The Green Book

VALUING NON-MARKET IMPACTS	ANNEX
INTRODUCTION

The valuation of non-market impacts is a challenging but important element of appraisal, and should be attempted wherever feasible. This Annex outlines techniques on how to value non-market impacts, and some typical applications such as time-savings, health benefits, prevented fatality, design quality, and the environment. These approaches can be complex but are equally as important as market impacts.

VALUING NON-MARKET IMPACTS

Where market values are not available for an identified cost or benefit, there are a number of approaches to attributing a value for inclusion in an appraisal, the most commonly used of which are outlined below.

Willingness to Pay and Willingness to Accept

The preferred method of valuation is to simulate the market by estimating the ‘willingness to pay’ (WTP) or ‘willingness to accept’ (WTA) a project’s outputs or outcomes. Willingness to pay for a little more of a service is a reflection of the value placed by consumers on an increment of that service. The amount consumers are willing to pay depends to a large extent on the levels of income available to them, so valuations are usually obtained by averaging across income groups.

The quantification of potential social, health or environmental impacts normally requires an alternative approach to valuation. Techniques to establish money values for this type of non-market impact generally involve the inference of a price, through either a revealed preference or stated preference approach.

Revealed preference techniques involve inferring an implicit price revealed indirectly by examining consumers’ behaviour in a similar or related market. Hedonic pricing is an example of this approach.¹ For example, the relationship between house prices and levels of environmental amenity, such as peace and quiet, may be analysed in order to assign a monetary value to the environmental benefit. Other examples are travel cost models (for recreational values) and random utility models (to value individual features of a site).

Stated preferences are normally obtained by specially constructed questionnaires and interviews designed to elicit estimates of the willingness to pay (WTP) for, or willingness to accept (WTA), a particular outcome.² When using stated preferences the main choice is between contingent valuation and choice modelling (CM). Contingent valuation studies elicit WTP or WTA via direct questions such as ‘What is the maximum amount you would be prepared to pay every year to receive good x?’ (the ‘open-ended’ format) or ‘Which of the amounts listed below best describes your maximum willingness to pay every year to receive good x?’ (the ‘payment card’ format). CM studies, on the other hand, elicit values by presenting respondents with a series of alternatives and then asking which is most preferred.³

The technique chosen will depend on the individual circumstances, and should be judged on a case-by-case basis. As a general rule, revealed preference methods are fairly reliable, and should be used where the relevant information can be inferred. However, they cannot estimate the value placed on an asset by people who make no direct use of it. In these circumstances, stated preference methods may be helpful. In some cases, it will be appropriate to use both techniques together, for example, to check the consistency of results.

back to top

Other approaches

When faced with a mix of both monetary values and quantified data (and probably some unquantified considerations as well), weighting and scoring can be used to bring data expressed in different units into the appraisal process. Using this technique, options can be ranked and the preferred option identified. This approach usually involves an explicit relative weighting system for the different criteria relevant to the decision.⁴ It often involves an implicit monetisation of different impacts, once the performance against the various criteria is compared to the costs considered worth spending to secure or to avoid them.

Where a direct assessment of the value of a benefit or cost is particularly uncertain, reference can be made to the costs of preventing the loss of, or replacing, a non-marketed good (such as a natural habitat or recreational facility). This does not provide a measure of its value but can provide a figure to focus discussion upon whether the good is worth as much as this expenditure.

In the absence of an existing reliable and accurate monetary valuation of an impact, a decision must be made whether to commission a study, and if so, how much resource to devote to the exercise. Key considerations that may govern a decision to commission research are:

	Tractability of the valuation problem: whether research is likely to yield a robust valuation;
	Range of application of the results of a study to future appraisals;
	How material the accuracy of the valuation is to the decision at hand. This may be gauged through sensitivity analysis around a range of plausible estimates; and,
	Scale of impact of the decision at hand. If the decision relates to a multi-billion pound programme or to regulation that will impose costs of similar scale upon industry, it is clearly worth devoting much more resource to ensuring that the valuations of the non-market benefits (and costs) are accurate than would be appropriate for a smaller scheme.

