Executive Summary

1 The Issue

The UK Government is committed to a systematic appraisal of environmental impacts from projects and policies. A substantial literature exists which offers guidance on how to appraise projects. A much smaller literature exists on how to appraise policies. The purpose of this report is document what guidance already exists, where the gaps in guidance are, and how they might be filled. The focus is on policies rather than projects, but project appraisal guidelines could be adapted to help with policy appraisal. Accordingly, project appraisal guidance is included in the assessment.

This document is not concerned with providing guidance where none exists or where it is judged inadequate. Nonetheless, some intimation has to be given of what 'new' guidance might look like.

Two quite distinct audiences are identified: those who need to engage in technical appraisal at a fairly detailed level of sophistication, and those who need what we call 'rapid appraisal' techniques. The latter may be middle and senior management, but will also include the final decision-maker, the politician. Because the needs of these two audiences are different they are addressed separately in the report. Chapter 3 and Annex 1 look at the various formal techniques available for environmental appraisal. Chapter 5 addresses the needs of managers giving direct advice on policy.

Throughout, the focus is on the environmental impacts of policies and, in general, on policies that do not have environmental benefits as their main purpose.

2 The Findings

Chapter 6 lists our general findings. These are:

Project and policy guidance
Virtually all guidance relates to projects rather than policies. In principle, the techniques required for policy appraisal may not appear to differ from those needed for project appraisal. Both require some balancing of gains and losses together with considerations of equity (fairness) and wider economic considerations, such as impacts on competitiveness. In practice we suggest policies and projects do differ in terms of the wider ranging impacts from policies compared to projects, the greater uncertainty of policy impacts, and the probable smaller chance of being able to quantify policy impacts. In turn, if policies go wrong they are likely to have wider impacts whereas projects (other than 'big' projects) may fail and not have major feedback effects on households and corporations.

Consideration could therefore be given to developing fresh guidelines on policy assessment as it relates to environmental impacts. Chapter 5 sets out some suggestions and these relate to (a) determining criteria for deciding whether environmental impacts are significant, (b) a 'logical framework' for determining what the environmental impacts of a policy are, and (c) some combination of these in a policy environmental statement.

Such statements can be made more complex and detailed so as to assist with more detailed technical guidance (the technical audience) and can be focused on salient issues to facilitate rapid appraisal.

At this stage, such policy assessments will be mixes of quantification, judgement and discourse. How far the more technical stages can be modelled in the future remains to be seen. There are developments that suggest a strong role for modelled and quantified impacts in a more holistic view of policy, but there is also a strong case to be made against excessive quantification. In particular, there is a need for flexibility in assessment so that there is always scope for judgement. Put another way, carried to extremes some appraisal techniques may appear to convey 'the answer' to a policy issue. This will especially appear to be the case if guidance is offered on the basis of 'black box' models in which there is little transparency relating to how the impacts have been quantified. Chapter 5 also notes the importance of process, ie of making decisions through procedures that incorporate the judgements of those who are expert in the area and those who have a clear stake in the outcome. Equally, Chapter 5 stresses the importance of giving a framework for decision-making. Neither process nor guidance alone is adequate. Both are needed.

Guidance on risk assessment
Risk assessment has developed into a sophisticated set of procedures and there are variations in types of risk assessment. Some, while incomplete as decision-making guidance, offer helpful insights and checks on rationality. Examples include the use of 'health-health' and 'risk-risk' analysis whereby policies are evaluated not just in terms of the risks they reduce, but also in terms of the risks they may create. Risk creation may come about because of the costs of the policy, costs that remove income from households who may then reduce their expenditures on risk avoidance activities. How far these issues are of empirical relevance in the UK appears to be an open question and one that deserves to be investigated.

More generally, the USA has had substantial experience of risk assessment and the US EPA's experience is now extensively documented. Existing procedures in the UK are clearly along the right lines and formalisation of guidance, eg via the HSE, is sound. But an overall set of guidelines on risk assessment would benefit from a close look at the US experience and at some of the newer forms of risk assessment.

