Blackwell Reference Online

Extract

The failure of the traditional time‐separable constant relative risk aversion consumption capital asset pricing model to explain the equity premium and risk‐free rate puzzles has led to numerous variations of the basic model being proposed. One successful approach which allows for non‐separability in utility over time is habit formation or habit persistence . In habit formation models it is not the absolute level of consumption which is important, but consumption relative to some benchmark level. Essentially, the representative agent's utility depends not only on current consumption but also on consumption in the previous period. Habit formation models typically define the utility function as taking the form U ( C t , X t ), where C t is consumption at time t and X t is the time varying habit or subsistence level which typically depends on previous consumption, X t = f ( C t −1 , C t −2 , …). The exact form of U ( C t , X t ) varies. Abel (1990) proposes that it should be a power function of the ratio C t / X t , while Constantinides (1990) , Sundaresan (1989) , and Campbell and Cochrane (1999) argue for a power function of the difference C t − X t . This distinction is important, since ratio models have constant risk aversion while difference models have time varying risk aversion. A further distinction between different types of habit formation model ... log in or subscribe to read full text