It is often difficult to assess the reliability of estimates emerging from a single study using a single method. Valuations may be unreliable because responses to questionnaires may be inconsistent or biased, or because valuations may take insufficient account of budget constraints. Estimates can be given more credence if different methods, or studies by different researchers, give similar results.

When using any technique, it is advisable to provide a range of values, and to subject the estimated values to a plausibility check with decision makers. The minimum or maximum valuation of a benefit or cost that would support a particular decision (‘switching value’) should be made explicit, compared with the real or implied valuations derived from previous decisions, and qualified by a statement of the robustness of the valuation techniques employed.

back to top

CURRENT RESEARCH / PLAUSIBLE ESTIMATES

Following are some areas where research has been undertaken to derive plausible estimates for particular non-market costs and benefits.

Valuing Time

Within central government, the Department for Transport’s (DfT) approach to valuing time in the appraisal of road schemes and other projects is well established.⁵ This approach uses different values for ‘employers’ time and ‘own’ time (or working and non-working time).

The value of employees’ time-savings (working) is the opportunity cost of the time to the employer. This will be equal at the margin to the cost of labour to the employer: the gross wage rate plus non-wage labour costs such as national insurance, pensions and other costs that vary with hours worked.⁶

The values for working time used in the appraisal and modelling of transport projects and policies, are based on the mileage weighted labour costs of users of each mode of transport. The National Travel Survey (NTS) contains detailed information on the distance and amount of time spent in travel by individuals in each earnings band to provide the appropriate weights for each mode of transport. The New Earnings Survey provides estimates of the earnings of drivers of commercial and public service vehicles. In theory, it is possible to collect data on the earnings of those who would use the project being appraised, although this is rarely practical.

It is accepted practice to use a national average standard value of non-working time (equity value of time-savings) for all modes of transport for appraisal purposes. The use of a project-specific value of non-working time might be preferable in cases where time-savings can be captured through revenue from fares. These will often form part of a commercial decision by, for example, a train operator assessing the case for accelerating a service.

For transport appraisals, journeys to and from work are included in non-working time. The value of savings in travel time for work is assumed to rise at roughly half the rate of real income.⁷ For non-work time, this assumption balances a number of factors that might either tend to increase or decrease the value of time-savings relative to income. These might include a decline in the marginal utility of money as incomes increase, changes in the length of the working week and changes in the quality of travelling conditions.

Some additional considerations when valuing time-savings include:

	People place a higher value on saving walking or waiting time than on saving time spent in a vehicle. Evidence suggests that walking and waiting time should be valued at double that used for in-vehicle time.⁸
	Time spent in overcrowded conditions on public transport also carries a higher weight, the value being determined by the severity of the overcrowding.
	Unreliability, measured in terms of deviations around the expected journey time, can also carry an additional penalty.
	Time-savings should be valued at the same rate per minute, whatever the extent of the saving or duration of the journey.

Using the estimated average values of travel time-savings from previous projects or proposals may not be appropriate if the characteristics of the client group are not similar to those of transport users, or if the circumstances differ significantly. Nevertheless, the estimates may serve as orders of magnitude.

back to top

Valuing Health Benefits

Health impacts are rarely a question simply of lives lost or saved. In policy areas that affect mainly health, an alternative approach is often used, to take account of changes in life expectancy (including expected life years where lives are lost or saved), and changes in the quality of life. This approach is known as the quality-adjusted life year (QALY).

The EuroQol instrument provides a simple and consistent framework for measuring general health and deriving QALY values and is the most commonly used measure of health benefits in Europe. It weights life expectancy for health-related quality of life over time.

