Economist claims Cornell smoking-ban study is flawed

Nation's Restaurant News, Nov 11, 1996 by Milford Prewitt

NEW YORK -- A Cornell University study that reported an increase in restaurant sales after New York City passed an anti-smoking ban was seriously flawed, according to a Northwestern University economist.

Michael K. Evans, head of the Evans Group think tank, was hired by the National Smoker's Alliance -- a smokers' rights advocacy group -- to conduct a thirdparty review of the Cornell research. Evans maintained that Manhattan restaurant operators were hurt by the April 1995 Smoke Free Air Act. The ban prohibits smoking in all restaurants with 35 seats or more.

Evans said that, in applying certain mathematical formulas to the same data the Cornell researchers used, he estimated a 9- to- 16-percent decline in restaurant sales as a result of the law.

The NSA said it had sought out an economist to analyze the survey because the Cornell research has been used subsequently in other cities to advance anti-smoking legislation.

Cornell University officials and the authors of the report defended their study, emphasizing that they would not change a single aspect about their methodology or conclusions if given the chance to do it all again. They insisted that their report is mistakenly being criticized as an economic impact study when,in fact, it is a survey of consumer behavior.

"The most troubling aspect of the Cornell survey is that it does not present a useful before and after picture," Evans wrote in his review, declining requests for interviews. "The Cornell study is seriously incomplete because it does not measure the change in the number of times per week that smokers and nonsmokers dined out before and after the ban or the amount spent before and after."

Responding to Evans' criticisms, Cornell refuted him point by point, even dismissing the assumptions behind a key mathematical formula the economist used to prepare his review. More to the point, the Cornell researchers charge that Evans did not understand the intent of the study and manipulated their work out of context in order to advance the political agenda of the NSA.

"Our study was not flawed or in error," said Cathy Enz, associate professor of management at Cornell's School of Hotel Administration and the principal author of the study. "But if you don't like the results of something, criticize it. It makes my blood boil that a study that was under complete objectivity is coming under this kind of name calling.

"We never said that our report was definitive or final. You do a study, you raise questions and then you do another study. This is just the beginning of our work."

Leo M. Renaghan, director of the Center for Hospitality Research at Cornell, which financed the study, said Enz would present an update on her research Nov. 12 at the Marriott Marquis in Times Square during the New York Restaurant Show. Renaghan said he hoped critics of the study show up to challenge the report.

"We unequivocally stand behind this study," Renaghan said. "What we have is an advocacy group distorting solid research, and I think it's a shame that the restaurant industry-is being held hostage by this political agenda.

What happens to restaurant sales in cities that pass smoking bans is one of the most contentious issues in foodservice. Up until now the pro and con factions in the debate have defended their arguments by compiling anecdotal reactions of customers or gathering immediate sales impressions from operators.

Recently, the NSA, for example, released the results of a national poll that was conducted by Roper Starch Worldwide, reporting that 83 percent of bar and tavern owners anticipated a loss of revenue if a federal smoking ban proposed by the Occupational Safety and Health Administration ever becomes law.

Antagonists on both side of the issues agree that the only objective way to conduct a thorough analysis of the impact of smoking bans on restaurant sales is to gather pre-ban sales and tax data and compare it against postban sales and tax data using the same restaurants that were in business before and after the law.

The current flare-up concerns the findings in "Should NYC's Restaurateurs Lighten Up?" published in the spring 1996 issue of the Cornell Hotel Restaurant Administration Quarterly -- a nonacademic journal whose reader base is primarily hotel and restaurant operators.

Tom Humber, president and chief executive of tee NSA, said proponents of smoke-free environments used the Cornell study in Mesa, Ariz., for example, where one of the country's toughest anti-smoking ordinances won voter approval earlier this year.

More recently, the Cornell research armed anti-smoking advocates in Montgomery County, Md.'s new smoking ban and was used in suburban Buffalo, N.Y., Humber reported.

"You have bad research and bad assumptions playing a big role in the passage of bad legislation," Humber said. "To me, that's political."

Humber joins the National Restaurant Association, the New York State Restaurant Association, the accounting firm Coopers & Lybrand and hospitality consultants, who also have criticized the Cornell research.