References
Air Resources Board (2004). Particulate air pollution and infant mortality. California Environmental Protection Agency ftp://ftp.arb.ca.gov/carbis/research/health/healthup/may04.pdf
Barnett, A. G., Williams, G. M., Schwartz, J., Neller, A. H., Best, T. L., Petroeschevsky, A. L. and Simpson, R. W. (2005). Air pollution and child respiratory health: A case-crossover study in Australia and New Zealand. American Journal of Respiratory and Critical Care Medicine, 171(11), 1272-1278.
Beacon Pathway Ltd (2008) Thermal Insulation in New Zealand Homes – A Status Report, Beacon Pathway Ltd, Auckland
BRANZ (2006) Energy End Use in New Zealand Households – Report on Year 10 Analysis for the Household Energy End-use Project (HEEP), Study Report 155, Building Research Association of New Zealand, Wellington.
Canadian Medical Association (2008). No Breathing Room: National Illness costs of Air Pollution. Ottawa.
Clinton, S., Jones-Lee, M., Metcalf, H., Loomes, G., Robinson, A., Covey, J., Spencer, A. and Spackman, M. (2007). Valuation of health and safety benefits: Dread risks, Health and Safety Executive Research Report RR541, page 95.
Cohen, A.J., Anderson, H. R., Ostro, B., Pandey, K. V., Krzyzanowski, M., Künzli, N., Gutschmidt, K., Pope III, C. A., Romieu, I., Samet, J. M. And Smith, K. R. (2004). Urban air pollution, in Ezzati, M., Lopez, A. D., Rodgers, A and Murray, C. J. L. (ed). Global and Regional Burden of Disease Attributable to Selected Major Risk Factors: Volume 1. Geneva. World Health Organization.
Fisher, G., Rolfe, K.A., Kjellstrom, T., Woodward, A., Hales, S., Sturman, A.P., Kingham, S., Petersen, J., Shrestha, R., King, D. (2002). Health effects due to motor vehicle pollution in New Zealand: Report to the Ministry of Transport. Ministry of Transport, Wellington, New Zealand.
Fisher, G., Kjellstrom, T., Kingham, S., Hales, S. and Shrestha, R. (2007). Health and Air Pollution in New Zealand: MAIN REPORT. A Research Project Funded by: Health Research Council of New Zealand Ministry for the Environment, Ministry of Transport.
Glinianaia, S. V., Rankin, J., Bell, R., Pless-Mulloli, T and Howel, D. (2004). Does Particulate Air Pollution Contribute to Infant Death? A Systematic Review. Environmental Health Perspectives, 112(14), 1365-1370.
Guria, J., Jones, W., Jones-Lee, M., Keall, M., Leung, J. and Loomes, G. (2003) The Value of Statistical Life and Prevention of Injuries in New Zealand. Wellington: Land Transport Safety Authority (unpublished draft report).
Ha, E. H,, Lee, J.T., Kim, H., Hong, Y. C,, Lee, B. E,, Park, H. S, and Christiani, D. C. (2003). Infant susceptibility of mortality to air pollution in Seoul, South Korea. Pediatrics 111(2), 284–290.
Jalaludin, B. B., O’Toole, B. I. and Leader, S. R. (2004). Acute effects of urban ambient air pollution on respiratory symptoms, asthma medication use, and doctor visits for asthma in a cohort of Australian children. Environmental Research, 95(1), 32-42.
Kappos, A. D., Bruckmann, P., Eikmann, T., Englert, N., Heinrich, U., Höppe, P., Koch, E., Krause, G. H. M., Kreyling, W. G., Rauchfuss, K., Rombout, P., Schulz-Klemp, V., Thiel, W. R. and Wichmann, H. E. (2004). Health effects of particles in ambient air. International Journal of Hygiene and Environmental Health, 207(4), 399-407.
Künzli, N.; Kaiser, R., Medina, S., Studnicka, M., Channel, O., Fillinger, P., Henry, M., Horak, F., Puybonnieux-Texier, V., Quenel, P., Schneider, J., Seethaler, R., Vergnaud, J-C., and Sommer, H. (2000). Public-health impact of outdoor and traffic-related air pollution: a European assessment. The Lancet, 356, 9232, 795-801.