Guidance on multi-criteria analysis
Multi-criteria analysis (MCA) seeks to account for the fact that policies invariably have multiple objectives not all of which can be reduced to indicators familiar to cost-benefit analysis or cost-effectiveness analysis. Examples might be policies where there is concern over cost but with goals to achieve given waste recycling rates, meet international waste reduction targets and create employment. MCA offers procedures for deciding which policies are best suited to achieving these multiple ends, whilst leaving the final 'trade-off' to judgement. The dangers with MCA are well known - a temptation to multiply goals perhaps to the point where they overlap, and the potential for arbitrariness through the use of weighting and scoring techniques. Nonetheless, MCA is attractive in principle.

Currently, no guidance exists on how to engage in MCA and how to decide when it is better suited to an issue than some other technique. Existing work for the Environment Agency may correct this omission. As such a 'watching brief' is perhaps called for to see if this will address what is needed.

Guidance on cost-benefit analysis
While there is no need for new guidance on cost-benefit analysis (CBA) as such, this report does suggest consideration of guidelines on how to carry out economic valuation, an integral part of CBA. In particular, there are few sets of guidelines on how to conduct questionnaire approaches to valuation (contingent valuation and contingent ranking). Establishing such guidelines could be important and the 'NOAA' guidelines in the USA have assumed considerable importance in the conduct of monetary damage assessment. Along with these guidelines on what constitutes 'good practice', there is evidence that contingent valuation studies are not always carried out in an acceptable manner. A guidance manual would help to avoid this problem, and this matters especially in a context where there is still some suspicion that these approaches are not valid.

Sectoral focus
Annex 1 reviews available sets of guidelines by sector. We note, for example, that the water sector has many existing or 'in the pipeline' manuals and guidelines. Other sectors appear to have few or none, eg air pollution and total ecosystem change.

3 The Appraisal Techniques

Chapter 3 sets out 'cameos' of the individual appraisal techniques. These are classified as

Environmental impact assessment (EIA) with and without impact weighting schemes;

Life cycle analysis (LCA)

Strategic environmental assessment (SEA)

Risk assessment (RA), including comparative risk assessment (CRA), risk-benefit analysis (RBA), health-health analysis (HHA) and risk-risk analysis (RRA).

Multi-criteria analysis (MCA)

Cost effectiveness analysis (CEA)

Cost benefit analysis (CBA).

The techniques may compete with each other in some contexts but Chapter 3 shows that they have a particular set of relationships one to the other. One cannot engage in cost-benefit analysis, for example, without the results of a risk assessment and without the results of an environmental impact assessment. Thus RA and EIA are essential ingredients to any rational decision making procedure. LCA can be combined with EIA and with CBA, and so on. Overall, then, different techniques serve different purposes. In turn, Chapter 4 shows how each technique may be 'matched' to a policy area.

CBA and MCA are more likely to compete although, in principle, MCA seeks the evaluation of policies from the standpoint of a wider array of objectives than does CBA. The latter tends to be confined to 'economic efficiency' goals and perhaps to some concern for distributional effects.

4 The Documentation

Annex 1 lists the documentation available. This ranges from very brief 'check lists' all the way through to entire volumes of detailed identification and measurement of impacts. As noted above, some sectors are well covered and others are hardly covered at all. From a practitioner's point of view one concern is that much of the guidance is not published through orthodox channels - it exists as a 'grey literature' - nor does there appear to be a central directory of what advice is available. Annex 1 is not complete: documentation was emerging almost daily during the few months of this project. This suggests that the Annex might form the basis for a document which is regularly updated and held 'in house' either at DETR or by some appointed institute. During the process of interviewing personnel for this report it was made clear to us on a number of occasions that most people are not aware of the full extent of the documentation available.