Daniel Shaviro
Published: 2023
Total Pages: 0
Get eBook
What time periods should we use in tax and other fiscal policy to evaluate people's circumstances, and thus to determine either how they are being treated, or how they ought to be? This question is both fundamental and pervasive.Standard economic reasoning offers grounds for entirely basing one's thinking on lifetime models. In particular, the closely related permanent income and life cycle hypotheses support employing a purely lifetime perspective in evaluating people's circumstances and treatment. The resulting model posits that people make decisions on a lifetime basis, seeking to optimize lifetime utility in the face of both (1) period-specific declining marginal utility of consumption, and (2) whatever preferences they happen to have as between consumption in different periods. Accordingly, in the presence of complete markets (including a lack of borrowing constraints), the question of when one earns a given dollar ostensibly makes no difference regarding when one spends it on consumption. And equivalently, when one pays a given dollar of tax will make no difference regarding how much one spends in any period.This model applies the same basic logic as a two-goods model in an Economics 101 casebook (featuring, say, pizza and movies), but in a far more complex setting in which its application is considerably more challenging. Despite its ruthless simplification, it likely has some degree of descriptive accuracy. People surely do make some plans across very long time horizons, such as early-life career choice, and subsequent planning (however imperfect it may be) for retirement.Yet the factors that undermine life cycle view's accuracy and normative relevance are not limited to borrowing constraints. Also of crucial importance are people's tendency to treat different periods as effectively separate, and a number of other constraints that would prevent them (even if so minded) from equalizing the marginal utility of consumption as between periods.In sum, therefore, the life cycle model is not sufficiently descriptively accurate to be treated as more than an important orienting benchmark. Like such other “it doesn't matter” theories as the Coase Theorem, the Efficient Markets Hypothesis, and the Modigliani-Miller Theorem, its value lies more in its showing us where to look for falsifying conditions, than in its actual empirical validity.