Lloyd R (2006) “Fuel Poverty in New Zealand”; University of Otago
Miller, T. and Guria, J. (1991) The Value of Statistical Life in New Zealand, Ministry of Transport, Wellington.
Ministry for the Environment (2004). Proposed National Environmental standards for Air Quality, Resource Management Act 32: Analysis of the costs and benefits. Wellington. Ministry for the Environment.
Ministry for the Environment (2005a) Warm Homes Technical Report: Detailed Study of Heating Options in New Zealand – Phase 1 Report; Wellington.
Ministry for the Environment (2005b) Warm Homes Technical Report: Social Drivers – Phase 1 Interim Report; Wellington.
Ministry for the Environment (2008) Report on Progress: National Environmental Standards for Air Quality – Volume 1 Main Report; Wellington.
Ministry of Health (2009). Mortality: 2005 (final) and 2006-2008 (provisional). http://www.nzhis.govt.nz/moh.nsf/pagesns/530
Ministry of Transport (2008). The Social Cost of Road Crashes and Injuries June 2008 update. http://www.transport.govt.nz/research/Documents/Social%20Cost%20June%202008%20update%20(final).pdf
PA Consultants (2001) “Valuing New Zealand’s Clean Green Image“; Ministry for the Environment, Wellington
Röösli, M., Künzli, N., Braun-Fahrländer, C. and Egger1, M. (2005). Years of life lost attributable to air pollution in Switzerland: dynamic exposure–response model. International Journal of Epidemiology 34, 1029–1035.
Rowlatt, P., M.Spackman, S.Jones, M.Jones-Lee, and G.Loomes (1998). Valuation of Deaths from Air Pollution, A Report for Department of Environment, Transport and the Regions, and Department for Trade and Industry.
Sommer H. 1999. Health Costs Due to Road Traffic-Related Air Pollution: Technical report on economy. Report prepared for the WHO Ministerial Conference on Environment and Health, London June 1999, Federal Department of Environment Transport, Energy and Communications, Bureau for Transport Studies, Bern.
Tailor Baines & Associates (2006) The impact on housing energy efficiency of market prices, incentives and regulatory requirements, Report to CHRANZ, Wellington
Wilton, E. (2003). Assessing the benefits of introducing standards for air. Ministry for the Environment.
Woodruff, T. J., Grillo, J. and Schoendorf, K. C. (1997). The relationship between selected causes of postneonatal infant mortality and particulate air pollution in the United States. Environmental Health Perspective, 105(6), 608-612.
Table 16 summarises the differences between the initial and updated cost benefit analyses.
|
Table 16 Coverage of initial and updated analyses Read a description of this image Text description goes here. |
The initial analysis had one principal benefit, the social willingness to pay value attached to reduction in premature death and hospitalisation (VOSL) and one much smaller benefit of avoided losses in GDP (direct and indirect, arising from deaths, hospitalisation and restricted activity days). These benefits understate the societal value of reductions in deaths and hospitalisations because they omit long term deterioration in quality of life and also omit the cost of medical treatment. The updated analysis includes the willingness to pay value (which includes income loss) and also medical treatments. It acknowledges but does not value long term quality of life and the indirect GDP benefits, which are small relative to VOSL.
The initial analysis had costs associated with standards for ambient air quality, prohibited activities and wood burners. Most of these are updated in the new analysis, except costs to schools and hospitals, which are no longer incurred, and the subsidy cost for wood burners. The new analysis values potential impacts on households for compliance with the wood burner standard. Potential for business forgone by consenting constraints cannot be valued without localised modelling.
We discuss here the major comments and Our response to those comments. The comments refer to estimation of benefits in terms of physical consequences and their valuations and estimation of costs.
Comment: The effects of are not just the effect of short term exposure but the effect of cumulative exposure to PM10 of over time. Hence, the benefit of reducing PM10 levels should be measured by improvement in life expectancy.
Response: There is merit in the argument that the health effects are the total effect of exposure over time. The standard is expected to keep the PM10 emission below a certain level. That should reduce the cumulative effect of the exposure over time. The estimates of health effects used in the report is based on earlier studies which used a methodology developed by Künzli et al (2000), which was based on cohort observations. The health effects were estimated as functions of annual average 24 hour maximum PM10 levels. The fatalities, as we understand, were linked to each year’s level. It does not necessarily give the cumulative effect. That part needs a separate study.
