To the extent possible, exposure-response functions, valuations, etc. are based on earlier studies for DETR (for review of the NAQS) and the Department of Health (COMEAP and EAHEAP). Where this is not possible alternative sources have been used, particularly the European Commission’s ExternE Project (European Commission 1995, 1999) which provides a framework for a holistic assessment and for a systematic and thorough consideration of uncertainties across all of the receptors of importance here.

5.4 Characterising uncertainties

As noted above, one of the main difficulties in cost-benefit analysis concerns the fact that, to take a holistic view of benefits, it is necessary to consider effects that can be quantified with limited confidence alongside those that can be quantified well. If all of the economic data on benefits are combined to give a single figure of benefit, it is not possible to gain an appreciation of the extent to which any comparison of cost and benefit may be considered robust.

In previous work a structured sensitivity analysis for air quality benefits appraisal was developed. This sought to retain understanding of the confidence associated with the different types of benefit to be gained from pollution abatement in a transparent manner. The same principle is applied here.

For this study the approach has been adapted in a manner that is more consistent with previous analyses conducted for DETR and the Department of Health, as regards the ranking and grouping of benefits. It has also been extended to provide a more explicit assessment of possible ranges in the overall benefits.

Table 14. Effects of abatement of emissions of SO₂, NO_x, NH₃ and VOCs. An assessment (albeit partially subjective) of the likely importance of each effect following a reduction in emissions is shown in the final column: ‘+++’ benefit from reduction, very important, through to ‘-‘ slight cost from abatement (e.g. through reduced fertilisation of crops with S or N). ‘?’ denotes effects for which particular uncertainty exists, for example concerning the existence of chronic effects of air pollutants on morbidity, or the importance of visibility reductions.

Effect	Importance of effects
Health
Ammonium sulphate and nitrate aerosols acute – mortality chronic – mortality acute – morbidity chronic – morbidity	+++ +++? ++ ++?
Ozone acute – mortality acute – morbidity chronic – morbidity	+++ ++ ?
SO2 acute – mortality acute – morbidity chronic – morbidity	+++ ++ ?
Direct effects of VOCs on mortality and morbidity	++?
Direct effects of NO2 on mortality and morbidity	++?
Altruistic effects of the above health impacts	+++?
Impacts on competitiveness of businesses linked to the above health effects (via changes in rates of absenteeism, demand for pharmaceutical products, consumer demand via changes in mortality rates)	++?
Materials
SO2 / acid effects on utilitarian buildings	++
Effects on cultural assets	+++?
Effects on steel in re-inforced concrete	+
Effects of O3 on paint	no significant effect
Effects of ozone on rubber	+
Macroeconomic effects	++?
Crops
Direct effects of SO2 and O3 on crop yield	++?
Indirect SO2 and O3 effects on livestock	+
N deposition as fertiliser	--
Interactions between pollutants, with pests and pathogens, climate...	--/++
Acidification/liming	+
Macroeconomic effects	++?
Forests
O3 effects on timber production	+?
Non-O3 effects	++?
Non-timber benefits of forests	++?
Exceedence of critical load for eutrophication	++?
Exceedence of critical load for acidification	++?
Other ecosystems
Exceedence of O3 critical level	++
Exceedence of critical load for eutrophication	+?
Exceedence of critical load for acidification	+++
Visibility
Change in amenity	++?
Secondary effects of pollution abatement measures on pollutants not considered under the Directive/Protocol (e.g. greenhouse gases)	--/+++

The benefits are quantified in a series of stages, starting with the aspects that can be quantified with the least uncertainty:

• pollutant exposures;
• assessment of the effects quantified by the Interdepartmental Group on Costs and Benefits (IGCB):
• critical loads exceedence;
• mortality and respiratory hospital admissions linked to short term pollutant exposures;
• damage to building materials;
• change in crop yields.

Analysis to this stage has previously been accepted by UK Government departments, for example in the review of the National Air Quality Strategy, and hence provides the results upon which the greatest stress is placed in this report. However, it excludes assessment of a number of effects that have been reported in the literature. This introduces a likely bias towards underestimation of the effects of air pollution. It also excludes valuation of health effects on the grounds that available data are too uncertain.

The application of uncertain data clearly provides the threat that the results derived using them could be misleading. However, the exclusion of effects, particularly those for which there is a strong logic(1) for inclusion, also provides a threat that results will be misleading; it could be said to imply an unwarranted degree of assurance that present conditions are acceptable; that the benefits of pollution abatement are less than they really are. With this in mind the analysis has been extended to a number of additional aspects, quantified in a sensitivity analysis that follows assessment of the issues identified above. These aspects, in order of increasing uncertainty, are as follows:

• assessment of additional health effects for which COMEAP provided exposure-response functions, but did not use them in its quantification exercise;
• application of the ranges in valuation data identified by EAHEAP. This monetisation will undoubtedly be controversial given the previous conclusion of Department of Health Ministers that available health valuation data were not robust enough to be applied in support of policy (DETR, 1999);
• application of additional exposure-response functions reported in the literature, but for which there are outstanding issues of interpretation and greater concern over the accuracy of predictions made.

In addition to this structured approach, individual elements of uncertainty are discussed as they arise in the next section and in the Appendices.

(1) For example, if one accepts, as COMEAP have done, that air pollution is causally linked to mortality and respiratory hospital admissions, it may well seem likely that it is also linked to less severe effects, such as increased prevalence of the symptoms of asthma.

Chapter 4          Chapter 6

---

Report and site prepared by the National Environmental Technology Centre, part of AEA Technology, on behalf of the UK Department of the Environment, Transport and the Regions