The comparison of health interventions may reveal the impact of different factors on clinical effects. For example, working out the relationship between dosage and response of a particular medicine is a necessary prior step to properly valuing a policy for the provision of that medicine. In some cases, such as when the benefits of an intervention are measured in ‘natural’ units (e.g. reduced incidence of a disease or lower blood pressure rates), it may be appropriate to undertake an appraisal on the basis of its cost effectiveness.⁹

It is difficult to determine whether a health programme should be funded, or how large it should be, without first allocating a monetary value to the projected health gains. Valuation is also important when health impacts are to be weighed against non-health impacts. There are a number of techniques available, including undertaking a survey to estimate an individual’s WTP for certain health benefits.¹⁰ Once WTP is known, appraisers can compare the marginal benefits of an intervention against its marginal costs.

An example of a broad approach to estimating acute health impacts is set out in Box 2.1.¹¹

BOX 2.1: MEASURING SHORT TERM HEALTH BENEFITS ASSOCIATED WITH REDUCTIONS IN AIR POLLUTION¹²

A FIVE-STEP APPROACH TO VALUING HEALTH IMPACTS

Estimate the annual average concentration of pollutants and resident population in each 1km grid square of the country.

Assign the baseline level of the given health-related and pollution affected events to each grid square e.g., daily deaths, hospital admissions for the treatment of respiratory diseases.

Combine the data from (1) and (2) and apply a dose-response function linking pollutant concentrations with the relevant effects. Dose-response functions are expressed as a percentage increase in the baseline rate of health outcome per unit concentration of pollutant. Three outputs can be derived:

3.1	The current effect on health of the relevant pollutant per grid square
3.2	The benefit to health per grid square produced by the fall in concentrations of air pollutants expected to occur
3.3	The benefit to health produced by reducing the concentration of pollutants in each grid square, in accordance with the proposed policies which aim to meet the objectives.

Sum the results obtained in (3) to estimate the total reduction in the number of cases of each health effect (which has an accepted dose-response function) associated with meeting or approaching the objectives.

Apply monetary values for each health effect to transform quantitative estimates into monetary estimates.

back to top

The Value of a Prevented Fatality or Prevented Injury

A benefit of some proposals is the prevention of fatalities or injuries. The appropriate starting point for valuing these benefits is to measure the individual’s WTP for a reduction in risk of death (or their willingness to accept a new hazard and the ensuing increased risk).

The willingness of an individual to pay for small changes in their own or their household’s risk of loss of life or injury can be used to infer the value of a prevented fatality (VPF). The changes in the probabilities of premature death or of serious injury used in such WTP studies are generally very small.¹³

In the UK, the main measure of VPF incorporates the ‘extra’ value placed on relatives and friends, and any further value placed by society on avoiding the premature death of individuals. Accordingly, the addition of an individual’s WTP for the safety of others to his ‘own’ WTP for ‘own’ safety may lead to double counting.¹⁴

A lower bound on the value of a prevented fatality may be determined by revealed preference and stated preference studies. This lower bound is useful for determining a threshold of value for money for safety expenditure and also for comparing proposals concerning increased safety.

Revealed preference studies can derive individual WTP for risk reduction from, for example, the size of wage differentials for more or less risky occupations; or price versus safety trade-offs in choosing transport modes; or WTP for safety devices such as smoke alarms or car air bags. However, in practice, these estimates of the revealed value of a prevented fatality are not precise. Stated preference approaches have also been used to provide estimates of VPF using questionnaires.¹⁵

In the UK, the Department for Transport (DfT) values the reduction of the risk of death in the context of road transport at about £1.145m per fatal casualty prevented (in 2000 prices).¹⁶ In addition to the WTP measures, these estimates include gross lost output, medical and ambulance costs.Values are uprated in line with assumed changes in GDP per head.

