ABSTRACT
There is some controversy associated with the language describing monetary values for changes in health or mortality risks, especially the term ‘value of a statistical life’ (VSL). We investigate if the general public distinguishes differences in language describing the concept of VSL using four different descriptive treatments. Based on a survey of willingness to accept (WTA) in which individuals receive payment to participate in a future study, results show that the general public does not perceive a difference in the language used to describe changes in risks. This suggests that public objections to the VSL may not be caused by the specific description used nor can be ameliorated by refinement. Further, the results alleviate concern that language chosen may affect respondent participation and welfare estimates. Because of its apparent absence in the literature, we also adapt and demonstrate a non-parametric conservative estimate of WTA.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 A search on Econlit of ‘Value of a Statistical Life’ for various works published in January 2011 or later produced 81 results.
2 For example, a ‘usual cash payout is generally $1–$3 for surveys that take anywhere from 5 to 45 min’ (ABC News Citation2006).
3 One suggestion was to model the probability of rejecting the question based on the information treatment but was deemed inappropriate and was confirmed by the infrequency of this selection, between 1.8% and 5% per treatment.
4 One could argue that this explanation ameliorates the potential protest of respondents. We posit that if the protest can be counteracted with 35 words of explanation and an example, then the potential magnitude of the issue was minor at the outset, but it is apparently not the case given the amount of media coverage and public discussion.
5 An e-mail address or phone number would create the strongest hurdle but violates company protocols of Personally Identifiable Information (PII). Some respondents provided additional PII without a request.
6 While the benefits of double-bounded dichotomous choice are known (Hanemann, Loomis, and Kanninen Citation1991), it was not employed in order to maintain the highest level of credibility and simplicity of the question.