Through the Broken Window

Smoke and Mirrors: Re: Hugh Waters' Study “The Economic Impact of Secondhand Smoke in Maryland”

By Christopher Goff

Hugh Water’s study on the economic impact of secondhand smoke in Maryland is a gross exaggeration of the real costs associated with environmental tobacco smoke (ETS). Not only is his rate of population attributable risk (PAR) artificially high, but he includes all illnesses that show any correlation with ETS whatsoever – never mind the fact there is no evidence that these illnesses are actually caused by secondhand smoke. Additionally, he includes burn hospitalizations, outpatient services, and deaths, none of which could possibly be attributed to secondhand smoke. Further, the study never takes into account the length of exposure, concentration, or dispersion of the purportedly lethal effluence, thereby grossly elevating the reported “costs”. Lastly, the overt disregard for the costs to Maryland that would result from a ban on public smoking is academically criminal. As Robert Levy of the Cato Institute has pointed out in regard to this sort of statistical legerdemain, “responsible statisticians agree that it is impossible to attribute causation to a single variable, like tobacco, when there are multiple causal factors that are correlated with one another”. This study stands in blatant disregard for the reality of the issue and should therefore, itself be disregarded.

The public health argument that Waters attempts to bolster is itself a flawed hypothesis that gained unmerited credence. In 1993, the Environmental Protection Agency (EPA) released the purported “grand-daddy” of all smoking studies. It declared that ETS was a carcinogen that causes 3,000 deaths annually. Five years later the U.S. District Judge William Osteen threw out the study, criticizing the EPA for “cherry picking” the data and demonstrating “no association between ETS and cancer”. He further pointed out that the EPA withheld “significant portions of its findings and reasoning striving to confirm its a priori hypothesis”.

In 1996, American Heart Association’s journal, Circulation, reported no increase in coronary heart disease associated with secondhand smoke “at work or in other settings”. Just two years later, the World Health Organization reported “no association between childhood exposure to environmental tobacco smoke (ETS) and lung cancer”.

Additionally, a 1999 editorial in the New England Journal of Medicine concluded that, “we still do not know, with accuracy how much or even whether [ETS] increases the risk of coronary heart disease”. Furthermore, Robert A. Levy points out that, the American Council on Science and Health has stated that the results, as mentioned above, are “consistent” with studies by the Center for Disease Control and Prevention.

In May 2003, the researchers at the British Medical Journal found that environmental tobacco smoke or ‘secondhand smoke’ had “no significant connection with heart disease or lung cancer death at any level of exposure at any time”. Lastly, MIT professor of economics Jonathan Gruber, stated that “the size of the health costs of secondhand some are quite ambiguous and controversial”. He further stated that, “[W]hile controversy exists within the literature, there is a fairly strong consensus that the externalities that have been measured are small on net” and that “issues such as secondhand smoke may justify public policies such as clear air laws that restrict smoking in public places, but the limited evidence on the impacts of secondhand smoke also raises questions about the widespread nature of the restrictions”. In other words, the proposed ‘all-out ban’ lacks significant scientific, statistical, or economic proof as a good public policy and should not be pursued as it simply infringes on the rights of consumers, business owners, and workers to no practical purpose.

As a matter of the alleged social costs to the state of Maryland, in 1998, Jane Gravelle of the Congressional Research Service stated that, “smokers do not appear to currently impose net financial costs on the rest of society”. In fact, according to Gravelle, “smoking has apparently brought financial gain to both the federal and state governments”. Also, W. Kip Viscusi, Professor of Law and Economics at Harvard Law School, asserts that smoking either “pays for itself” or in fact, “generates revenues for the states”. He also claimed that “[A]ll states now earn a net financial profit from cigarettes rather than incurring a loss”.

So without significant scientific evidence for substantial costs incurred by secondhand smoke or “passive smoking”, how is it the Hugh Waters’ paper seems to lend so much credence to the proposed Maryland ban? It is simple: he only looks at the alleged savings (benefits) to Maryland citizens and not at any of the costs incurred by individuals and businesses. It is very easy to come out on top, when you neglect to mention costs in a cost-benefit analysis or a cost effectiveness ratio. And even with this flawed data in hand, many policy wonks are still pursuing this agenda of lowering consumer welfare, limiting labor choices and wage options and increasing business expenses. Any limit to choice reduces consumer welfare and business opportunities. Also, you could expect wages and employment opportunities to decline. Additionally, this will have the unintended consequences of creating new “winners” and “losers”, by forcing smaller businesses who can’t adapt to simply fail. In fact, as Arthur Foulkes pointed out in his July 2003 article in The Freeman, “…all the studies supporting smoking bans are based on aggregated restaurant sales data; they look at the “restaurant industry” in the smoke-free communities. They largely ignore what might be happening under the surface to individual businesses”.

There are also distributional concerns. As Jonathan Gruber points out, “the smoking rates of the lowest income quartile are roughly twice those of the highest quartile” and recognizing that, we have to acknowledge that this will serve as a welfare loss to the lower income groups and would have the same effect as a regressive tax. No amount of paternalistic fervor will make this reality disappear.

Whether or not they recognize the sheer absurdity of the proposition that public bans will impact public health, I implore the legislators of Maryland to recognize that this regulation can only cause more economic harm than benefit. Like U.S. District Judge William Osteen did in 1998, throw out this study and rely on the fact that numerous industries have adopted smoke-free work areas without coercion from above. Do not forget that these business owners are in business to please their customers and maximize worker productivity, and if they believe that banning smoking in their particular venue is profitable, they will choose to do it freely -- without the heavy hand of government.

Christopher Goff is a statistician and holds a masters degree in economics from North Carolina State University.