DfT also attributes monetary values to the prevention of non-fatal casualties, based on a WTP approach. Serious and slight casualties are valued separately and the values are uprated in line with changes in GDP per head. Values currently in use for preventing a serious and slight road injury are £128,650 and £9,920 respectively (at 2000 prices).¹⁷ Costs of police, insurance and property damage are added to these casualty values to obtain values for the prevention of road accidents. The HSE tariff of monetary values for pain, grief and suffering begins at £150 for the most minor non-reportable injury.¹⁸

There is evidence that individuals are not indifferent to the cause and circumstances of injury or fatality. For example, in their estimate of benefits from asbestos proposals, HSE currently doubles the VPF figure to allow for individual aversion to dying from cancer, and the additional associated personal and medical costs.¹⁹

back to top

Valuing Design Quality

Design quality is an important element of all public sector building projects and should be assessed during appraisal. Limiting property valuation to traditional methods without consideration of the costs and benefits of design investment can distort the decision making process. Good design will not always result in the lowest initial capital cost. However, over the period of the contract a higher initial investment can, when expressed as a discount value, result in the lower whole life costs.

The benefits of good design include:

	Simplification and savings in cost, by ensuring that capital costs are competitive and that savings can be achieved on running costs;
	Increased output and quality of service through enhancement of the environment in which a service is provided; and
	Staff recruitment and retention.

Where good design has a direct economic impact, such as staff retention or patient recovery times, it may be possible to calculate the costs and benefits directly. However, it is often difficult, if not impossible, to calculate the monetary value of many of the benefits of good design, such as civic pride, educational achievement or user experience. In such instances, it may be necessary to use contingent valuation or a similar technique. For smaller projects, where contingent valuation may prove too complicated, research studies can help with comparisons and benchmarking to ensure good design is accounted for.

	DETAILED GUIDANCE ON EVALUATING AND DELIVERING DESIGN QUALITY CAN BE FOUND IN:
		The Value of Good Design, CABE
		Achieving Well Designed Schools Through PFI, CABE
		Better Civic Buildings and Space, CABE
		Treasury Guidance Note 7: How to Achieve Design Quality in PFI projects
		Improving Standards of Design in the Procurement of Public Buildings, CABE/OGC
		The CABE website (http://www.cabe.org.uk)

VALUING ENVIRONMENTAL IMPACTS

The valuation of environmental costs and benefits is constantly evolving, with new research continually being funded by the UK government and its agencies. Research covers both methodological development and the estimation of values. There are a number of valuable reference sources that discuss valuation issues in depth.²⁰ The following paragraphs provide information on government research and guidance on the quantification and monetisation of impacts, including which departments are sponsoring research. As this is a developing field, policy makers are encouraged to refer to the Green Book homepage, in order to locate the most up to date information.

Impacts of policies and measures on greenhouse gas emissions

Current methodologies for assessment of the effects of policies and measures on greenhouse gas emissions are policy specific with no standard guidance available. There are some models available that can be used to assess the effects of particular types of proposals on emissions (e.g. National Road Traffic Forecasts).²¹

The impact of a new policy, project or programme on emissions should be expressed in terms of carbon savings, or in terms of additional emissions, measured in million tonnes of carbon-dioxide equivalent (MtC02).

In cases where quantification of the climate change effect is impractical, an assessment of whether the policy is likely to increase or decrease emissions, combined with a qualitative assessment of the significance of this change, should be included in the appraisal.

Once the emissions impact of a proposal has been quantified, current research informs the calculation of illustrative values for the social damage cost of carbon.²² This can then be used to estimate the monetary value of the impacts.

back to top

Assessing vulnerability to the impacts of climate change

In 1997, the UK government established the UK Climate Impacts Programme (UKCIP) to help public and private organisations assess their vulnerability to climate change. UKCIP, together with Defra, can provide the latest information on climate change predictions and assessments. This includes guidance on how to identify and assess the risks and uncertainties posed by a changing climate, and a methodology for costing the impacts of climate change.

Key policy areas where climate change might be a particularly important consideration include: investment appraisal for long-term planning and infrastructure projects, regulatory and planning frameworks, contingency planning and long-term policy frameworks.

Air Quality

Assessing the impact of particular policies on air quality is a complex science. Sophisticated modelling tools exist to forecast emissions from different sources and estimate the impact on ambient concentration levels of different pollutants at different locations.²³ Government departments and agencies may need to consider air quality impacts in the design of their policies. For example, the Highways Agency’s Design Manual for Roads and Bridges can be used to forecast the impact of new or existing road schemes on emissions of key pollutants from road transport.

