Smoking ban research: policy kills

A recent issue of The Economist includes a small blurb on new research published in the Journal of Public Economics by two Wisconsin economists. Scott Adams, of UWM, and Chad Cotti, of UW-Oshkosh, have found that when public smoking bans are implemented at the county level, drunk driving fatalities in adjacent counties can rise. Presumably, this is due to smokers no longer patronizing local establishments and driving to and from bars in the counties that have not prohibited public smoking.

For instance, they found that when Colorado's Boulder County banned smoking, fatal accidents in Jefferson County, between Boulder and Denver, went up by 40%.

Smoking ban opponents will find this research useful in arguing that the ban's benefits may not be outweighed by its other impacts. Smoking ban proponents may use the findings to argue that a statewide ban is the most appropriate, so as to reduce the potential for cross-county drunk-driving by smokers.

However, the economists also looked at statewide bans, finding, for instance, that accidents increased 26% in Pennsylvania's Delaware County after the neighboring state of Delaware introduced a smoking ban in 2002.

However, it's unclear which way this finding cuts for Wisconsin, as our neighbors Minnesota and Illinois both have new smoking bans in effect this year, while Iowa's goes into effect soon. Would a statewide smoking ban endanger our citizens on the road, forcing them to drive to Michigan's Upper Peninsula? Or, without a statewide ban are we (at least in our southern and western border counties) about to see an influx of intoxicated drivers from the Twin Cities and Chicagoland?