Correcting for bias when average values are used to compute changes in consumer surplus

Abstract

Calculation of a change in consumer surplus is often used when evaluating the welfare impact of a change in the level of a nonmarket good. Unfortunately, because of a lack of data, estimation of the required demand curve is often infeasible. Instead, a proxy measure is often used that combines a measure of average consumer's surplus, such as recreational visitor‐day values, with a prediction of changes in use levels. This article examines the potential biases and cases in which this can be corrected by the use of this benefits transfer technique.

Keywords:

• bias
• consumer surplus
• linear model of demand
• semilog model of demand