Predicting Experimental Choice Behavior and Life Outcomes from a Survey Measure of Present Bias

Abstract

Using a representative sample of the German adult population, this paper investigates the extent to which a survey measure of present bias predicts present-biased choice behavior in incentive-compatible experiments and real-world outcomes related to investments in financial assets and human capital. The results are threefold. First, the survey and experimental measures of present bias are significantly related. Second, the survey measure predicts choices between immediate and delayed monetary payoffs in an incentive-compatible experiment, but not between payoffs at two future points in time. Third, the survey measure of present bias is a good predictor of the propensity to save money, to obtain a university degree, and to maintain a healthy life style. In most specifications, the survey measure tends to be a stronger predictor of real life outcomes than the experimentally elicited measure of present bias.