Although the contingent valuation [i]modus operandi[/i] has been widely used to value a diverse array of nonmarket environmental and natural resource commodities.

Although the contingent valuation [i]modus operandi[/i] has been widely used to value a diverse array of nonmarket environmental and natural resource commodities, late empirical evidence suggests it may not accurately estimate real economic values. The hypothetical nature of environmental valuation observes typically results in responses that are significantly greater than actual payments. Economists have had mixed succes in developing techniques designed to check for this "hypothetical bias." This paper highlights the part of experimental economics in addressing hypothetical bias, and identifies a gap in the existing literature by way of focusing on the underlying causes of this bias. chiefly of the calibration techniques used today lack a theoretical justification, and therefore these manner of proceedings need to be used with caution. We argue that what is yet to be experimental research should investigate the reasons hypothetical bias persists. A better understanding of the causes should enhance the effectiveness of calibration techniques.

Key Words: contingent valuation, experiments, hypothetical bias, stated preference

Consider the challenge faced through a contingent valuation (CV) practitioner who is interested in estimating the economic value of a non-market serviceable such as visibility at a National Park or the protection of habitat for an endangered species. The CV overlook is carefully designed and instituteed (e.g., Mitchell and Carson, 1989; Champ, Brown and Boyle 2004) and the be deriveds are produced with the latest estimation techniques (Haab and McConnell 2003) We now have an estimate for the economic value of the good-but is this value accurate?

The answer to this question has stirred considerable, and sometimes contentious debate, as highlighted by means of litigation resulting from the 1989 Exxon Valdez oil spill in Prince William vigorous [see Diamond and Hausman (1994); Hanemann (1994); and Portney (1994) for a synthesis of the debate]. Using simply field CV data, we cannot be certain that value estimates are accurate. Why? Since CV reviews are hypothetical in both the payment for and provision of the well adapted in question, we do not know whether what an individual says she would do in a hypothetical setting matches what she will do when actually given the opportunity to do so1 And, without the ability to take note of the latter, it is difficult to confirm whether the values elicited from a hypothetical contemplate accurately reflect the real economic value of the upright Some researchers have expressed touch that this lack of a consequential economic commitment in CV examines often leads to hypothetical bias in which economic values are overstated. For example, as Harrison and Rutstr?¶m (forthcoming) assert: "As a matter of logic, if you do not have to pay for the useful but a higher verbal willingness-to-pay replication increases the chance of its provision, then verbalize away to increase your awaited utility!"

Economics experiments offer the potential to shed a certain light on the accuracy of replys to hypothetical CV questions. Experimental research has a well-established framework which was widely recognized when Vernon Smith became a co-recipient of the Nobel Memorial Prize in Economics "for having established laboratory experiments as a tool in empirical economic analysis." What distinguishes experiments from other empirical techniques are direction and replication. The ability to sway the environment under which individuals make economic decisions is what gives experiments power. The experimenter can vary treatments to proof hypotheses about the effects of different explanatory variables forward individual choices. Unlike a typical field CV take a view of a carefully designed experiment can include the couple hypothetical and real payment scenarios. on comparing outcomes in these pair settings, one can make one inferences about the existence of hypothetical bias, its causes, and ways to mitigate its consequences Moreover, other researchers can replicate, and perhaps increase the experiment to test its robustness. Generally, it is the carcass of experimental evidence, rather than a single close attention that allows us to draw more reliable conclusions about what we do and do not know (Roth 1988)

The existence of hypothetical bias has been welldocumented in the two laboratory and field settings. In a modern survey of the literature, Harrison and Rutstr?¶m (forthcoming) set up a positive bias in 32 of 39 observations. There are pair meta-analyses of the experimental hypothetical bias literature: List and Gallet (2001) and Murphy et al. (2003) The latter cogitation updates the List and Gallet data for an coding differences and conducts a sensitivity analysis of their conclusions The results of both metaanalyses are consistent with findings reported by the agency of Harrison and Rutstr?¶m, suggesting that mean hypothetical values are about 25 to 3 times greater than actual values (but this take rises from a highly skewed distribution with a median ratio closer to 15) Figure 1 nears the distribution of this bias for 83 observations from 23 studies (Murphy et al., 2003)