Essays on the fiscal policy implications of habit formation

A large and rapidly growing literature across many subfields of macroeconomics has recently argued that models with habits in consumption fit a variety of empirical data better than the standard model with intertemporally separable utility. But to date there has been very little research on how optimal economic policy is different in a world where consumers have habits. My dissertation explores the implications of habit formation for fiscal policy, with a focus on the economic and policy consequences of population aging.; I first analyze the implications of past and projected demographic changes for the U.S. saving rate using an infinite horizon model with habits. In contrast to the standard frameworks employed in previous studies, the habit formation model is shown to be able to match the historical data on savings during the 1970--1980 period. I use this model to simulate the optimal behavior of the national saving rate in the 21st century, which exhibits a smooth but significant decline over the next five decades. This contrasts with the immediate drop in savings generally predicted by models without habits.; Second, I focus on the fiscal policy and welfare implications of habits using an overlapping-generations model. My theoretical analysis suggests that when individuals are finitely-lived and have intergenerational habits, the optimal fiscal policy takes the form of a social insurance scheme, allowing present and unborn cohorts to share in the consequences of macroeconomic shocks. The essay concludes by pointing out that if habits exist but policy is conducted disregarding habits, the welfare losses for society can be substantial.; Finally, I explore the implications of habits for social security reform using a model augmented with life-cycle income and a pension system. It shows that reforms such as cutting benefits, increasing taxes, or gradually transiting out of the current pay-as-you-go system involve small costs to current generations in exchange for larger gains to future generations, and these gains/losses are magnified with habits. Moreover, if individuals have habits, reforming social security results in social gains that can be more than 100% higher than what the standard model predicts.