Sustainable Consumption and Production - Development of an Evidence Base

Project Ref.: SCP001 Resource Flows

Dr Thomas Wiedmann1, Mr Jan Minx1, Dr John Barrett,1 Mr Robin Vanner,2 Professor Paul Ekins2

May 2006

Final Project Report

1Stockholm Environment Institute - YorkSally Baldwin Building - D Block

University of York, HeslingtonYork, YO10 5DD, UKTel.: +44 1904 43 2899

Email: jcm120@york.ac.uk

2Policy Studies Institute50 Hanson Street,

London, W1W 6UPTel: +44 20 7911 7500

Email: r.vanner@psi.org.uk

1 Executive Summary 9

2 Introduction: Background on Policy and Research Issues 13

SCP001 1/130

2.1 General Policy Context and Project Background 132.2 Material Flow Analysis 152.3 Why Matter matters - Motivation and Classification of MFA methodologies172.4 Changing Patterns – MFA and the government’s framework for Sustainable

Production and Consumption 19

3 Approach to Assessment 22

4 Review of Resource Flow Methodologies 244.1 Introduction 244.2 Choice of Methods 254.3 Assessment 26

4.3.1 Economy-wide Material Flow Analysis 274.3.2 Bulk Material Flow or Material Systems Analysis – Assessing the

methodological framework of the Biffaward studies 294.3.3 Analysis of Material Flows by Sector: NAMEAs, Generalised Input-Output

Models, and Physical Input-Output Analysis 314.3.4 Lifecycle Inventories 344.3.5 Substance Flow Analysis 374.3.6 Integrating approaches – Hybrid Methodologies 394.3.7 Approaches with sustainability reference points 42

4.4 From Weight to Impact – Changing Perspective on Material Flows 444.5 Policy Analysis 464.6 Resource Flow Models in the UK 474.7 Synthesis – Painting the Bigger Picture 504.8 Answering DEFRA’s specific tender questions 524.9 A Brief Guide to Policy Makers: on the choice of MFA tools - Choosing MFA

methods for SCP policy use – from general to specific 544.10 Establishing strategic evidence sources for informing the prioritisation of SCP

policies 554.10.1A general framework for tackling the most important material flows first 564.10.2Using MFA methodologies for informing prioritised MFA policies 574.10.3Towards a comprehensive and strategic evidence-based material flow

approach for informing SCP policies 59

4.11 Examples – Applying a strategic evidence-based material flow approach 614.11.1Example 1: Sector Benchmarking – Monitoring Success 614.11.2Example 2: Integrated Product Policies 624.11.3Example 3: Sustainable Waste Management 64

4.12 Towards a Research Agenda for Tracking Strategic Materials in the UK 674.12.1Policy Background and Motivation 674.12.2Identifying key materials based on environmental impacts 674.12.4Prioritising materials based on environmental impacts 694.12.5Towards a modelling framework for tracing strategic materials 724.12.6Challenges for Data Collection and Modelling 744.12.7Further considerations and recommendations for tracing strategic materials in

the SCP policy context 76

4.13 Recommendations 78

SCP001 2/130

4.14 References 80

5 Review of Biffaward Studies 875.1 Introduction 875.2 The Biffaward Studies 875.3 The Assessment 88

5.3.1 The process of assessment 885.3.2 Assessment specifications 89

5.4 The relationship with the Biffaward Movement report 905.5 Assessment results 905.6 Discussion of Results 93

5.6.1 Sectoral Studies 935.6.2 Product Studies 945.6.3 Consumption Studies 945.6.4 Waste Studies 955.6.5 Material Studies 95

5.7 Discussion and recommendations 965.7.1 The policy cycle and the Biffaward studies 965.7.2 Specific policy points arising from particular studies 965.7.3 The value of the Biffaward studies 985.7.4 Recommended additional funding of work 99

