A 'Natural Experiment' Approach to Investigating Scope Sensitivity in Values for Risk Reductions: Private and Public UV Health Risk Reduction Strategies in Low and High Risk

A 'Natural Experiment' Approach to Investigating Scope Sensitivity in Values for Risk Reductions: Private and Public UV Health Risk Reduction Strategies in Low and High Risk
of 39
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Title: A ‘Natural Experiment’ Approach to Investigating Scope Sensitivity in Values for Risk Reductions: Private and Public UV Health Risk Reduction Strategies in Low and High Risk Countries Authors: Ian J. Bateman 1,2 , Roy Brouwer  1,3 , Stavros Georgiou 1 , Nick Hanley 4 , Fernando Machado 5 , Susana Mourato 6 , Caroline Saunders 7 Affiliations: 1 Centre for Social and Economic Research on the Global Environment (CSERGE), University of East Anglia, Norwich, UK. 2 Centre for Environmental Decision Making and Centre for Economic and Behavioural Analysis of Risk and Decision (CEBARD), University of East Anglia, Norwich, UK. 3 Corresponding author. Current address: National Institute for Integrated Inland Water Management and Wastewater Treatment (RIZA), National Water Policy Division, PO Box 17, 8200 AA, Lelystad, the Netherlands. E-mail: r.brouwer@riza.rws.minvenw.nl. 4 Dept. of Economics, University of Glasgow, Glasgow, UK. 5 Faculdade de Ciências, Economicas e Empresaríais, Universidade Católica Portuguesa, Lisbon, Portugal. 6 Department of Environmental Science and Technology, Imperial College of Science, Technology and Medicine, London, UK. 7 Commerce Division, Lincoln University, Canterbury, New Zealand. Acknowledgements CSERGE is funded by the UK Economic and social Research Council (ESRC). Further funding was  provided by the authors’ host institutions for which we are grateful. The authors also wish to thank the team of interviewers who collected data in the countries surveyed as part of this research. 1    Abstract A common finding amongst studies of willingness to pay for changes in risk exposure is that values are relatively insensitive to the magnitude of that change. One hypothesised reason for this result is that subjects do not comprehend the nature and scope of the risk change  presented to them. This presentation is typically made through some reference to risk  probabilities (conveyed through a variety of devices) with scope sensitivity being tested through variation in those probabilities. An alternative, more ‘natural’ experiment, is to define the sample to include groups of subjects who face substantially different levels of risk. By moving all subjects to a common, readily comprehended risk level (e.g. reducing all risks to zero) sensitivity to risk reduction can be assessed by comparisons across groups. To test this hypothesis we conduct a common-design contingent valuation study across four countries for which the skin cancer risks of exposure to solar UV radiation differ substantially. Survey respondents were presented with both a private and public good route for affecting risk reduction. In both cases results confirm that, once adjustment had been made for expected relationships with other covariates (such as income and risk averting expenditure), valuation responses for both the private and public good conformed to expectations with the ordering of values reflecting the ordering of scientifically established health risks. This suggests that theoretically anticipated links between values and objective health risks may be observed within such situations and provides a justification for continuing research into more natural representations of risk and risk reductions in order to yield consistent and robust measures of risk reduction values. Key words: Health risks, contingent valuation, scope sensitivity, natural experiments, UV radiation. 2  1. Introduction Recent years have seen a number of contingent valuation (CV) and similar stated preference studies examining individuals’ values for changes in risk exposure. A common problem encountered by such studies is an insensitivity of stated values to the magnitude, or ‘scope’, of the change in risk considered (Hammitt and Graham, 1999). Illustrations of such findings are provided by Jones-Lee et al (1993), Baron and Greene (1996), Jones-Lee and Loomes (1997), Beattie et al. (1998) and Carthy et al. (1999). One reaction to insensitivity to the scope of a risk reduction has been to encourage respondents to attend to the quantitative information concerning risk probabilities which is  presented to individuals 1 . However, in testing such an approach, Beattie et al. (1998) report that willingness to pay (WTP) sums increased by just one-third when moving from one risk reduction scheme to another offering a three-times-larger risk reduction. An alternative reaction to such findings of insensitivity to scope is to question the fundamental approach  being used in such exercises. There is ample evidence that many members of the public find certain types of quantitative information, including probabilities, difficult to accurately understand (e.g. Tversky, Sattath and Slovic, 1988). Therefore, instead of attempting to make respondents attend to quantitative information about hypothesised changes in risk, we could consider cases where different groups of individuals are at different, pre-existing levels of real-world risks. By moving all subjects to a common, readily comprehended, risk level (the obvious level being zero risk) we remove much of the need to provide quantitative information. Such ‘natural experiments’ (Rosenzweig and Wolpin, 2000) are only possible within a certain restricted set of applications. However, in the face of such widespread evidence of insensitivity of values to objective risks, it seems a prior and necessary test to examine whether or not consistency can be observed in even the most conducive of circumstances. 