Value of statistical life and cause of accident: A choice experiment
2008 (English)Conference paper (Refereed)
The results from this study are used to compare the marginal willingness to pay for risk reductions and calculate corresponding values of statistical life for road, fire and drowning accidents in Sweden. The values were estimated using a mail survey with 5200 respondents where each respondent was required to answer nine different dichotomous willingness to pay questions which were varied according to accident type, base line risk, risk reduction and cost. The risk reductions were expressed as mortality risk for adult Swedes. The questionnaire also contained socioeconomic and risk related questions.
We find that
- Women are more likely to take actions in order to reduce their mortality risk.
- Older respondents are more likely to take actions in order to reduce their risk of dying by fire and road accidents, but not drowning.
- Respondents who live in apartment blocks are less likely to take actions to reduce their mortality risk.
- Respondents who have had a previous bad experience with road accidents are more likely to take mortality risk reducing actions.
- Respondents who found the valuation questions easy to answer are more likely to take actions.
Using results from at probit regression model we find the value of statistical life for road accidents to be 20.2 million Swedish kronor (â¬ 2.2 million), 13.3 million kronor (â¬ 1.4 million) for fire accidents, and 12.4 million kronor (â¬ 1.3 million) for drowning accidents. Our results therefore suggest that different values should be used in a cost-benefit framework depending on the accident type.
The value of statistical life for road accidents is within the span of other Swedish estimates where several studies using contingent valuation methods with WTP-questions have arrived at values between 17 and 24 million Swedish kronor (Persson et al, 1996; Persson et al, 2001; Hultkrantz et al, 2006; Svensson, 2007). Further comparison is not possible because we are unaware of any other studies that estimate the value of statistical life using WTP-questions for fire and drowning accidents.
The respondents were also asked whether they believed they could affect the level of risk and also questioned on the extent of their concern regarding the different types of accidents. The responses indicate that while subjects believe that they can affect the risk levels of all three accident types, they believe that they can control the risk for road accidents to a lesser extent for road accidents than for the others. Similarly, although subjects do not worry about the mortality risks to any large extent, there is a clear difference between road accidents and the other causes. It may be the case that these two factors cause the differences in value of statistical life estimates for road accidents on one hand and fire and drowning accidents on the other hand. The results that subjective beliefs of the risk such as dread and voluntariness influences the value of statistical life is in line with suggestions by Slovic (1987), Savage (1993) and Chilton et al (2006)
Place, publisher, year, edition, pages
Research subject Economics
IdentifiersURN: urn:nbn:se:kau:diva-25431OAI: oai:DiVA.org:kau-25431DiVA: diva2:599209
ECHEROMA 2008 HEALTH ECONOMICS AND GLOBAL RENAISSANCE