5.8 References 101

6 Development of an Indicator for Emissions and Impacts associated with the Consumption of Imported Goods and Services 1026.1 Introduction 102

6.1.1 Principles of emissions accounting and ‘embedded emissions’ 1026.1.2 About this part of the project report 1046.1.3 Requirements for an indicator of embedded emissions and impacts 104

6.2 Assessment Results and Discussion 1056.3 Specification of an indicator for embedded emissions 107

6.3.1 Main issues 1076.3.2 Data handling 1076.3.3 Interpolating time series data 1096.3.4 Outline of a pragmatic start for a model for embedded indicators 1096.3.5 Feasibility of extending the model 1116.3.6 Advantages and strengths of the specified model 1136.3.7 Assumptions, limitations and weaknesses specific to MRIO modelling 114

6.4 Policy and Other Applications 1156.4.1 Indicator for environmental impacts embedded in trade 1156.4.2 Sustainable procurement and business options – greening international supply

chains 1176.4.3 Other applications 118

6.5 Conclusions and recommendations 1196.6 References 122

SCP001 3/130

List of FiguresFigure 2.1 - Material Flows of the Domestic Economy 14

Figure 2.2 - Human induced physical flows 16

Figure 2.3 - Classification of MFA Methodologies 18

Figure 3.1 - Illustrative example of assessment matrix 23

Figure 4.1 - Software interfaces of REAP (left) and REEIO (right) 50

Figure 4.2 - MFA methodologies in relation to dematerialisaton, detoxification and policy analysis 52

Figure 4.3 - MFA methodologies in relation to policy demands on different level 54

Figure 4.4 - Framework for prioritising SCP material flow policies 56

Figure 4.5 - MFA methodologies in an integrated material flow approach 58

Figure 4.6 - Studying the physical flows associated with aluminium and steel production 73

Figure 5.1 The groupings for the Biffaward studies 88

Figure 5.2 - Assessment Matrix: the Biffaward studies 91

Figure 5.3 - The policy cycle and the Biffaward studies 96

Figure 6.1 - Consumer (“consumption emissions”) versus producer (“production emissions”) responsibility for CO2 emissions in the UK 102

Figure 6.2: Data handling protocol for a multi-region input-output model 108

List of TablesTable 2.1 - The SCP policy agenda and other relevant environmental agendas...................................21

Table 4.1 - Material Flow Methodologies included in this review........................................................26

Table 4.2 - Summary economy wide material flow analysis................................................................29

Table 4.3 - Summary bulk material flow/ material systems analysis....................................................30

Table 4.4 - Overview ecological footprinting and environmental space...............................................42

Table 4.5 - Examples: results ES analysis.............................................................................................43

Table 4.6 - Comments on linkage of MFA methodologies to environmental impact categories..........45

Table 4.7 - Overview assessment results DEFRA tender questions......................................................53

Table 4.8 - Pollutant agents and their environmental impacts...............................................................69

Table 4.9 - Environmental impact categories applied by Van der Voet et al. (2005)...........................70

Table 4.10 - Ranks of environmental impacts per kg of materials........................................................71

Table 4.11 - Ranks of total environmental impacts of materials..........................................................71

Table 4.12 - Shortlist of strategic materials...........................................................................................72

SCP001 4/130

Table 4.13 - Specific Recommendations derived from the review of the individual methodologies....79

Table 4.14 - General Recommendation.................................................................................................79

Table 5.1 - Studies which van be used directly to inform policy..........................................................92

Table 5.2 - Studies which can be used for policy with an appreciation of the study's limitations........93

Table 5.3 - Studies which should not be used to inform policy without further work..........................93

Table 5.4 - The robustness of the evidence base provided by the Biffaward studies............................93

Table 6.1 - Overview of approaches and assessment results...............................................................105

List of BoxesBox 2.1 - Tender Questions...................................................................................................................15

Box 4.1 - Policy questions economy-wide material flow analysis........................................................27

