Time non-separable utility in life -cycle consumption and portfolio choice.

The inability of canonical models of consumption and portfolio allocation to yield empirically consistent results has led to some of the larger puzzles in the fields of macroeconomics and finance. Over the past decade, time non-separable utility has come to the rescue of rational expectation life-cycle models. This dissertation is a theoretical and empirical examination of time non-separable utility and its role in household life-cycle decision-making. The first two chapters of this dissertation examine habit formation in consumption and portfolio choice. Past consumption creates a liability that effectively reduces net wealth. The habit liability is the cost of the maintaining an accustomed standard of living. Effective wealth is defined as the difference between net wealth and the habit liability and shown to play the same role in decision making for a habit forming household as net wealth plays for a non-habit forming household. The first chapter provides life-cycle simulations indicating risky asset demand is greatly reduced by the habit liability both in levels and as a share of net wealth. Empirical results based on aggregate data provide evidence for habit formation in risky asset demand but none for consumption. The second chapter provides evidence of the habit liability's role in the demand for risky assets using household survey data. Results suggest that among households with the same wealth and future non-asset income, those with higher past consumption own less risky assets. The third chapter of this dissertation examines time non-separable utility in the form of a bequest motive. The goal of this chapter is to parametrically compare the consumption profiles of the elderly where both the existence of a bequest motive as well as the magnitude of how such a motive enters the decision making process is not assumed but is instead estimated. It is shown that between 50 and 70% of the elderly single population has a bequest motive. Of these households, 2 to 5% have sufficient wealth that they will almost certainly die with positive net worth. The remaining households with a bequest motive may die with positive net worth depending on their length of life.Ph.D.Economic theoryEconomicsFinanceSocial SciencesUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/123227/2/3068914.pd