Commitment in intertemporal household consumption: a revealed preference analysis

We present a revealed preference methodology for analyzing intertemporal household consumption behavior. In doing so, we follow a collective approach, which explicitly recognizes that multi-member households consist of multiple decision makers with their own rational preferences. Following original work of Mazzocco (2007), we develop tests that can empirically verify whether observed consumption behavior is consistent with (varying degrees of) intrahousehold commitment. In our set-up, commitment means that households choose consumption allocations on the ex ante Pareto frontier. The distinguishing feature of our tests is that they are entirely nonparametric, i.e. their implementation does not require an a priori (typically non-verifiable) specification of the intrahousehold decision process (e.g. individual utilities). We demonstrate the practical usefulness of our methodology by means of an empirical application. For the data at hand, our results suggest using a so-called limited commitment model that allows for household-specific commitment patterns. Importantly, our application also shows that bringing intertemporal dynamics in the empirical analysis can substantially increases the discriminatory power of the revealed preference methodology.