THE VALUE OF LIFE: A CONTINGENT VALUATION STUDY OF RISK AVOIDANCE AND UTAH COUNTY AIR POLLUTION

by

Print Friendly, PDF & Email

Christopher Michael Lee, Economics

Introduction

While prices established in competitive markets provide the most traditional and widely accepted valuation of goods and services, many commodities cannot be valued with price studies because no competitive market exists for those commodities. Economist have developed some methods by which valuations for non-marketed commodities can be estimated. These non-market methods are significant because they can estimate values for public goods, non-traded goods and services, and other items or amenities which the market does not or cannot allocate.

Survey Instrument Design

A telephone survey of 113 Provo and Orem residents was performed during February, 1994. Using funds received from the BYU Office of Research and Creative Work, two BYU students were employed to call these residents, read the survey questionnaire, and record responses.

After asking for participation in the survey, the surveyor presented a brief paragraph of information regarding the health effects of particulate emissions in Utah County. At the end of this information, the surveyor cited a “risk of death’.’ due to air pollution. This figure was based simply on the number of deaths estimated to have been caused by air pollution per year divided by the population of Utah County. Thus, the figure does not take into account varying risks for those with respiratory problems or for age differences in risks. Instead, each respondent is asked to evaluate his or her willingness to pay for reduced risk of mortality based on one standard probability of one in 3,000. The surveyor then gave a comparison of risk or mortality from driving in a car in Utah County, to orient the respondent somewhat to the magnitude of the air pollution risk. The Utah Highway Safety Department provided data for each of the past six years. The average number of deaths per year was 33, making the overall “risk of death” about one in 6,000 (Dame 1993).

After providing this brief bit of information, the surveyor then proceeded with the hypothetical questions. Specifically, two types of surveys were performed. Three main questions were asked each group, in opposite order for each group. The two orders of questions allows for the determination of a specific type of instrument bias. The first question asked the individuals’s valuation of reduced air pollution, provided that only the respondent him or herself receives the benefits of reduced risk. The second asked the valuation of reduced air pollution, provided that benefits are accrued by the respondent and his or her immediate family. Finally, the third question asked the valuation of reduced air pollution with benefits to the individual and his or her family and friends (the community as a whole).

Discussion

Details and tables of the responses are found in a larger document submitted to the Honor’s Office at Brigham Young University. Descriptive statistics, a discussion or potential bias, and copies or the instruments can be found there.

Although the study does offer some interesting avenues for analysis, it appears to demonstrate inaccuracy on the basis of ordering bias, unexplained variation, and high life value estimates. This inaccuracy has been shown by other researchers to pose significant problems for researchers using the open-ended questionnaire for contingent market studies.

The hypothetical nature of this type of study does not enable persons to respond with orientation or bearings on actual budget constraints, life valuation, or other factors. Instead, an individual, when faced with an open-ended question such as this study posed, simply blurts out some rounded number that seems plausible. For this reason, the median for both questions was $100. Individuals just say the first number that comes to them in response to the open question. $100 is perhaps the most round of numbers.

One possible solution to this problem of open-ended questionnaires for future researchers is to ask for a variety of valuations for reduced risks, thereby perhaps causing individuals to think about trade-offs in bidding including their own budget constraints. In addition, future use of this data may present interesting findings regarding altruistic behavior related to the three bid types in this study.

Conclusions

Several conclusions may be drawn from this study. First, although value-of-life estimates seem high compared with other studies, this may demonstrate that residents of Utah County place a high value on air quality. This value reflects both mortality rate reductions and quality of life/ morbidity rate improvements. Using air pollution as a vehicle for determining life valuation has this as its major fault: there is no way of determining which part of an individual’s bid is due to reducing mortality risk and which part is due to improvements in morbidity rates and quality of life. Second, the data reveals a high variance in responses. This may be attributed to the high amount of-information or -awareness-that respondents-ofthe survey had prior to the questionnaire_ being_give_n them. This prior awareness was not measured by the survey, so it could not be included as an explanatory variable. The variance perhaps could have been reduced by increasing the sample size.

However, the high variance also points to problems implicit in this type of open-ended questionnaire in which respondents must pnll numbers “out of the air” to make a bid. The findings of this study support the results of previous studies of open-ended questionnaires.

Third, the data offers little opportunity to explain the variance of bids. The ten to fifteen explanatory variables used in the econometric model give no conclusive explanation of individual, family, or community bids. This problem is especially significant since variables which intuition suggests should have explanatory power do not explain bids.

Bibliography

  • Brookshire, D.S., B. C. Ives, W. D. Schulze. 1976. The Valuation of Aesthetic Preferences. Journal of Environmental Economics and Management. March. 325-346.
  • Dame, Richard. 1994. Personal interview with author.
  • Loomis, John B. and Richard G. Walsh. 1986. Assessing Wildlife and Environmental Values in CostBenefit Analysis: State of the Art. Journal of Environmental Management. 125-131.
  • Maddala, G.S. Introduction to Econometrics. 2nd ed. 1991. New Jersey: MacMillan Publishing.
  • Miner, Dean and C. Arden Pope III. 1986. Willingness to Pay for Improved Air Quality in Utah County. Research Report, Ezra Taft Benson Agriculture and Food Institute, BYU.
  • Randall, Alan, B. C. Ives, and C. Eastman. 1974. Bidding Games for Valuation of Aesthetic Environmental Improvements. Journal of Environmental Economics and Management. January. 132- 149.
  • Rowe, d’ Arge, Brookshire. 1986. An Experiment on the Economic Value of Visibility. Journal of Environmental Economics and Management. July. 1-119.
  • Seller, Christine, John R. Stoll, and Jean-Paul Chavas. 1985. Validation of Empirical Measures of Welfare Change: An Comparison of Nonmarket Techniques. Land Economics. May. 156-17 5.
  • Stoll, John R., Webb Smathers, and Robert N. Shulstad. 1983. An Overview of Current Issues in Nonmarket Valuation. In Nonmarket Valuation: Current Status, Future Directions. Southern Natural Resource Economic Committee. 1-29.
  • Thayer, M.S. 1981. Contingent Valuation Techniques for Assessing Environmental Impacts: Further Evidence. Journal of Environmental Economics and Management. August. 27-44. U.S. Dept. of Commerce. 1991. Statistical Abstracts of the United States. Washington: U.S. Dept. of Commerce.
  • Viscusi, Kip. 1992. Survey or Risks to Life and Health. Risk Valuation: Concepts and Estimates. Chicago: University of Chicago Press.