Box 4.2 - Policy questions environmental input-output analysis..........................................................31

Box 4.3 - Policy questions LCI/LCA.....................................................................................................34

Box 4.4 - Policy questions substance flow analysis..............................................................................37

Box 4.5 – Key recommendations environmental impact assessment....................................................46

Box 4.6 - Key recommendations policy analysis..................................................................................47

Box 4.7 - Good practive sector benchmarking......................................................................................62

Box 4.8 - Good practice prioritising product groups according to environmental themes....................64

Box 4.9 - Good practice: The Japanese waste input-output model.......................................................66

AcknowledgementsThe authors are grateful to Jonathan Nobbes and David Aaron Thomas for their valuable contributions to the review of the Biffaward studies (Section 5). We are further indebted to Dr. Manfred Lenzen for his active involvement in Section 6.3 and Appendix III.3 and his very useful comments on the report as well as Dr. Karen Turner, University of Strathclyde for her contributions to Appendix III.3. We would also like to thank the Steering Group members and the reviewers for their comments on the interim report as well as Dr. Klaus Hubacek, University of Leeds, for commenting on earlier drafts.

The work has been funded by the UK Department for Environment, Food and Rural Affairs (DEFRA) through the programme ‘Sustainable Consumption and Production – Development of an Evidence Base’, Project code CTX0505/SCP1.1.

SCP001 5/130

GlossaryACORN A classification of residential neighbourhoods

BedZed Beddington zero energy development

BRE Building Research Establishment

BMFA Bulk material flow analysis

CDM Clean Development Mechanism

CO2 Carbon Dioxide

CSR Corporate social responsibility

Defra Department for Environment, Food & Rural Affairs

DMC Domestic material consumption

DTI Department for Trade and Industry

EA Environmental accounts

EC European Community

EF Ecological Footprint

EIOA Environmental input output analysis

EIOLCI Environmental input-output life-cycle inventories

EMFA Economy-wide material flow accounting

EPA Environmental Protection Agency

ES Environmental Space

EU European Union

Eurostat Statistical Office of the European Communities

GDP Gross Domestic Product

GINFORS Global Inter-industry forecasting system

GIOA Generalised input-output analysis

GTAP Global Trade Analysis Project

IEA International Energy Agency

IIOHLCI Integrated input-output hybrid life-cycle inventories

IO Input-output

IOHLCI Input-output hybrid life-cycle inventories

IPA Impact Potential Approaches

IPP Integrated product policy

IPPC Intergovernmental Panel on Climate Change

ISIC International Standard Industrial Classification

ISO International Organisation for Standardisation

SCP001 6/130

IVEM Centre for Energy and Environmental Studies, University of Groningen

JI Joint Implementation

LCA Life-cycle analysis

LCI Life-cycle inventories

MFA Material flow analysis

MIOTs Monetary input-output tables

MOSUS Modelling opportunities and limits for restructuring Europe towards sustainability

MRIO Multi-regional input-output

MSA Material system analysis

NAMEA National accounting matrix including environmental accounts

NHS National Health Service

NISP National Industrial Symbiosis Programme

NOx Nitrogen oxides

OECD Organisation for Economic Co-ordination and Development

ONS Office of National Statistics

PIOA Physical input-output analysis

PM10 Particulate Matter of less than 10 millionths of a metre

POITs Physical input-output tables

Prodcom Products of the European Community

REAP Resources and Energy Analysis Programme (Stockholm Environment Institute)

REEIO Regional Environment-Economic Input Output model (Cambridge Econometrics)

REMAT Resource Management Tool (University of Surrey)

RMFA Regional Material Flow Accounting Model (University of Surrey)

