Dynamic Choice and TimingIndependence an experimental investigation..pdfVIP

Dynamic Choice and Timing-Independence: an experimental investigation. by Massimo Paradiso* and John D. Hey** Abstract: Timing-independence implies that individuals are indifferent between a sequential choice problem and a planned choice problem which are strategically equivalent except for the timing of resolution of the uncertainty. This paper reports an experiment in which we investigate whether the timing of resolution of the uncertainty affects individual preferences. We elicit individual preferences for three strategically equivalent decisions problems. The experimental results suggest that timing-independence is an inappropriate assumption of individual preferences. The paper discusses possible implications of such findings. JEL classifications: D81, C91, D90 Keywords: Dynamic decision making, Experiments, Timing-independence, Strategic equivalence. *University of Bari and **Universities of Bari and York We wish to thank the European Community under its TMR Programmes Savings and Pensions and ENDEAR for financing the experiment. We thank Antonella Trisorio for assistance in the conduct of the experiment. An early version of this paper has benefitted from helpful suggestions by Robin Cubitt and participants at the FUR Conference in Mons in July of 1997. Introduction The interpretation of a sequential (that is, dynamic) choice problem as a planned choice problem is widely supported by economists and decision theorists. The sequential choice problem is reduced to a non-sequential choice problem by the adoption of a plan (that is, of a decision strategy) as a tool to manage time in dynamic decision making. This reduction requires the decision-maker to be indifferent between choice problems that are strategically equivalent