Forecasting among alternative strategies in the management of uncertainty

This paper explores why expectations elicited from corporate decision-makers may fall short of the rational expectations ideal: the formation of an informed forecast may just not be part of the optimal management of uncertainty. The analysis of an investment decision shows that when forecasting is expensive, this strategy is likely to be superseded by the possibility to wait, to switch between projects, or the alternative to eliminate inferior projects over time. Even at low forecasting costs it may be optimal not to forecast and instead to diversify over different projects. © 1998 John Wiley & Sons, Ltd.