ROW Rest of the world

RSWT Royal Society for Wildlife Trusts

SCP Sustainable consumption and production

SCPnet Sustainable consumption and production Network

SEI Stockholm Environment Institute

SD Sustainable Development

SFA Substance flow analysis

SPA Structural Path Analysis

TIOLCI Tiered input-output life-cycle inventories

TMI Total material input

TMC Total material requirement

UK United Kingdom

SCP001 7/130

UN United Nations

UNCTAD United Nations Conference on Tariffs and Trade

WRAP Waste and Resources Action Plan

WSSD World Summit for Sustainable Development

WTO World Trade Organisation

SCP001 8/130

Executive Summary

Part A – Review of Biffaward Studies (by Robin Vanner and Paul Ekins)

This part of the report introduces the Biffaward series of reports which were commissioned following the publication in 1997 of an overview publication (Biffa 1997) relating to resource use and waste management in the UK. The rationale for the investment of more than £10.6 million from the landfill tax credit scheme was that information about resource flows though the UK economy is of fundamental importance to the cost effective management of those flows, especially at the point at which those flows become ‘wastes’. Now that the Biffaward programme of research is nearing completion, the purpose of this assessment is to identify those studies which are both sufficiently robust, and relevant to future policy agendas. The results of the assessment have been the subject of a formal consultation with the reports’ authors, as well as a dialogue with practitioners and policy makers through a project workshop. Of the 48 studies which have been assessed:

7 studies can be used to inform policy directly.

29 studies can be used to inform policy, but with a note of caution about some aspect of the study (in many cases this is due to the results representing modelled data’.

12 studies were assessed not to be usable in relation to future policy without further work.

Considering the results by how they relate to the chain of production and consumption:

The product, material and consumption type studies were all assessed to be robust and policy relevant, although many require a note of caution in relation to some study limitations.

There were mixed assessments of studies relating to production. There was only one completed primary production study available for assessment. All but one of the secondary production studies were considered to be usable in policy, although the age and lack of updateability of some of the datasets was identified as a significant limitation for some. The tertiary or service sector studies faced an even greater challenge in terms of the collection of data and only two of them were assessed as usable in policy,

10 out of the 15 waste studies were considered to be usable in policy in some way. It is difficult to make generalisations for such a varied group of studies. However, the studies relating to research and model development tended to be more robust than the sectoral level waste studies, due to the higher quality of the data that they used.

This part to of the project goes on to conclude that the Biffaward programme of studies generated many benefits for a large range of users and showed how a comprehensive or consistent classification of material flows within the economy might be achieved. A full and comprehensive reading of the reports has led to the conclusion that, in industry, there may be potential to enhance industrial symbiosis utilising the present knowledge base, through the development of industrial networks. Such networks need to be developed and shown to lead to mutual industrial advantages and benefits, on the basis of current knowledge, before further investment is made in extending the quantification and specification of industrial material flows. The process of mapping out consumption is now well developed but needs to be supported by better data sources so that the findings relate even more closely to actual consumption patterns.

SCP001 9/130

Part B – Review of Resource Flow Methodologies(by Jan Minx, John Barrett and Thomas Wiedmann)

Resource Flow or Material Flow Analysis (MFA) can be considered as the application of the biological concept of metabolism to human society. Society is depicted as a living organism which continuously withdraws resources from nature, digests them in the transformation processes of production and consumption and finally releases them back to nature as wastes/residuals. MFA can be seen as an attempt to describe and analyse all or a subset of these metabolic (material and energetic) processes quantitatively.

Part B of this report reviews the MFA literature and assesses the robustness and policy relevance of the underlying MFA methodologies for informing the government’s SCP agenda and their potential to contribute to the build-up of a physical evidence base. Recommendations are provided on how to exploit existing, and fund future, research. While the focus of the review is on five general MFA methodologies (economy-wide material flow analysis (EMFA), bulk material flow or material systems analysis (BMFA/MSA), environmental input-output analysis (EIOA), life cycle analysis (LCA) and substance flow analysis (SFA)) and the hybrids which can be derived through methodological integration, it also briefly covers methodologies with a direct sustainability reference, such as the Ecological Footprint and Environmental Space and some existing UK specific MFA models.

