The saving behavior of the wealthy has received remarkably little academic attention in the past twenty years or so. This is probably largely attributable to a relative lack of good data: The Survey of Consumer Finances is virtually the only publicly available source of detailed data on wealthy households, and even the SCF has only a few hundred really wealthy households in each triennial wave. Despite recent neglect, the topic is an important one for scholars of saving behavior, for at least two reasons. First, wealthy households should provide a powerful means of testing whether the standard model of consumer behavior, the Life Cycle/Permanent Income Hypothesis, is adequate as a universal model of saving and consumption. This is an application of the general scientific principle that models should be tested under extreme conditions; if they do not hold up, a new model (or an extended version of the old one) is called for. The second reason for studying the wealthy is that they account for a large share of aggregate wealth. In fact, some understanding of the saving behavior of the wealthy is probably indispensable to any credible attempt to account for the magnitude of aggregate wealth.
Although the primary source of evidence in this paper will be the four Surveys of Consumer Finances conducted in 1983, 1989, 1992, and 1995, the inevitable limitations of those data will be apparent. The paper therefore also relies to a considerable extent on unorthodox kinds of evidence, ranging from information in the annual Forbes 400 tabulation of the richest American households, to quotations from and about the very rich, to the results of a “focus group” meeting with a set of wealthy individuals who were directly asked their reasons for saving.