Cost-benefit analysis often requires us to predict the future. Whether it is desirable to begin a project depends on what we expect will happen after we have begun. But, as mere mortals, we rarely are able to make precise predictions about the future. Indeed, in many situations analysts can be certain that circumstances largely beyond their clients’ control, such as epidemics, floods, bumper crops, or fluctuations in international oil prices, will greatly affect the benefits and costs that would be realized from proposed policies. How can analysts reasonably take account of these uncertainties in CBA?

In this chapter, we consider three topics relevant to uncertainty: expected value as a measure reflecting risks, sensitivity analysis as a way of investigating the robustness of net benefit estimates to different resolutions of uncertainty, and the value of information as a benefit category for CBA and as a guide for allocating analytical effort. Expected values take account of the dependence of benefits and costs on the occurrence of specific contingencies, or “states of the world” to which analysts are able to assign probabilities of occurrence. Sensitivity analysis is a way of acknowledging uncertainty about the values of important parameters in our predictions—it should be a component of almost any CBA. When analysts have opportunities for gaining additional information about costs or benefits, they may be able to value the information by explicitly modeling the uncertainty inherent in their decisions. A particular type of information value, called quasi-option value, is relevant when assessing currently available alternatives that have different implications for learning about the future.