The assessment shows that all methodologies can contribute valuable evidence to inform SCP policy and outlines in which context each might be best applied. It highlights that all methodologies have their individual strengths and weaknesses. In order to overcome these limitations there is a need for the government to take a leading role in:

Establishing good practice for methodologies, where standardisation is not sufficiently developed;

Encouraging the integration of methodologies where possible to increase their robustness and applicability to the SCP agenda;

Orchestrating methodologies in an integrated material policy approach, which makes best use of the individual strengths of the methods and closely responds to the needs on the SCP agenda.

With regard to the government’s emphasis on focusing first on physical flows with the most significant environmental impacts, the report proposes a strategic material flow approach for informing SCP policy. This requires linking MFA methodologies as closely as possible with environmental impact (potential) assessments and underlines the importance of focussing SCP policies not only on the dematerialisation of human (economic) activities, but also on their detoxification through the substitution of harmful with less harmful physical flows in an integrated approach.

The key to supporting the SCP agenda with strategic MFA evidence is the build-up of comprehensive models which allow the prioritisation of policies as well as the use and further build-up of evidence. Two complementary models are outlined, which the government should consider to adopt – one based on a EIOA methodology with a particular strength in prioritising physical flows within the domestic supply-chain context, the other one based on a hybrid LCA-EMFA methodology with a particular strength on the prioritisation of (finished) materials.

Finally the report follows the government’s request made in later stages of the project to identify a selection of key materials with particularly high environmental impacts and devise a methodological framework in which they can be traced. Based on available UK evidence found in a study by Van der

SCP001 10/130

Voet et al. (2005) a list of 5 materials is identified in a transparent approach. However, concerns about the robustness of the data used for the identification are highlighted.

Part C - Development of an Indicator for Emissions and Impacts associated with the Consumption of Imported Goods and Services(by Thomas Wiedmann, Jan Minx and John Barrett)

When people in the UK consume goods or services they also produce carbon dioxide emissions, not only directly, for example when driving a car, but also indirectly. This is because the production of these goods or services will, most likely, have involved the emission of carbon dioxide earlier on. These indirect “emissions from consumption” can occur anywhere, even in foreign countries when goods or services are imported to the UK. In a manner of speaking, these emissions are “embedded” with the consumption activity.

There are a number of good reasons why national Governments should report environmental (and other) impacts embedded in trade, e.g. to improve sustainability reporting indicators, to enable meaningful international comparisons, to instigate country dialogues and assist in trade negotiations or to generally raise awareness amongst consumers.

It is difficult, though, to measure these embedded emissions. At present there is no standardised calculation method and therefore no reliable indicator that quantifies total “emissions from consumption” is currently available.

This project set out to identify the most appropriate approach for constructing such an indicator and an extensive literature review was undertaken. The assessment of the findings was aimed at identifying the strengths, weaknesses, assumptions, limitations, data requirements and suitability of different approaches.

Many studies assume that all imports are produced using UK production technologies. This is called a ‘single-region’ assumption. However, foreign countries use different production technologies and energy mixes and therefore the results are not very helpful. A sound respond to this problem is to extend the basic single-region framework to the international case and to employ a multi-region model, ideally covering all trading partners of the country under investigation. In addition, the review showed that such a model should also use input-output analysis, a well established method that automatically takes into account the impacts of all steps that were required to produce the goods and services that are consumed.

For these reasons, a ‘multi-region input-output’ (short: MRIO) approach seems to be the most suitable and promising choice for a model that calculates an embedded emissions indicator. A specification for the development of a MRIO indicator for embedded emissions is given in the report and options for the actual implementation, including cost reduction options, are presented.