1  An alternative approach is to try to simplify the task by breaking it down into more manageable component  parts. Such an approach is implemented by Baron (1996) and Carthy et al. (1999). While this appears to improve value consistency, such methods are highly labour intensive and have the disadvantage that if errors or biases creep into any component question, they are liable to be compounded when the components are ‘chained’ together. 3  One source of risk which conforms to these requirements is that of solar UV radiation. This has proven links to a variety of skin disorders (Slaper et al., 1992), of which skin cancers including malignant melanoma are the most rapidly growing forms of cancer amongst non-elderly or non-infirm populations (Elwood, 1989). Furthermore, death rates vary significantly across countries and ethnic groups. Certain southern hemisphere Australasian countries are at  particularly elevated risk due to ozone depletion. Furthermore, fair skinned populations are  particularly at risk, while darker skinned populations are at relatively lower risk (Harras et al., 1996), a factor which, as we will demonstrate here, is also to some extent reflected in their  behaviour. Here then is a risk for which different groups have different exposure levels. By presenting these groups with scenarios in which risk is reduced to a common and readily comprehended new level (zero risk) we obtain our natural experiment. While we deliberately avoid attempts to explain the actual risk probabilities to which individuals are exposed, we test the hypothesis that values for such risk reductions will nonetheless be consistent with these varying risk levels across different groups of people. In order to capture the variation in risk necessary to our scope test we undertake identical CV surveys in four countries across which objectively measured risk levels vary substantially;  New Zealand, Scotland, England, and Portugal 2 . Respondents are asked their WTP for a new sun protection product, which protects the individual user from all the harmful effects of sun exposure for a given period (a private good). They were also asked their WTP for a different risk reduction scenario, namely reductions in emissions of ozone depleting substances through an increase of general taxation (public good). Awareness, attitudes, perception and  behaviour are measured using conventional measurement scales from socio-psychology (e.g. Torgerson, 1958; Summers, 1970; Nunnally, 1978; Slovic, 1987). The incidence of skin cancers is particularly high in New Zealand, partly as a result of its  predominantly fair-skinned immigrant population being exposed to relatively high levels of 2  This cross-country CV approach to risk valuation has resonance with another recent study conducted by Ready et al. (1999). However, as the subtitle of the latter study indicates, this looked at risks for which all respondents, irrespective of country, faced the same exposure such that consistency across different groups of people was the expectation. 4  UV radiation due to substantial losses of stratospheric ozone (Bentham, 1993). About one thousand new cases of melanoma are registered every year in New Zealand, while over 200  people die annually from the disease (McGee et al., 1995). Together with Australia, these are now the highest incidence and mortality rates in the world (Globocan, 2001). Table 1 illustrates incidence and mortality rates for each of the countries surveyed in our study. Incidence rates are more than four times higher in New Zealand than in the UK and 10 times higher than in Portugal. Separate figures are not given for Scotland and England although epidemiological studies have shown that rates are significantly higher in the former than in the latter country (Langford et al., 1998). The high incidence rates observed in New Zealand are to some extent mitigated by well established health education and treatment  programmes. In New Zealand 12% of the skin cancer cases result in death, compared to over 27% in the UK and Portugal. Nevertheless, the substantially higher incidence rates mean that melanoma mortality rates for the New Zealand population are more than twice those in the UK, which in turn are more than double those in Portugal. Adjusting for the age structure of the countries involved merely acts to increase the disparity between incidence and mortality rates in New Zealand relative to those in Europe (Globocan, 2001). TABLE 1 ABOUT HERE In summary therefore, scientifically established UV related health risks are substantially higher in New Zealand than in the other European countries considered in our study. Within Europe, melanoma incidence and mortality rates are two to three times higher in the UK than in Portugal. As mentioned, an important determinant of this variation is differences in skin tone with darker skins offering greater protection, although sun exposure behaviour also plays a significant part (e.g. Armstrong, 1984; Bentham and Aase, 1996). Taken together these findings map out a continuum of health risks with populations in New Zealand bearing the highest risks followed by Scotland and England, although differences between these countries are relatively small. Finally, Portugal is the lowest risk country in our sample, providing a  base case for our subsequent comparative analyses. Our central research focus is to test whether this pattern in scientifically established health risks is reflected in WTP for risk reductions across these countries. 5
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks