Debunking the WISPIRG study: part 1

This will be a multi-part series that I’ve been working on aimed at countering the “no economic damage” study that pro-ban activists at the Wisconsin Public Interest Group (WISPIRG) recently released.

Part of the problem with studies like this one is that they simply look at other studies and compile them into a meta-study. This allows them to cherry pick through the data and include only the studies that confirm what they’re trying to prove. Worse yet, they don’t always do their homework on the studies they include and tend to miss the whole story.

So part 1 will look at the inclusion of the “Lexington, KY” study:

Here is the claim that WISPIRG uses:

“In general, selected key business indicators in Lexington restaurants, bars and hotels have not been affected by the smoke-free law…”

The study is a 15-page look at some of Lexington’s economic indicators performed by Ellen Hahn, a nursing professor. It’s a bit light on the data and has more graphics than analysis. Compare that to the 50-page study by Richard Thalheimer, economist.

Since I have no need to reinvent the wheel here, this is what the Bluegrass Institute, an independent think tank and hardly a friend of tobacco, had to say about it:

The actual effects of smoking bans are even more inconclusive than the science regarding secondhand smoke. Pro-ban advocates claim no negative effect on business and often go as far as claiming an economic benefit.

These allegations are spurious to say the least. Such economic studies purporting to show no effect of an enacted smoking ban have multiple and often fatal flaws in their research. For one thing, smoking bans rarely appear the same in different localities. As no two communities are exactly alike, policymakers must carefully evaluate the economic comparisons of smoking bans between them.

Before-and-after comparisons have attempted to show that business activity does not decline in establishments that prohibit smoking following the enactment of a smoking ban. However, very few of the studies attempting to make such comparisons follow the standard rigor and precision required for this type of research. Usually, the time frames are too short to be measured, few external variables are taken into consideration or economic modeling is not utilized or is simply wrong.

For example, a study by University of Kentucky nursing professor Ellen Hahn attempted to demonstrate an absence of negative effects on business activity after Lexington’s smoking ban took effect in 2004. While the report is widely quoted in the media, it has been soundly discredited by researchers across the state. Dr. Paul Coomes, a leading University of Louisville economist, said the Hahn report “is less an econometric study than a short running narrative surrounding a few charts.”

Hahn’s paper contains no rigorous economic model, uses a very short time span and fails to account for many variables such as longer operating hours for bars. After warning of the dangers of making before-and-after comparisons, Hahn proceeds to do just that, claiming the results are instead conclusive. Thus, her conclusions are anything but incontrovertible.

Conversely, a study by University of Louisville economist Richard Thalheimer does contain a rigorous economic model while also accounting for many variables in play. Thalheimer’s study finds a 9 percent to 13 percent drop in demand for alcohol in bars and restaurants after Lexington’s smoking ban was enacted.

Thalheimer was unable to release specific details about the information he reviewed because it contains propriety sales data. Also, he was unable to account for 100 percent of alcohol sales. However, his study does contain analysis on a majority of alcohol sales in Lexington.

So while Thalheimer’s report showing a significant drop in demand is not perfect, it’s the most rigorous and competent analysis of Lexington’s smoking ban. While his study may have a crack in the windshield, Hahn’s paper is missing the entire front half of the car.

So which makes more sense when looking at the economics of a smoking ban? The Hahn study written by a nursing professor, or the Thalheimer study, written by an economist?

In another article, Hahn’s study is once again cited, this time for its apparent conflict of interest:

A current study by University of Kentucky scholar Ellen Hahn purports to garner research on the effects of smoking bans on employee turnover and training. But Hahn will gather her data from employees of Vice Mayor Scanlon’s 75 Applebee’s restaurants. Can she spell “c-o-n-f-l-i-c-t-o-f-i-n-t-e-r-es-t?”

Hahn has received nearly a half-million dollars from the Robert Wood Johnson Foundation to fund her effort to promote smoking bans in all restaurants just as Scanlon has done at Applebee’s.

So now we have questionable economics, funding from a notorious anti-smoking group and a severe conflict of interest.

But the question that remains to be answered is why the WISPIRG study doesn’t include the Thalheimer study or any of the extra information surrounding the Hahn study.