The implementation and application of a full MRIO framework poses three basic challenges: data availability, data reconciliation and computability. These issues and possible practical solutions are discussed in detail in the report. The crucial part of an operational MRIO framework is a code protocol that processes data of any kind in a highly efficient way. In essence, this is a sophisticated computer programme that can ‘digest’ data from different countries and years in different classifications and valuations with data gaps and inconsistencies.

SCP001 11/130

In order to minimize resource requirements, it is reasonable to explore possibilities to simplify the model in a way that still provides the advantages and holds the option of extending it to a full-scale model later on. The objective of any future research should therefore be to implement an initial, relatively small, data and model framework that is easily expandable without major adaptations. Reductions in data and therefore resource requirements can be made by cutting down the number of countries/regions, years, sectors or indicators. A ‘slim’ option, for example, could initially feature just two regions (e.g. UK and the rest of the world) and be expanded over time. The relatively low data requirements of a two-region model allow the early construction of a time series of embedded emissions.

We also describe numerous policy and other applications of the specified approach. Most importantly, the suggested model allows for a robust, reliable and reproducible quantification and analysis of environmental impacts embedded in the international trade of goods and services – not only carbon dioxide emissions, but ultimately all sorts of environmental and social impacts. This is a novel approach. Studies so far either covered trade only partially or used short-cut estimates that do not really withstand scientific scrutiny and that cannot be truly relied upon in international negotiations about (embedded) emissions.

SCP001 12/130

1 Introduction: Background on Policy and Research Issues

1.1 General Policy Context and Project BackgroundThe Sustainable Consumption and Production (SCP) movement initiated in Agenda 21 with the acknowledgement that “the major cause of the continued deterioration of the global environment is the unsustainable patterns of consumption and production, particularly in industrialised countries, which is a matter of grave concern, aggravating poverty and imbalances” (UN, 1992: 4.3). Therefore, policies were called for that help

1. “To promote patterns of consumption and production that reduce environmental stress and will meet the basic needs of humanity;

2. To develop a better understanding of the role of consumption and how to bring about more sustainable consumption patterns” (UN, 1992: 4.7).

This was re-enforced at the World Summit for Sustainable Development (WSSD) with new global commitments on SCP. In the aftermath the UK government re-viewed its own efforts and followed up with a national SCP framework called ‘Changing Patterns’ (DEFRA, 2003). This was taken down to an implementation level in the Sustainable Development Strategy in 2005, in which a general action plan for SCP was proposed in chapter 3 (DEFRA, 2005a). At the heart of all actions is the overriding objective to decouple economic growth and resource utilisation and to detoxify economic processes/economy/society. Hence, SCP as understood by the UK government is about breaking the links between economic growth and environmental degradation as well as reducing the environmental impact per unit of physical flow. This requires a consequent shift towards new products and services with lower environmental impacts and therefore a general re-thinking of how we produce and consume. Hence, action needs to be induced in three major areas:

“better products and services, which reduce the environmental impacts from the use of energy, resources, or hazardous substances,

cleaner, more efficient production processes (doing more with less), which strengthen competitiveness, and

shifts in consumption towards goods and services with lower impacts” (DEFRA, 2005: 44).

The whole spectrum of stakeholder groups in the UK need to be involved in this challenge and make their contribution. This goes all the way from individuals, who need to become more aware of their roles as responsible citizens and resource consumers, through businesses, which need to develop new eco-innovative business models, and to government itself. One important contribution from the scientific community is the provision of a sound and robust evidence base and a useful set of analytical tools to inform decision makers in the best and most reliable way. Such an evidence base needs to provide organised and comprehensive information about the usage of natural resources by human activities and the arising environmental impacts caused during the process of resource utilisation in a lifecycle perspective, from extraction to disposal. A lifecycle perspective also requires taking into account the environmental impact caused abroad by domestic consumption of imports. All this is vital to the success and delivery of the overall SCP programme.

