The Value of Information is driven by the Intensity of Preference

According to previous research, the value of information about a risky decision does not seem to depend in any consistent or predictable fashion on the features of the decision problem. For instance, the relationship between value of information and the degree of risk aversion can be non-monotonic, seemingly counterintuitive in some cases. This paper shows that value of information obeys a simple and general principle. The value of information is systematically driven by the intensity of preference between the decision alternatives: it will be maximal when the decision maker is indifferent between the prior alternatives, and lower as the preference for one of the alternatives is stronger. This result holds for any given choice alternatives, information structure, random background wealth, and utility function. The result bears some intuition. It also indicates that quasi-concave relationships should generally be expected between value of information and specific parameters of the decision problem.