Intertemporal choice under uncertainty: A behavioral perspective

This dissertation explores the relationship between delay and uncertainty in risky intertemporal decision-making. In the first part, I investigate the interaction between delay and uncertainty effects. Contrary to normative predictions, I find that delay and risk discounting are not independent from each other. In particular, the delay discount rate increases for uncertain future outcomes and the risk discount rate decreases with an increase in delay. Thus individuals are more impatient for gambles than for certain outcomes and less risk-averse (more risk-neutral) for delayed outcomes than for immediate ones. The findings also suggest that the simultaneous presence of delay and uncertainty leads to a higher discounting of risky future outcomes than predicted by either effect separately. The second part of the dissertation focuses on two applications of this result: risk mitigation investments in individual decision-making and new product development in managerial decision-making. Based on the general framework developed in the first part, I examine the interaction between delay and uncertainty in these two domains. In both applications, I find little empirical support for the normative prediction that the effects of delay and uncertainty are independent of each other. The results are consistent with the findings from Part I. In the first study, a comparative analysis of risk mitigation measures and energy-efficiency reveals that investments with a risky stream of future benefits (e.g. risk mitigation) are discounted more heavily than those with a certain stream of future benefits (e.g. energy efficiency). This is consistent with the result that individuals are more impatient for risky outcomes than for certain ones. The second study, which focuses on new product development, shows that the effect of delay and uncertainty is in the same direction as in the generalized case, leading to more discounting than either effect separately would predict. The dissertation concludes that the time and risk preferences depend on each other and the interaction between delay and uncertainty causes a higher discounting of risky future outcomes than predicted by either effect separately