SCP001 13/130

Figure 2.1 - Material Flows of the Domestic Economy (from Stahmer et al., 1998)

Therefore, DEFRA is putting a significant amount of resources into the development of such an evidence base – which draws on the wealth of existing research and includes new work – to stimulate a more knowledgeable and informed debate. In this development process DEFRA is currently establishing long term evidence needs and priorities by commissioning work to review existing literature, compile quantitative and qualitative evidence, identify gaps, and define new research priorities. This project focuses on the potential contribution to the evidence base of approaches and studies which use physical information to inform the decision-making process about the UK’s exchange relationship with the natural environment. They have frequently been summarised under the heading of resource or material flow analysis (MFA). Hence, MFA is understood as a family of methodologies which comprises all with the measuring and understanding the physical exchange processes between the national economy and the environment (Femia and Moll, 2005). The most important physical exchange relationships relevant for MFA are summarised in Figure 2.1.

SCP001 14/130

This review will first focus on general MFA methodologies and how they can be used within the Government’s SCP programme. In the second part the review will concentrate on an array of material flow studies, which have been commissioned between 1997 and 2006 within the Biffaward Mass-Balance scheme. This review will assess not only the individual qualities of different methodologies, but also how the results can be used to support the UK’s SCP agenda.

Figure 2.1 also highlights the UK’s physical integration with the rest of the world through trade patterns. The domestic consumption of imports causes environmental pressure elsewhere through resource extraction and waste/pollution release. This is particularly relevant when a fair burden sharing is concerned. However, the estimation of environmental pressure abroad causes a whole array of methodological problems. Therefore, a third part of this report will also review existing methodologies to estimate the “embedded CO2 emissions” of imported goods and services, and will propose a robust estimation framework for such an indicator for the UK. The particular objectives of the different project parts are outlined below in Box 2.1.

Part (A) Biffaward Mass Balance StudiesObjective A1) Assess the quality of the evidence presented by Biffaward Mass Balance projects.Objective A2) Assess policy relevance of the evidence presented by Biffaward Mass Balance projects and

recommend further action in terms of exploiting existing or funding further research.

Part (B) Resource Flow StudiesObjective B1) Select those approaches, methods and tools that are most relevant for SCP policy.Objective B2) Assess the quality and the robustness of these approaches and of the data which they have both

used and generated.Objective B3) Identify potential benefits for policy formulation and recommend further action in terms of

exploiting existing or funding further research.

Part (C) Embedded EmissionsObjective C1) Review and assess existing approaches to estimate emissions embedded in traded goods and

services.Objective C2) Provide detailed specification for the development of an indicator for embedded emissions.

Box 2.1 - Tender Questions

Please note that the below review will start with part B, proceed with A, and end with part C, reflecting the intention of the authors’ to move from the general to the specific. The next Section will provide a brief introduction to material flow analysis and link this to a clear definition of the project scope.

1.2 Material Flow AnalysisThe biological concept of metabolism can be used to depict society like a living organism, which continuously draws resources from nature, digests them (in the processes of production and consumption) and finally releases them back to nature as wastes/residuals. In material terms nature and society therefore constitute a materially closed system (see Jackson, 1996), with society usually being conceptualised as embedded in the environment (see Figure 2.1). The current/contemporary notions of ‘industrial’ (Ayres and Simonis, 1994) and ‘societal metabolism’ (Fischer-Kowalski, 1997) are prominent examples for such metabolic conceptualisations of the physical nature-society relationship.

Material Flow Analysis (MFA) can be seen as an attempt to describe and analyse all or a subset of these metabolic (material and energetic) processes quantitatively. Hence, we might want to define

SCP001 15/130

MFA following Brunner and Rechberger (2004: 3) in the most general way as the “systematic assessment of the flows and (changes in) stocks of materials within a system defined in space and time”. MFA puts an emphasis on tracing and modelling the linkages between sources, pathways and sinks of all, or a well-defined selection of, physical flows, and comprises all methodologies which try to do so. Therefore, it might be best understood as a meta concept, or family of methodologies1 that can be applied as a useful framework for quantifying and analysing the metabolic flows between the human and the natural system on various levels of aggregation (Daniels, 2002: 88).