We have discussed in detail the drawbacks of using the difference in life expectancy in .
Comments: The estimate of VOSL used is based on risk reduction in traffic crashes. There are a few differences between this scenario and PM10 effects.
- There is a considerable time lag between exposure and time of death;
- The risk profile may not be the same and that may affect the VOSL for PM10 effects.Response: It is possible that the VOSL used is on the lower side because of the longer time difference between the event and death, as people suffer more during this time (see for more details).True, the risk profile may be different. The current VOSL is an update of the value established in 1991 based on a survey carried out in 189/90. At that time annual road toll was varying between 650 an 750. We have not got the estimate of deaths from pollution during this time period. The estimate for 2003 (872) is higher. If the pollution related death was similar during 1989/91, then the base risk would be lower and that would possibly have some effect but not large enough to substantially affect the cost benefit relativity analysed in the paper.
Comments: There is a likelihood that the number of deaths are under estimated in the 2004 CBA and in this report as uses the same estimates under the status quo scenario, because all urban areas were not covered in the 2004 estimates.
Response: If higher levels of estimated number of deaths were used, the benefit cost ratio and the net benefit would be higher.
Comments: Some tables need better explanation.
Response: We have explained them further and also revised the table presentations for better understanding.
The value of statistical life (VOSL) used in the analysis is based on that used in transport project appraisals in New Zealand, which was derived from a willingness to pay (WTP) survey carried out in 1989/90 (Miller and Guria 1991), and updated since then by indexing it to the ordinary time wage rate (Ministry of Transport 2008).
Before 1991, project appraisals used value of life based on the so called “human capital approach”, in which the value of a premature fatality avoided (life saved) is the present value of expected output forgone over the rest of life of the person. The WTP based VOSL is fundamentally different in approach, in being derived from the amount of money people are willing to pay to reduce their risk of death by a certain proportion. The WTP surveys probe in various ways how people change their behaviour in face of risk. It is very unlikely that respondents think about the number of life years left them or adjust their willingness to pay accordingly: not only do they not know when they would otherwise die but also they have to survive today to enjoy any more life beyond then. Surveys carried out to estimate this value lead people to think about the level of risk they face and the scope of reducing the risk and how much they would be willing to pay based on what they can afford, but there is no evidence their answers are framed in terms of a value per life year.
The value of statistical life does not purport to place a value on any identified, discreet individual’s life. When subjecting a project to cost benefit analysis, the safety benefit is expressed as a reduction in the risk of accidental injury or death. A reduction in risk of death of 1 in 10,000 is equivalent to saying that one more person will be saved out of a group of 10,000 people. In that sense only is the VOSL used as an expression of the value of a risk reduction that is expected to save one life – the value of a statistical life saved.
Some researchers derive a value of life per year from the WTP based VOSL, in which they assume that the VOSL is the discounted present value of values of life years lost if the person died at that point. In our view it is a wrong approach since the VOSL is based on the value of risk reduction now, not securing years of life ahead. For similar reasons, characterizing the benefit of air quality improvement as an extension of life years before death at some distant point in the future will understate the value of risk reduction now: people do not know in advance when they are going to die with current air pollution and when with pollution reduced, so any expressed willingness to pay is for a generalized reduction in risk that is being incurred now.
The formula of Kunzli et al (2000), which is used to estimate the number of deaths in this study, is based on observed numbers of deaths in the 30 years and over age group. Many of these will not have died immediately after being exposed to high level of pollution, rather they endure respiratory conditions, heart disease and cancer for many years before death. The value of pain and suffering for them would be considerably higher than for those facing risk of instantaneous death. Studies find that people are more willing to pay to save a person from cancer death than a pedestrian death (Rowlatt et al 1998, Clinton et al 2007).
The social cost of pollution is the total loss to society if one person dies as a result of exposure to pollution, so the benefit of a pollution reduction policy is estimated by the expected number of lives saved as a result. Some of these lives would have been lost immediately after exposure (comparable to pedestrian death) but in many cases they would die after years of suffering. Compared to road crashes, in which only deaths which occur within 30 days of a crash are counted as road crash deaths, the loss to society of the average premature death from pollution is likely to be greater than that from the average road fatality.