Impacts on air quality are generally expressed in terms of either the total volume change in emissions of a particular pollutant from a particular source; the likely impact of this change on levels of ambient air quality in the affected area; or the total number of households likely to be affected by these changes.

In cases where detailed modelling is not possible, a reasoned statement of whether or not a particular policy is likely to result in greater or lesser emissions of particular pollutants should be included in the appraisal.

Research has been funded to develop a methodology for quantifying and monetising, where appropriate, the health and environmental impacts of air quality changes.²⁴

Landscape

Landscape includes townscape, heritage, and other related matters. Guidelines for assessing the impact of policies, projects and programmes on landscape have been devised by English Heritage and the Countryside Commission.²⁵ The Commission for Architecture and the Built Environment (CABE) may also be able to provide guidance.²⁶

Research has also been commissioned Defra to estimate the value of environmental landscape features associated with agri-environment schemes. Contingent valuation techniques have been used, producing an Environmental Landscape Features (ELF) model. This constitutes a first attempt at a benefits transfer tool for appraising agrienvironment policy.²⁷ Features covered include heather moorland, rough grazing, field margins and hedgerows. The model provides estimates of WTP for these features on an area basis, and estimates of their diminishing marginal utility.

back to top

Water

It is not easy to derive economic values for damage costs of water pollutants. The complexity of the way in which pollutants entering the water environment affect chemical water quality and ecological status means that it is difficult to devise simple dose-response functions. Furthermore, there are several ways in which the benefits of improving water quality are location-dependent and it is not easy to determine the relevant population to use for grossing up values, or how to take account of decay functions to represent 'distance decay'.²⁸ Therefore, water valuation studies do not generally produce ‘marginal damage cost’ estimates for specific pollutants; they are more geared towards producing values for observable changes in environmental quality.

Numerous studies have attempted to estimate the economic value of changes to water quality or flow rates/levels in water bodies,²⁹ but establishing values that can be transferred is difficult. New research is planned by Defra, the Environment Agency and Ofwat to value the environmental benefits of changes in water quality.

Biodiversity

The benefits of biodiversity can be difficult to measure, define and value. However, if these benefits are disregarded or given a low priority in appraisal work, there is a risk of excessive and potentially irreversible degradation of natural resource stocks.

Defra and the Forestry Commission fund research on the valuation of biodiversity that is concerned both with developing methodological approaches and deriving empirical estimates.³⁰

Noise

Assessing the impact of noise can be complex, not least because of the subjective nature of many of its effects. Despite this, a number of approaches to quantifying the impact of changes in noise according to the source, the scale and nature of the proposals have been developed. For example, the impact of new transport infrastructure or industrial developments can be quantified according to the number of people/households affected by an increase or decrease of noise levels measured in average decibels (dB(A)). This approach can also be used to assess the impact of changes to traffic control measures.

This is a rapidly developing area and studies are being taken forward to obtain monetary values for noise.³¹ Recent studies across Europe have yielded a range of values, many of which lie in the range of

20 - 30 per household per decibel per year. The median value from those studies is

23.5 per household per decibel per year (2001 prices).³²

Recreational and amenity values for forests

In 1992, the Forestry Commission established a value for recreational visitors to forests of £1 per visit. More recent work on the recreational value of forests in Northern Ireland suggested that mean willingness-to-pay (WTP) varies between £0.60 and £1.74 per visit, depending upon the location of the forest, its attributes and socio-economic characteristics of the visitors.³³ If a high level of accuracy is required, recreational values need to be more sensitive to the attributes of individual forests, the location and availability of substitutes, and the characteristics of the visitors in the catchment area. However if a broader estimate is sufficient, the 1992 value (£1 per visit) indexed to the year of the appraisal should suffice.

