Expert: Problem gambling study flawed

By Liz Benston

Monday, March 31, 2003 | 10:54 a.m.

Another gambling policy expert is weighing in on the debate about how to quantify the costs to society of addicted gamblers.

Bill Eadington, an economics professor at the University of Nevada, Reno and director of its Institute for the Study of Gambling and Commercial Gaming, expects to publish a report in an upcoming academic journal that takes issue with how University of Nevada, Las Vegas public administration professor and gambling expert Bill Thompson has attempted to estimate the social costs of problem gambling.

The report follows a study released last month by Thompson and UNLV economics professor Keith Schwer, director of UNLV's Center for Business and Economic Research, estimating that addicted gamblers in Southern Nevada cost the state from $300 million to $470 million each year and about $8,000 per problem gambler each year.

The study's methodology sparked criticism from casino insiders who also suggested that the data could be used as fodder for anti-gambling crusaders or as an excuse to hike casino casino taxes at a time when Nevada and other states nationwide are facing budget deficits.

Earlier this month, the Nevada Resort Association -- the chief lobbying group for Nevada casinos -- commissioned a rebuttal report by Georgia College & State University Assistant Professor of Economics Douglas Walker, who said the results of Thompson's study were "unreliable because their analysis is seriously flawed."

Eadington and Walker have long been critical of social cost studies that Thompson has conducted for other states grappling with how to measure the effect of legalized gambling in recent years. But the debate has taken on a bigger public role in recent weeks because Thompson -- claiming to be the first to estimate problem gambling costs in Las Vegas -- has turned his lens on the nation's gambling capital.

Eadington said his analysis, which will appear in the summer issue of the Journal of Gambling Studies, originated from a paper presented at a gambling conference in British Columbia more than two years ago and was not prompted by Thompson's recent report on Las Vegas. The journal issue will be dedicated to papers presented at the British Columbia conference concerning problem gambling's social costs.

The issue also will feature a separate report by Walker that criticizes a study Thompson conducted several years ago on gambling addiction's social cost in South Carolina.

Economists have come up with a precise definition of "social costs" for everything from unemployment to air pollution, Eadington said.

The definition excludes the effects of lost productivity at work, theft, bad debts and welfare payments caused by a person's gambling problem -- all factors economists call "transfers of wealth" from one individual or entity to another rather than money lost by society as a whole, he argues in his report.

"This is not to say these are not real costs to the individuals or they're not regrettable to the individuals," he said. "But it's confusing to call them social costs."

If a gambler borrows $100,000 and loses it gambling, there's still no net loss for the economy because the casino has the money and the individual or the lender does not, for example, he said.

If the person is arrested because of the gambling problem and processed through the courts, society has to allocate resources to pay for that, which translates into a "social cost," he added.

Thompson's studies have arbitrarily included other costs that don't fit the standard definition, creating data that is meaningless and at worst, irresponsible, he said.

At least 50 percent of the costs included in Thompson's $300 million to $470 million estimate should be rejected under the economic model, he said.

Similarly, Walker's report said that Thompson has relied on "numerous arbitrary assumptions" to come up with his Nevada estimate.

Excluding those assumptions, Walker reduced Thompson's social cost estimate to $12 million for problem gamblers in Southern Nevada per year.

Problem gambling is a relatively new area of academic study that has emerged, largely through casino contributions, to analyze the effect of legalized gambling. The few researchers who work in the field -- with the exception of Thompson -- have so far been reluctant to come up with their own social cost estimates.

Eadington said he won't take that step until he has done further analysis of the issue. Other factors -- such as suicide -- have social costs that are difficult to parcel out from costs borne by other individuals, for example, he said.

Thompson said both Eadington and Walker have raised "some good issues and some things to sort out."

Instead of criticizing his analysis, both should be developing their own cost estimates, however, he said.

"Whether we're calling it $12 million, $50 million or $300 million, the cost to the economy and the cost to society is significant. Let's go after the compulsive gambling with a real program. And let's not neglect the bottom line -- which is that there's a problem."

Thompson said his study wasn't intended to portray social costs as those derived by economic models.

"We said these are costs to other people, costs to government as well as economic costs."

Thompson said he intends to conduct further research that will separate costs into all three categories.

Social cost studies are controversial because they attempt to assign values to human suffering and also suggest the industry is to blame, observers say.

Such studies are still valuable tools to explore the issue of gambling addiction, said Chris Armentano, a board member of the Institute for Problem Gambling, a Connecticut-based nonprofit that offers education and training to employers and other groups in that state.

"If the government is going to invest money to mitigate problems caused by gambling, they have to be convinced that a problem exists," Armentano said. "The only way then can do that is with sound research."

"We need more studies and somebody -- whether it's the government or somebody with the money -- needs to come up with credible research," he said.

While Nevada is contending with the effect of its gambling-infused culture, other states are moving forward with legalized gambling "without doing any serious evaluation beforehand" of its effect on society, he said.

But publishing misleading information is probably worse than publishing nothing, Eadington said.

"You take a number like that and it somehow takes on a life of its own," he said. "But what does it really mean? This needs more scientific study."

Underlying the issue of social costs and benefits is a moral debate that is even more daunting for governments: that is, whether to authorize activities that are in some cases self-destructive, said Eadington, who has helped craft gambling policies worldwide.

"With gambling, you have a choice of either applying constraints, prohibitions or other significant limitations" that apply to everyone or focusing policy on people "damaged by self-abuse," he said.

"All that subtlety gets lost when you do what Bill Thompson did. Using these numbers really demonizes gaming, which may in some cases deserve it but not to the extent that the figures suggest."

Thompson said he takes issue with how the casino industry has reacted to the study.

"This, to me, smacks a bit to me of the tobacco industry response," he said. "Rather than saying there's a problem and let's work on it -- maybe let's explore what we can do to bring the costs down -- they've hired somebody to say that there's no problem."

Walker was paid by the Nevada Resort Association for the report. Eadington did not receive industry compensation for his analysis, which conducted under the auspices of UNR. Thompson and Schwer compiled their research using their own funds.