Essays on habit formation and catching up with the Joneses

The starting point of relative consumption models is to postulate that consumer preferences are influenced not only by a household's own current consumption, but also by its own past consumption, or by average current and past consumption among all households in the economy. When a household's utility is affected by its own past consumption, the household is said to exhibit "habit formation". If the household's utility is affected by either average contemporaneous or average lagged consumption in the economy, then households are "keeping up with the Joneses" or "catching up with the Joneses" respectively. My dissertation focuses on evaluating the effect of relative consumption on household decision making. The first chapter focuses on summarizing the existing research in these areas. The following two chapters consist of an empirical study examining the effect of habit formation on household food consumption, and a theoretical study analyzing the effect of habits and catching up with the Joneses preferences on optimal taxation and the associated welfare gains.;Much empirical research has been done to investigate the existence of habits in the household preference structure, but this research has always specified a particular functional form for a utility function. A parametric model may not adequately capture the size of the habit formation parameter. One of the objectives in my dissertation is to examine a model of habit formation, where current and lagged consumption are nonlinearly related but without imposing a functional form for the model. A test for neglected nonlinearity on the same data verifies that there is a nonlinear relationship between current and lagged consumption. Therefore, without specifying a particular functional form, a semiparametric model will better represent the relationship between the variables.;Next I consider a model where a household's utility function exhibits habit formation and catching up with the Joneses preferences. The competitive equilibrium of this economy is not Pareto optimal since households do not internalize the consumption externalities they generate. Consequently, there is room for government intervention in the form of taxes on capital and labor income. Given this preference structure, I derive the conditions for the optimal tax rates that will induce agents to choose consumption levels that are the same as the social planner's solution.