Taking these factors into account, using the VOSL is more likely to under-state than over-state the cost to society of lives lost as a result of exposure to pollution.
A reviewer has questioned the relevance of using the value of statistical life (VOSL) derived from transport accident studies for application to mortality changes from air quality improvements, on grounds that such “benefit transfers” (i.e. using benefit estimates from one situation in analogous situations elsewhere) are subject to large discrepancies unless the populations at risk, and the risk profiles they face, are reasonably similar. While issue-specific valuations would be preferable, the survey-based studies to obtain them do not come cheap and no alternative value is currently available. The willingness to pay based value of statistical life is estimated from realistic risk and risk reduction, and respondents’ willingness to pay to gain that risk reduction. There is no particular reason what the value in present context should be drastically different, other than the likely higher level of pain and suffering before death. In that case, the VOSL should be higher not lower.
The transport-related VOSL is the only such estimate used in official estimates in New Zealand, and it is widely used with due caution as a practical guide in situations removed from its origins in road transport, such as assessments of aviation and maritime safety. The international literature suggests that willingness to pay to reduce risk is lower in road transport than in situations where respondents feel they have less individual control – for instance in public transport, or when exposed to involuntary risk. For this and other reasons the current transport related VOSL is more likely than not to underestimate the public benefit of mortality reduction in these other settings.
In the case of air quality, the risk from exposure is largely involuntary and out of control of individuals, and may also result in long lingering illness rather than sudden death. Studies have found that people consider deaths from cancer or heart disease worse than pedestrian deaths which occur in most cases within a short time from the crash, and a similar aversion to chronic illnesses brought on by air pollution is likely to increase the public willingness to pay to reduce risk of exposure.
Only a fraction of total pollution-related deaths occur in the first year after exposure and the rest occur in subsequent years. Some argue that the Value of Statistical Life (VOSL) should be discounted to take into account the delay in death.
This argument is based on only part of the total effect. The VOSL is estimated from the amount people are willing to pay to reduce the risk of death. In transport crashes, the number of deaths is based on those dying within 30 days from the date of accident. There can be more premature deaths due to crashes but they occur after 30 days. In some cases, people potentially involved in them and society in general would be willing to pay more to reduce the risk of such deaths, because of the pain, grief and suffering they cause to the people killed prematurely and those close to them.
In case of pollution, people who die in the second year, third year or a later date suffer more than those who die within a short time from the time of exposure. Some of these are due to being affected by cancer, ischaemic heart disease and cerebro-vascular disease. In these cases, people suffer longer before death than do those involved in a traffic crash. If the value of this extra suffering were added, the VOSL for such cases would be much higher. This may be a reason why the EPA in the USA uses much higher VOSL than that used by the Department of Transport.
One of the peer reviewers suggested that only the change in life expectancy should be considered. This would drastically reduce the value of reducing the risk of pollution effects. It is undoubtedly the wrong approach.
- First the willingness to pay based VOSL does not proportionately vary with remaining life expectancy. The remaining life expectancy reduces with age. The two past studies in New Zealand did not find any relationship with age. Some European studies found an inverted U shaped relationship, i.e., increasing with age up to a certain age and then slowly reducing.
- Calculating the value of a life year by dividing VOSL by the life years forgone by the average premature fatality means the value of a life year is the same for all: this approach would indicate very low value of remaining life for the elderly and very high value for children. Although some people may think such differences in relative values are appropriate, empirical evidence on the amount people are willing to pay to reduce their risk of death does not give any such indication.
Consequently it is inappropriate to conceptualise the benefit of air quality improvements as adding life years at some point in the distant future, or of manipulating the VOSL as if that is what it shows. The VOSL reflects individuals’ willingness to pay to reduce risk as a guide to society’s willingness to pay to reduce risk through collective action that reduces that risk.
The discussion suggests that the use of VOSL estimated from traffic risk based studies does not over estimate the social cost of air pollution effects. In fact there is a likelihood that it underestimates the social cost.
Comparison of initial and updated cost benefit analysesResponses to peer review commentsValue of loss of life due to pollution