Bracelet-Based Policy Making
Posted by The Situationist Staff on September 29, 2008
Ever wonder why so much time is spent comparing jewelry and telling personal anecdotes in presidential debates (see short video above), even as many of the larger policy questions remain largely unexplored? There are, of course, many reasons (some of which have been noted in previous Situationist posts), but Friday’s wrist-off reminded us of one key contributor: “the identifiable victim effect” – greater sympathy is felt for identifiable victims than for statistical victims.
George Loewenstein, Deborah Small, and Jeff Strnad have an excellent 2005 paper, “Statistical, Identifiable and Iconic Victims and Perpetrators” (available for free downloading on SSRN), discussing that effect and its role in policymaking. Their introduction includes this paragraph.
In this paper we focus on yet another reason for why taxation and government spending can go awry: human psychology, and specifically the lack of proportionality between human sympathy and the wants and needs of those toward whom the sympathy, or lack thereof, is directed. As Adam Smith observed in the Theory of Moral Sentiments, we often feel little sympathy toward people who are highly deserving of it. He illustrates the point vividly with the hypothetical case of a European man who gets more upset over losing his little finger than over a calamity that wipes out the entire population of China. However, the disproportionality can also go in the opposite direction. As Smith also points out, “we sometimes feel for another, a passion of which he himself seems to be altogether incapable,” as illustrated by the dismay of the mother of a sick child which, as he puts it, “feels only the uneasiness of the present instant, which can never be great” . . . . Smith adds dryly that “we sympathize even with the dead, who themselves experience nothing”. . . . Our main focus is on one specific source of arbitrariness in human sympathy: the disproportionate sympathy and attention to identifiable as compared with statistical victims.
Here is the abstract.
We first review research showing (1) that people respond more strongly to identifiable than statistical victims even when identification provides absolutely no information about the victims, (2) that the identifiable victim effect is a special case of a more general tendency to react more strongly to identifiable others whether they evoke sympathy or other emotions, and (3) that identifiability influences behavior via the emotional reactions it evokes. Next, we discuss the normative status of the effect, noting that, contrary to the usual assumption that people overreact to identifiable victims, identifiability can shift people’s responses in a normatively desirable direction if people are otherwise insufficiently sympathetic toward statistical victims. Finally, we examine implications of the identifiable victim effect for public finance. We show that the identifiable victim effect can influence the popularity of different policies, for example, naturally favoring hidden taxes over those whose incidence is more easily assessed, since a hidden tax has no identifiable victims. Identifiable other effects also influence public discourse, with much of the debate about government spending and taxation being driven by vivid exemplars – iconic victims and perpetrators – rather than any rational calculation of costs and benefits.