatment) was more than 10 years in the study, we reasoned the severity of problems had to be much greater in the years immediately preceding joining GA. Other researchers [2] have used three or four years as the length of severe problems; here we use four years. We annualized the costs by dividing the totals by four. We also only considered the debts of those who went through bankruptcy. We used national crime statistics regarding the costs of arrest and incarceration, and used reported incomes to evaluate costs of lost labor due to gambling. From the information presented, we calculated an annual cost for one pathological gambler to be $19,711 (see Table 4).

Table 4 Annual Cost Impact Profile of the Average GA Gambler
	Missing Work	$2,364*
	Quitting Jobs	$1,092*
	Being Fired	$1,581*
	Unemployment Comp	$87G
	Debt/Bankruptcy	$9,493
	Thefts	$3,379
	Civil Suits	$777G*
	Arrests	$95G*
	Trials	$85G*
	Jail Time	$80G*
	Probation	$170G*
	Food Stamps	$50G
	Costs of Welfare	$84G
	Costs of Treatment	$372*
	TOTAL	$19,711

We define the costs as the disamenities borne by persons other than the gambler. However, we acquiesce with critics who indicate that these are not all "deadweight" costs that subtract wealth for the entire economy, but instead many are costs that are merely transferred from one person to another. Therefore, we have marked those deadweight economic costs with an asterisk. They total $6,616, or 33.6 percent of the total. Costs to the government (marked with a "G"), whether transfer costs or deadweight economic costs, total $1,428, or 7.2 percent of the total.

Critics can also look at this data and say these are only numbers. They can say what we say about other numbers, "So what?" But there is meaning.

One level of meaning derived from the numbers is simply that the costs are major. If we have 20,000 compulsive gamblers in southern Nevada, the economy is losing $120 million or more per year and governments are losing over $28 million (or 51 percent of these amounts if the problem gamblers on the street are less severe than those in GA). The numbers justify government appropriations of more than $250,000 if we are to make a serious effort to lessen problem gambling.

Alcohol and Other Addictions
Another level of criticism - often from the industry - is that studies such as ours [3] (and others in at least six states) neglect the fact that compulsive gamblers have other addictions as well, and that these other addictions may contribute to, or account for, the costs identified. Good point. Accordingly, we asked about other addictions. Table 5 shows the responses.

Table 5 Percent of GA Respondents Reporting Additional Addictions
	Alcohol	22%
	Tobacco	16%
	Drug	9%
	Food	28%
	Shopping	9%

We went back to our numbers to see if those with other addictions exhibited higher costs. For our analysis of the effects (or associations) of comorbidity, we eliminated some individuals who did not fully report on other addictions. Therefore, for the analysis of associations, we used a base figure of $19,585 in annual costs per pathological gambler. To place Table 5's figures in context, we must note that some respondents exhibited few, or even no, costs on their profiles. This made the average for others much higher.

Only alcohol and drug addiction showed statistically significant relationships regarding costs. [4] The findings suggest a pattern of complementary and substitute comorbidity as it relates to problem gambling. For the alcoholics in the survey, we found that the extra addiction added $14,460 to their cost profiles, holding all other factors constant. However, we also found a contrary result from respondents who indicated that they were addicted to drugs. Their gambling cost profile was reduced an average of $19,156, other factors being held constant. Accordingly, a person with gambling, alcohol and drug addiction would have a cost profile $4,696 below the average ($19,156 minus $14,460).

The results from the analysis of shopping, food and tobacco addiction were not significant. Nonetheless, they are worthy of note. The shopaholics had cost profiles $4,234 below average. Food addictions added $10,887 to the profiles, while tobacco addictions added $456.

We can surmise that alcohol use complements gambling, as the two activities are expected to go together at the same time and in the same place. Casinos tacitly acquiesce to alcohol addiction; some would likely say that many casinos promote the comorbidity, as drinks are often served to gamblers at the gambling site. In Las Vegas, drinks are considered a free amenity for gamblers. Certainly, a person who is in a drunken state would be denied free drinks at a casino, as a person who outwardly appears to be intoxicated might create a liability situation. (Note: Nevada does not have a "dram" law assigning tort liability to those distributing alcoholic beverages when the person drinking does harm to another.) Thus, other than for individuals who have passed the insobriety threshold, gambling and drinking are accepted, if not encouraged.

On the other hand, one is not likely to see a casino acquiescing to drug activity. Indeed, drug activity is neither permitted nor tolerated on casino floors by customers or employees. Drug use also is more expensive than alcohol use, and a person high on drugs would probably be considered disruptive to casinos.

