Essays on consumption with habit formation

by Seckin, Zennube Aylin

The aim of this thesis is to examine the implications of habit formation in a consurnption-saving mode1 with income uncertainty. By introducing habit formation, we have shown that habit formation itself leads to prudent behavior, in addition to the usual precautionary saving motive against incorne uncertainty. We have shown that with habit formation, even in the quadratic case where the precautionary saving premium is zero, there is habits-induced saving. in the exponential case, we have obtained a closed fonn solution for consumption, which will provide a better framework for empirically testing the behavior of consumption over the business cycle. By allowing habit-forming preferences in consumption, we did not need to assume Iiigher levels of income uncertainty to explain the excess sensitivity, the excess smoothness and the excess growth puzzles of consumption. We have also shown that the greater is the strength of habit formation the lower is the effect of income uncertainty on consurnption. By introducing liquidity constraints into the model with habit formation. we have shown that the effects of present and future shadow prices of liquidity constraint will be present and rise with the strength of habit formation. In a consumption-leisure choice model with a Cobb-Douglas utility function and habit formation in consumption, the intertemporal marginal rate of substitution between consumption and ieisure and the intertemporal marginal rate of substitution of leisure are higher the higher is the strength of habit formation. This result implies that the individual is less willing to intertemporally substitute current leisure with future leisure and with current consumption. With the liquidity constraints depending on wage eamings, the intratemporal marginal rate of substitution between consumption and leisure decreases whenever the constraint binds. FinaIly, in a two-period model, we have shown that habit formation leads to a larger prudence premiurn and greater precautionary saving, provided that the individual has decreasing absolute prudence. However, we have found that it is not possible to conclude, even with the utility function exhibiting a simple form of habit formation, that habits increase or decrease the precautionary premiurn "in the sense of Kimball" when there exists multiple periods. This work has left me with a Iarge debt of gratitude. First, 1 wish to express rny thanks to Professor Fanny S. Demers and Professor Michel Demers for their immense patience. understanding, guidance and encouragement. Without their support. it would have been difficult for me to accomplish this work. 1also thank Professor Steve Fems for kindly accepting to CO-supervise my thesis. His critical comments allowed me to improve substantially the early drafts of this work. 1am grateful to Ginette Harte, the Secretary of Graduate Studies at the Department of Economics, for her kind support and help in fulfilling periodic formalities throughout my years at Carleton. 1 thank to my parents who supported me in numerous ways while 1 was undertaking this endeavor and constantly motivated me to complete it. They telephoned me every Sunday Morning from Istanbul to wish me luck and courage. They have been always by my side. Finally, 1 wish to express my ever-standing gratitude to my husband Kaan who suffered and enjoyed with me the various stages of my studies, and constantly motivated me to accompIish this important step of my professional carrier. He always gave me the strength and hope that I needed to keep going during the most difficult times of distress. He helped me in every ways to facilitate my Iife and nsver complained about my repeated periods of desperation. 1 am deeply indebted to him for his immense love, compassion and understanding.