Linde Institute/SISL Seminar: Shachar Kariv
Our work is motivated by the need to provide a better, positive account of preferences. To provide that account, we need a choice environment that is rich enough to allow a general characterization of the patterns of individual behavior. In addition, to characterize behavior at the level of the individual subject, it is necessary to generate many observations per subject over a wide range of choice sets. In prior work, we developed a graphical interface for exactly this purpose. With the interface, subjects see on a computer screen a geometrical representation of a standard consumer decision problems (selection of a bundle of commodities from a standard budget set) and choose portfolios through a simple "point-and-click." The experimental method is applicable to many types of individual choice problems. In this paper, we study, intertemporal choice. In the experiment, subjects choose, from a budget line, bundles consisting of some income sooner and some income later. The budget lines vary randomly and cross often, now reflecting different interest rates as well as endowments.
The long-standing interest in intertemporal choice has, in recent years, been further fueled by evidence of non-constant time discounting and a better understanding of its theoretical consequences. Several studies, can be interpreted to show that time discount rates decline as tradeoffs are pushed into the temporal distance. In particular, subjects often choose the larger and later of two rewards when both are distant in time, but prefer the smaller and earlier one as both rewards draw nearer to the present. There are, however, several important and, to our knowledge, unexplored questions about choice over time that seem fundamental to questions about the form of time discounting: How consistent are intertemporal choices with any economic model of utility maximization? How do differences in the rationality of intertemporal choice relate to parametric estimates of time-discount rates for the same individuals? We show that apparent non-stationarities in intertemporal tradeoffs (e.g. β≠1 in the beta-delta model) in fact reflects inconsistency with utility maximization rather than non-stationarity in flow utilities or time discount rates.