The Forestry Commission commissioned a further study to estimate the range of non-market benefits associated with forestry. This reviewed existing methodologies and research to determine the best approach to valuing the non-market benefits of UK forestry and made recommendations on non-market values for recreation, landscape, amenity, biodiversity and carbon sequestration.³⁴

back to top

Valuing disamenity

Activities including the transport and disposal of waste and the quarrying of minerals and aggregates give rise to a range of undesirable impacts that can undermine public enjoyment of an area. A number of studies have attempted to value these, which together can be considered disamenity impacts and which may include noise, traffic disturbance, dust, odours and visual intrusion.

The former DETR commissioned a study to inform the decision on whether to impose a tax on aggregates and, if so, at what level (See Box 2.2).³⁵

BOX 2.2: SUMMARY OF DETR STUDY

The study estimated how much people valued avoiding the adverse environmental effects of quarrying for construction aggregates, such as crushed rock, sand and gravel, both in their locality and in landscapes of national importance.

Ten thousand respondents were picked at random from areas surrounding 21 sample quarries and other extraction sites. They were asked how much they would be willing to pay, in the form of increased taxes over a five year period, for the local quarry to be shut down, assuming that the site was restored in keeping with the surrounding landscape, and that the workers found new employment. A further 1,000 respondents, chosen at random from 21 English postcodes not near aggregates production sites, were asked what they would be willing to pay to close a quarry in a National Park (the Peak District and the Yorkshire Dales were used as examples). These results show the value attributed to the environmental damages of quarrying by people not themselves directly affected.

The environmental effects which people were asked to value included: adverse effects on nature, such as loss of biodiversity; noise from quarry transport and blasting; traffic and dust levels; and visual intrusion.

From the results of the surveys, national estimates were calculated for the average amount that people are willing to pay for the environmental benefits obtained from early closure of a quarry. These are shown below for each category of sample site:

‘Willingness to Pay’ estimates:

Case Study Sites	£/tonne
Hard rock	0.34
Sand and Gravel	1.96
Quarries in National Parks	10.52

The national average amount which individuals were willing to pay for the closure of all types of quarry sites, weighted by the type of output, was calculated to be £1.80 per tonne.

