Unintended consequences

Increase In Drunk Driving Fatalities Followed Ban On Smoking In Bars

ScienceDaily (May 21, 2008) — A ban on cigarette smoking in bars is meant to save lives by reducing patrons’ exposure to secondhand smoke. But it may actually be having an unintended consequence, according to a study done at the University of Wisconsin-Milwaukee (UWM).

By comparing data from a variety of locations around the United States where laws requiring smoke-free bars exist with locations without bans, economists Scott Adams and Chad Cotti found a relative increase in fatalities caused by drunk driving following ban enactment.

The results of their study appear in the June issue of the Journal of Public Economics and have also been reported in the May issue of The Economist.

While the results at first seemed surprising to Adams, a UWM assistant professor of economics, and Cotti, now at the University of South Carolina, literature on consumer behavior suggests an explanation: Smokers are willing to drive longer distances to an establishment that allows smoking.

“Like they would to buy fireworks, lotto tickets or, in some cases, alcohol, people will often go to a neighboring jurisdiction that doesn’t have a ban,” says Adams. The number of smokers willing to drive extra distances offsets any reduction in driving from smokers choosing to stay home following a ban, he adds.

Using fatalities as a gauge in the study is more accurate than using data on DUIs, since drunk-driving laws are not uniformly enforced, he says.

The study’s evidence suggests that consumers are driving longer distances to smoke and drink, but this does not exclude other potential explanations.

“We can’t rule out the explanation that smoking bans might reduce the propensity to drink in moderation,” Adams says, “But in each and every instance of ‘border shopping’ we found, the increase in fatalities was true.”

The study is the second on the topic of smoking bans for the pair of economists. The first study focused on whether smoking bans have an effect on bar employment. Results from that study showed that restaurants were helped by smoking bans, especially in warmer climates and in warmer months. But bar employment fell.

They then decided to examine the effect of the bans on drunk driving, says Adams. “The thinking was that the bans might have additional health benefits if there’s a reduction in driving associated with it.”

It still appears that the positive health effects of smoking bans outweigh the negative, he says, but the real conclusion is that a universal smoking ban would eliminate the danger of people trying to avoid the individual bans.

Twenty states have universal bans, and Gov. Jim Doyle has proposed one for Wisconsin.

Adams and Cotti have begun a new study to see if smoke-free bars are associated with evidence of a reduction in heart disease. Preliminary evidence strongly suggests it does.

“I view economics very much as a social science – the costs associated with people’s behavior,” says Adams, who specializes in health and labor economics. “Public economists are concerned with the externalities and whether what affects you also has an impact on others – without those costs being accounted for.”