The Utility of Wealth and the Utility of Windfalls.

Most economists, when forced to use a cardinal function, assume that the marginal utility of wealth declines continuously as wealth increases. This assumption is intuitively acceptable and mathematically convenient in many contexts. However, the fact that people gamble casts doubt on the validity of this assumption [3, 6, 1]. Richardson [9], in a discussion of Friedman and Savage [3], argues that their analysis ignores the factors which make planning important in the real world, and that they therefore overlook the "dislocation costs" associated with unanticipated changes in wealth which necessitate plan revision. In this paper models are considered in which a dislocation cost is formally incorporated; this is achieved by limiting consumption to the services of a single durable good which has a selling price, for the consumer, below its purchase price. If the consumer desires to consume more, or less, he must "trade in" his current "model" for a larger or smaller one. This terminology is clearly influenced by the automobile market: houses are a more important source of the effects we are studying. Although it is hard to imagine a world in which such goods are the only consumer goods there might be some demi-paradise in which all non-durables were free goods. [ABSTRACT FROM AUTHOR]