Modigliani 1988 a similar theory was proposed by

This preview shows page 251 - 253 out of 259 pages.

Modigliani (1988). A similar theory was proposed by another Nobel winner,Milton Friedman. His (1957) theory is called thepermanent incomehypothesis.The distinctions between the two theories are not particularlyimportant for the points raised in this chapter.2The marginal propensity to consume, or MPC, is the fraction of anincremental dollar that is spent rather than saved. If an individual receives awindfall of $100 and spends $95, saving $5, then the MPC is .95.3This chapter draws heavily on my joint work with Hersh Shefrin, Thaler andShefrin (1981) and Shefrin and Thaler (1988). Details of a model of savingsbehavior based on mental accounting and self-control are available in thelatter paper.4In a provocative paper, Deaton (1987) has argued that consumption isactually too smooth, rather than too variable. However, Deaton does notdispute the fact that consumption depends too much on current income.Rather, he argues that innovations in labor income areunderestimatesofchanges in permanent income, so consumption changes should begreaterthan income changes. The correct interpretation depends on the stochasticproperties of income.5It is worth noting that even the permanent income consumers in their modelhave an intertemporal elasticity of consumption of close to zero. A similarresult in the context of a permanent income model is obtained by Hall (1988).6Actually, since 1975 the benefit levels were indexed to the CPI, so thechanges were quite predictable months before the announcement date.7It is frustratingly difficult to test the idea that the timing of income flowswithin the year might influence consumption behavior. Perhaps because ofthe prediction of economic theory that such matters are irrelevant, nostandard data set includes questions about the magnitude and size of irregularincome flows such as bonuses.8The authors argue strongly that the bonuses should not be consideredtransitory income since they are well anticipated. They also use expectationsdata to test the hypothesis that workers spend unanticipated bonusesdifferently than expected bonuses, but find no evidence to support this view.9Most households have little in the way of liquid assets, even when they first
reach retirement age. This fact, in and of itself, supports the view that self-control issues are paramount in studying saving. The vast majority ofhouseholds do virtually no long-term “discretionary” saving.10There are two important components of pension wealth in the U.S., SocialSecurity benefits and private pensions. There is a large literature in eachdomain estimating this savings offset. The estimation problems are muchmore difficult for Social Security wealth because an individual’s SocialSecurity wealth is so highly correlated with age and prior earnings. Aftercontrolling for these two factors, there is essentially no cross-sectionalvariation in Social Security wealth. I will therefore just summarize theliterature on private pensions. However, see Barro (1978) (which contains areply from Feldstein) for a review of Social Security-savings literature.

Upload your study docs or become a

Course Hero member to access this document

Upload your study docs or become a

Course Hero member to access this document

End of preview. Want to read all 259 pages?

Upload your study docs or become a

Course Hero member to access this document

Term
Fall
Professor
ALAN JEFF
Tags
Economics, Game Theory, Public Good, Daniel Kahneman, Richard H Thaler

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture