The Revealed Preference Approach to Collective Consumption Behavior: Nonparametric Testing and Sharing Rule Recovery (Revised version of CentER DP 2007-73)

We present a nonparametric 'revealed preference' methodology for empiri- cally analyzing collective consumption behavior. First, we introduce an integer programming (IP) methodology for testing data consistency with collective con- sumption models that account for publicly as well as privately consumed goods. This IP methodology can include information on 'assignable quantities' for pri- vate goods. Next, we show that the IP methodology allows for recovering the personalized (Lindahl) prices for the public goods and the personalized quan- tities for the private goods. In turn, this implies recovery of the sharing rule (i.e. personalized income share levels). Numerical examples and an empirical application demonstrate the practical usefulness of the methodology.