Ordinarily, a February 29 post wouldn't make it to the blog (I have several of these 2/29 posts that have vanished in previous common years).
What follows is the introduction to my chapter in the Cherry, Kroll and Shogren edited volume, Environmental Economics, Experimental Methods, that was orginally titled "Confessions of a Contingent Valuation Economist."
As all environmental economists well know, ad nauseam, on March 24, 1989 the Exxon Valdez spilled 11 million gallons of oil into Prince William Sound. Less well known, on the same day I was busy working on my dissertation on the effects of substitute environmental goods on existence value. For better or for worse, I bumbled into one of the key issues surrounding the magnitude of natural resource damages associated with the oil spill and, more broadly, one of the most active research literatures in the annals of environmental economics.
This unfortunate event shaped the research agendas of many environmental economists during the 1990s due to the magnitude of the disaster and the pecuniary and non-pecuniary richness of the intellectual activity. What became known as the "contingent valuation debate" focused on the issue of what were then known as existence values, then non-use values and finally have come to be known as passive use values. Passive use values are the values that people have for the environment even if they do not use the environmental resources on-site (e.g., a fishing trip). People may value the existence of an environmental resource without revealing their value through behavior (e.g. a fishing trip) which makes the measurement of passive use values a considerable challenge. The contingent valuation method, a hypothetical survey approach in which "willingness to pay" questions are asked, is still the most commonly used method to assess the extent of passive use values.
Back in those dark days experimental economists were already attempting to debunk the notion that hypothetical willingness to pay statements had much validity at all. The publication of Mitchell and Carson’s CVM book in 1989 was a trumpet call for contingent valuation economists (i.e., those that faithfully “do” CVM) to circle the wagons and claim that hypothetical willingness to pay statements are equivalent to true willingness to pay statements if CVM is done correctly (i.e., conducted by like-minded folks with significant research budgets). A large number of experimental and survey researchers have disputed this claim, and they are probably correct.
Fifteen years later, most all contingent valuation economists, including myself, have grudgingly accepted the fact that hypothetical bias exists and is the biggest challenge confronting the use of CVM estimates for policy analysis. Attention has turned to what to do about hypothetical bias and other problems. Grudgingly, I must admit that much has been learned from the experimental economics laboratory. Indeed, many contingent valuation economists (not me) have become real scientists and conducted their own laboratory experiments.
This self-absorbed story serves to illustrate the main contribution of experimental economics to applied environmental valuation research. The CVM is a very simple and powerful valuation methodology. You can estimate the use values and passive use values of the craziest things. But, in the wrong hands, it can be embarrassingly abused. Experimental economists can bring contingent valuation economists out of their orbit around a far off hypothetical planet (e.g., Pluto) and back down to earth.
And there ya go.