1	For more information on hedonic pricing see CSERGE publication, Day (2001) ‘The Theory of Hedonic Markets; Obtaining Welfare Measures For Changes In Environmental Quality Using Hedonic Market Data’: ‘Report for the EU Working Group on Noise’.
2	Guidance on the use of stated preference techniques can be obtained from ‘Economic Valuation with Stated Preference Techniques: Summary Guide’ available on the DfT website (http://www.dft.gov.uk).
3	The term Choice Modelling encompasses a range of stated preference techniques. The term includes choice experiments (often preferred because of their firm base in welfare economics), contingent ranking, contingent rating and paired comparisons. Further detail is contained in Economic Valuation with Stated Preference Techniques: Summary Guide, see DfT website: http://www.dft.gov.uk.
4	An introduction to multi-criteria decision analysis – weighting and scoring – is given in Multi-Criteria Analysis: A Manual available from the ODPM website: http://www.odpm.gov.uk (see DTLR archive)
5	See DfT website for additional guidance: http://www.dft.gov.uk
6	DTI uses 27 per cent as an adjustment for non-wage labour costs, while HSE uses 30 per cent. See Labour Cost Survey (LCS) 1992
7	See DfT website: http://www.dft.gov.uk
8	See DfT website: http://www.dft.gov.uk
9	It is also possible to appraise a proposal on the basis of its ‘cost utility’ if there is an appropriate measure of the benefit of an intervention in terms of human welfare.
10	The interim Interdepartmental Group on Costs and Benefits (IGCB) report, ‘An Economic Analysis of the National Air Quality Strategy Objectives’ provides an example of how to conduct an economic analysis including health benefits.
11	Further guidance on the assessment and valuation of health impacts is given in the Department of Health’s (DH) ‘Guidance on Policy Appraisal and Health’ (1995) and ‘Evaluation of Health Technologies for Use in the NHS: Good Practice Guidelines’ (1999). HSE guidance on the valuation of health impacts is included in GAP23, ‘Regulatory Impact Assessment – Policy Appraisal’, June 2002.
12	See An Economic Analysis to Inform the Review of the Objectives for Particles Air Quality Strategy available on the Defra website (http://www.defra.gov.uk).
13	Franklin (2000), chapter 7, suggests that individuals systematically undervalue small risks, possibly introducing a downward bias in estimating VPF.
14	This augmentation of the ‘own’ WTP-based figure is legitimate only if concern for others’ safety takes the form of ‘safety-focused altruism’ where despite being concerned for others’ safety, people are indifferent to other determinants of their overall well-being. For cases that are intermediate, some augmentation of the ‘own’ WTP-based figure is justifiable. (M W Jones-Lee, 1992))
15	For additional information, refer to HSE (2000a), ‘Valuation of Benefits of Health and Safety Control, Final Report’, which describes an approach used to update the DfT value for reduction in risk of a fatality.
16	Dft See. Highways Economic Note No 1. 2000) ‘Valuation of The Benefits of Prevention of Road Accidents And Casualties’). Available on the DfT website (http://www.dft.gov.uk).
17	ibid
18	See HSE website: http://www.hse.gov.uk
19	There is currently no evidence to support this adjustment. HSE has commissioned a study to investigate public preferences for preventing fatalities due to ‘dreaded’ risks to inform this issue.
20	See, for example, “Economic Valuation with Stated Preference Techniques: Summary Guide”, available on the DfT website at http://www.dft.gov.uk.
21	Contact Defra for further advice on assessing the effects of a proposal on emissions.
22	A Government Economic Service working paper ‘Estimating the Social Cost of Carbon Emissions’ suggests illustrative values for the social damage cost of carbon that can be used to estimate the monetary value of impacts once they have been quantified. A copy of this working paper is available on the Treasury’s website http://www.hm-treasury.gov.uk. Defra can provide an associated guidance note on how to use these values in policy appraisal.
23	For a technical reference on the approach to air quality mapping and modelling, see “The Air Quality Strategy for England, Scotland, Wales and Northern Ireland”, Defra, January 2000,
24	Guidance can be found on the Defra website (http://www.defra.gov.uk). Defra has also sponsored research to generate empirical estimates of UK WTP for reductions in health risks associated with air pollution.
25	These guidelines draw extensively on the Guidance on the Methodology for Multi-Modal Studies (GOMMMS) available from the DTLR archive accessed from the ODPM website: http://www.odpm.gov.uk.
26	See website: http://www.cabe.org.uk
27	“Estimating the Value of Environmental Features”, Reports to MAFF, January 1999 and June 2001.
28	“Distance decay” refers to the observation that people living further away from an environmental impact care less about it and therefore express lower valuations.
29	For example, “Valuation of Benefits to England and Wales of a Revised Bathing Water Quality Directive and Other Beach Characteristics Using the Choice Experiment Methodology”, Eftec report to Defra, 2002. Also, the Environment Agency has a register of 50 water valuation studies which covers values for recreation, water quality, flood defence, navigation and fishing (Netcen 1998).
30	Guidance is also available from two OECD publications, “Handbook of biodiversity valuation: a guide for policy makers” and “Valuation of Biodiversity Benefits: Selected Studies.”
31	The results of DfT noise studies in the UK and guidance on how to implement values when undertaking appraisal are published on the DfT and Defra websites.
32	Summarised in the 2002 report to the European Commission DG Environment “The State of the Art on Economic Valuation of noise” by Stale Navrud .
33	Summarised in a report to the Forestry Commission. “Non-Market Benefits of Forestry, Phase 1”. (See http://www.forestry.gov.uk)
34	ibid.
35	London Economics (1999) The External Costs and Benefits of the Supply of Aggregates: Phase II. Report for DETR, now found on the ODPM website (see http://www.odpm.gov.uk)