The decrease in costs of gambling among compulsive shoppers has two possible explanations. First, a shopper cannot gamble while shopping. Second, shopping, like drug use, demands financial resources that may divert one from gambling activity. Shopping and drug use are substitutes for gambling, not complementary activities.

Food addiction causes a large increase in the cost profiles, though it is not statistically significant. While the dynamic is quite different, casinos are certainly associated with food. Casinos use food, mostly in the form of low-cost specials or buffets, as specific, advertised attractions to bring players into the gambling atmosphere. As free drinks are given to players in Las Vegas, so too, we find that meals are likely the second or third most prevalent free gift.

Tobacco has traditionally been associated with casinos as well. Like drinking, smoking is something one can do while engaged in gambling. While the cost increase for tobacco addicts was not large, it was expected. The increase was not large, perhaps because a quest for tobacco use may lead players to take breaks in their play in order to "light up" or procure cigarettes.

Our study did not address the impact of gambling activity on the costs of alcohol abuse or other addictions. A separate study of crime and gambling, however, found a significant association between the presence of casinos in Wisconsin and the incidence of arrests for drunken driving in the counties that housed them or were adjacent to them. This was the case even where casinos did not serve alcohol, as the gambling activity seemed to present an incentive to seek out drinking opportunities at nearby bars and taverns. [5]

Other factors were also related to the cost profiles. Income adds to the severity of the economic dimension of pathological gambling, as one might expect. Players with higher household income have slightly, but statistically significant, higher cost profiles. So do players with credit from bookies. Those with credit lines at casinos have especially higher profiles - and significantly so - with added costs of $25,051 per player. Those who steal from work have costs $14,522 higher than average, a significantly higher amount. This would be expected analytically, as the amount stolen is figured into the cost profile. Higher education also adds to costs, but not in a significant manner.

Contrary to expectations, those who played mostly in major casinos exhibited a significant lessening of cost profiles, while the lessening of costs for those playing at neighborhood casinos was not significant. Players in bars or stores had a significant increase in the cost profile, adding $1,971. We might speculate that casinos offer greater human interaction and much more noise and light that may distract one from episodes of binge playing, while disassociated activity more easily accompanies gambling in isolated places, such as bars and grocery stores, where machine play occurs.

These findings from our cost study point to the critical role of policies relating to alcohol use at gambling venues, suggesting that credible programs to address problem gambling call for the re-evaluation of current alcohol practices.

We now know that the co-morbidity of alcohol magnifies the economic dimension of problem gambling. What we may not fully understand is how each element of alcohol availability in gambling venues contributes to addictive behavior. Questions may arise as to the variation in costs with differing availabilities of gambling and alcohol. For example, does the wide distribution of gambling machines in local bars and taverns, compared with a more limited gaming district, significantly impact the incidence and the magnitude of costs? What policies to identify and address alcohol at gambling sites prove most successful?

The suggestions in our research tend to make us support actions to reduce alcohol consumption at casinos or, if alcohol is not currently present, to support a continuation of such prohibitions. Recently a Native American casino in Iowa did precisely that. Conversely, gaming policymakers in the United Kingdom have enacted new rules to permit drinking on casino gaming floors. The policymakers in Iowa seem most likely to embrace notions that will keep the win-win game working for them in the future.

Note: Much of the content of this article was drawn from the authors' presentation, "Economic Costs of Gambling," given by R. Keith Schwer to the Economics Forum at the University of Paris Deux, April 22, 2007.

William N. Thompson is a Professor of Public Administration at the University of Nevada, Las Vegas.

R. Keith Schwer is a Professor of Economics and Director of the Center for Business and Economic Research at the University of Nevada, Las Vegas.

[1] Westphal, James, L. J. Johnson (1998). Estimating the Social Cost of Gambling for Louisiana. Baton Rogue: Louisiana State University Medical Center.

[2] See: Grinols, Earl (2004). Gambling in America: Costs and Benefits. Cambridge: Cambridge University Press, 167-174; Leiseur, Henry (1998). "Costs and Treatment of Pathological Gambling." Annals of the American Academy of Political and Social Science. March, 153-171.

[3] Ibid.

[4] We used several statistical tests; the results reported here are based on scores from IV Tobit analyses. Calculations are available via email from william.thompson@unlv.edu.

[5] Thompson, William N., Ricardo Gazel and Dan Rickman (1996). Casinos and Crime in Wisconsin: What’s the Connection? Mequon, Wisc.: Wisconsin Policy Research Institute.