Wednesday, March 13, 2013

Jolls on image-based warnings on cigarette packs

Christine Jolls, Product Warnings, Debiasing, and Free Speech: The Case of Tobacco Regulation, 169 Journal of Institutional and Theoretical Economics 53-78 (2013)

Abstract:

The Family Smoking Prevention and Tobacco Control Act of 2009 requires the display of graphic health warnings on cigarette advertising and packaging in the United States. Debates over the permissibility of these new mandated health warnings under the unusually broad Free Speech Clause of the United States Constitution have paid insufficient attention to empirical evidence --- to be presented in this article --- of the warnings’ salutary effects in reducing consumers’factual mis-perceptions about smoking risks. Although such empirical evidence does not, by itself, settle the First Amendment debate, this evidence warrants more attention in that debate than it has received to date.

Her primary conclusions:

First, the empirical evidence suggests that the FSPTCA’s mandated health warnings enjoy at least some efficacy in reducing consumers’ factual misperceptions of smoking risks. Second, the factual accuracy of risk perceptions is not automatically or easily increased by even gripping, highly salient warnings – providing some suggestion that the danger of overcorrection of consumers’ factual misperceptions in this domain is relatively small.

…To the extent that this line of reasoning [in the cases striking down image warnings] condemns graphic images as distinctly ill-suited, by comparison with text statements, to increasing the factual accuracy of consumers’ risk perceptions, the argument is not easy to square with a sensible understanding of factual misperceptions and their amelioration through mandated disclosures. On the one hand, decades of empirical studies suggest that purely textual material may produce different beliefs about factual matters by virtue of details such as the choice of font; there is not a clear, pristine mapping from text statements to individuals’ perceptions about matters of fact. On the other hand, there are many instances in which a photograph, drawing, or other graphic image is undoubtedly better suited than text alone to conveying factual information; the poison symbol used on many products and the appearance of pictures and diagrams throughout a typical medical or scientific text provide ready examples.

No comments: