The value of useless information

Abstract

There are many situations in which individuals have a choice of whether or not to observe eventual outcomes. In these instances, individuals often prefer to remain ignorant. These contexts are outside the scope of analysis of the standard von Neumann-Morgenstern (vNM) expected utility model, which does not distinguish between lotteries for which the agent sees the final outcome and those for which he does not. I develop a simple model that admits preferences for making an observation or for remaining in doubt. I then use this model to analyze the connection between preferences of this nature and risk-attitude. This framework accommodates a wide array of behavioral patterns that violate the vNM model, and that may not seem related, prima facie. For instance, it admits self-handicapping, in which an agent chooses to impair his own performance. It also accommodates a status quo bias without having recourse to framing effects, or to an explicit definition of reference points. In a political economy context, voters have strict incentives to shield themselves from information. In settings with other-regarding preferences, this model predicts observed behavior that seems inconsistent with either altruism or self-interested behavior.