Welfare Measures When Consumers Can Learn: A Unifying Theory

We extend Graham (1981, AER, 715-725)'s welfare analysis under uncertainty to a dynamic environment where the agent can delay and obtain information to reduce the uncertainty. Our analysis yields a dynamic willingness to pay (WTP) locus that unifies the distinct concepts of option price, dynamic WTP, commitment costs, and quasi-option value. Option price corresponds to the ex ante WTP when the agent ignores the learning opportunity, and dynamic WTP is the value when such opportunity is recognized. The commitment cost is the difference between option price and dynamic WTP, which is also the annualized and monetized quasi-option value. The dynamic locus shows that there is an added value of state contingent contracts because such contracts allow trading among agents with heterogeneous degrees of access to future information and delay opportunities. We also show that making quasi-option value adjustments in project analysis, when project values are obtained through nonmarket methods, could result in double counting of quasi-option values