The contingent valuation method (CVM) is a stated preference approach to the valuation of non-market goods. It has a 50+-year history beginning with a clever suggestion to simply ask people for their consumer surplus. The first study was conducted in the 1960s and over 10,000 studies have been conducted to date. The CVM is used to estimate the use and non-use values of changes in the environment. It is one of the more flexible valuation methods, having been applied in a large number of contexts and policies. The CVM requires construction of a hypothetical scenario that makes clear what will be received in exchange for payment. The scenario must be realistic and consequential. Economists prefer revealed preference methods for environmental valuation due to their reliance on actual behavior data. In unguarded moments, economists are quick to condemn stated preference methods due to their reliance on hypothetical behavior data. Stated preference methods should be seen as approaches to providing estimates of the value of certain changes in the allocation of environmental and natural resources for which no other method can be used. The CVM has a tortured history, having suffered slings and arrows from industry-funded critics following the Exxon Valdez and British Petroleum (BP)–Deepwater Horizon oil spills. The critics have harped on studies that fail certain tests of hypothetical bias and scope, among others. Nonetheless, CVM proponents have found that it produces similar value estimates to those estimated from revealed preference methods such as the travel cost and hedonic methods. The CVM has produced willingness to pay (WTP) estimates that exhibit internal validity. CVM research teams must have a range of capabilities. A CVM study involves survey design so that the elicited WTP estimates have face validity. Questionnaire development and data collection are skills that must be mastered. Welfare economic theory is used to guide empirical tests of theory such as the scope test. Limited dependent variable econometric methods are often used with panel data to test value models and develop estimates of WTP. The popularity of the CVM is on the wane; indeed, another name for this article could be “the rise and fall of CVM,” not because the CVM is any less useful than other valuation methods. It is because the best practice in the CVM is merging with discrete choice experiments, and researchers seem to prefer to call their approach discrete choice experiments. Nevertheless, the problems that plague discrete choice experiments are the same as those that plague contingent valuation. Discrete choice experiment–contingent valuation–stated preference researchers should continue down the same familiar path of methods development.
Measuring network effects in mobile telecommunications markets with stated-preference valuation methods
