An Example of the Butterfield Paradox?

OpinionJournal.com editor James Taranto coined the eponymous term "The Butterfield Effect" after New York Times crime reporter Fox Butterfield, who could not understand why the number of inmates in federal and state prisons was increasing when crime rates were falling. Taranto concluded that perhaps Butterfield inadvertantly reversed the cause with the effect, viz., the correct way to interpret the phenomenom in question is that incarcerating more criminals (thus increasing the Sing-Sing population) reduces the crime rate by taking rapscallions out of circulation.

Now comes an interesting story from Canada that may represent another example of The Butterfield Effect:

Tobacco Marketers Targeting Teens Near Schools

A research team from the University of Alberta chose 81 randomly selected high schools across Canada and surveyed 22,000 high school sophomores and juniors about their smoking habits, while investigating the in-store cigarette advertising of 400 retail stores located near these 81 schools. What did they find?

"At the time of the study, we found that, compared to retail stores near schools with low smoking prevalence, stores near schools with high smoking prevalence had significantly lower prices per cigarette, more in-store promotions and fewer government-sponsored health warnings," said University of Alberta researcher and study co-author Candace Nykiforuk.

The authors conclude that retail shops who display tobacco-related promotional items, called "point-of-purchase" marketing, increase the incidence of cigarette smoking by high school students.

Schools with a smoking prevalence greater than 20.6 per cent had more neighbourhood stores with in-store tobacco promotions and access to lower prices on cigarettes. "We also observed that schools with a lower smoking prevalence had more stores in the neighbourhood that posted government health warning signs about smoking," said Nykiforuk.

The researchers say the strength of this study is in highlighting the relationship between increased PoP [point-of-purchase] activities in retail stores in the school neighbourhood and school smoking prevalence.

Here's an exact quote from the abstract, published in the Canadian Journal of Public Health:

Schools with low smoking prevalence had more stores that posted government health warning signs and higher cigarette prices.

Hey, I'm just a rank amateur, but if I was a marketing executive for a cigarette company and wanted to increase sales I would find retail stores near schools with low smoking incidence and then 1.) bribe the owner to remove the anti-smoking signs, 2.) saturate the store in point-of-purchase marketing items and 3.) lower the price of cigarettes there. Once sales to young smokers climbed I would then raise prices and reduce my budget for PoP promotional items, thus making more dough for Big Tobacco.

Could the researchers have misinterpreted their data? Isn't it possible that stores located near high schools containing a higher percentage of smokers receive lots of promotional items from cigarette companies in order to compete for brand loyalty? And if less consumers are buying a product, doesn't it make sense to lower prices? Why would a store near a low smoking prevalance school keep their cigarette prices high if they wanted to entice kids to start smoking?

Maybe the retail stores are simply abiding by standard rules of economics - the more prevalent smoking is in a school, the more they work to keep smokers happy and coming back to their particular store for their next pack, by keeping prices down and making them feel important with all these PoP marketing gimmicks.

I guess one way to test the theory is to increase point-of-purchase ads and lower cigarette prices in stores near schools with low smoking prevalance, and vice versa near the higher smoking schools. Good heavens - what am I saying? Fox Butterfield, call your office immediately!