Figure 2.2 - Human induced physical flows

On a conceptual level there are two systems of interest for MFA: (1) the economy2, which is driven by humans, and, (2) the environment, which is driven by nature. We can therefore assign physical flows to four different domains to describe human induced flows as shown in Figure 2.2. They can either occur within the economy or within the environment, or they might be in the process of crossing the system boundaries; either from the environment to the economy as resource inputs or vice versa as wastes/residuals. All four flow domains are therefore in the scope of MFA.

However, two general approaches for handling anthropogenic flows have emerged in the literature (see Fischer-Kowalski, 1997). The first one (approach 1) has analysed the physical society-nature (exchange) relationship from a socio-economic perspective usually with the aim to learn where the physical flows come from, how they transmit through the socio-economic system and where they end up, with the ultimate goal of devising policy measures for intervention. This approach has usually focused on the first three quadrants in Figure 2.2. The second one (approach 2) has started the inquiry from the ecological system and tried to assess the impacts of anthropogenic flows in the environment. This approach has focused on quadrants 2 to 4. This review mainly focuses on methodologies that have emerged from approach 1. That is to say, attempts to assess the impact of human induced physical flows in the environment will be left aside initially, even though in later parts some aspects will be touched upon due to their relevance in the process of compiling and using the SCP evidence base most effectively to inform SCP policy.

Fundamental to all different MFA family members is the existence of some established rules, which allow for a consistent accounting of the physical flows in a first instance. The basic accounting principle underlying the measurement of physical flows is the law of the conversation of mass as officially established, for example, in the System of Integrated Environmental and Economic

1 A classification and distinction of these approaches will be provided later on in this Section.2 The notions of Society and Economy will be used interchangeably.

SCP001 16/130

Accounting SEEA (United Nations, 2003: chap. 3) or in the Eurostat Guide for economy-wide material flow accounting (Eurostat, 2001). It says that if we measure the total flows of materials supplied to the economy, this must be equal to the total used by the economy and vice versa. Alternatively, it could also be said that the sum of physical flows from all origins must equal the sum of flows to all destinations (United Nations, 2003). This is commonly known as the Material Balance Principle. This principle provides a logical basis for all physical book-keeping efforts. It is important to note that the acronym MFA is also widely used in the literature to refer to material flow accounting. However, in this report it will be exclusively reserved to refer to “material flow analysis,” which subsumes material flow accounting under its heading, but also comprises all methodologies that help to study and understand these flows ex-post. Like environmental impact analysis policy (and therefore ex-ante), analysis will only be briefly touched upon despite its crucial role in reducing the large amount of uncertainty associated with (future related) policies and the best choice of policy instruments.

1.3 Why Matter matters - Motivation and Classification of MFA methodologies

The importance of environmental problems is well established at the political level. International conferences on the state of the environment have been taking place since the 1950’s with increasing frequency, environment departments have been established in most countries of the world, multi-national and also many small and medium sized enterprises have adopted environmental management systems and started to assess their performance in environmental terms and most countries have established some sort of environmental regulation. Ultimately the interest in the issue is rooted in concerns about human well-being – today and in the future. It is not surprising that environmental policies have, for example, often tended to be readily enforced when pollution has immediate adverse effects on a local community. It has shown to be much more difficult to tackle environmental problems whose impact on human well-being lies in the future, or those which are jointly caused by a consortium of countries – as with transboundary and global pollution problems.

However, despite their differences, all these problems share the fact that they arise in the course of the physical exchange and transformation processes taking place within the societal metabolism. Hence, it is not very difficult to establish that a solution for today’s environmental problems requires a sound understanding of the physical metabolism of societies – resource extraction, transformation and waste release. This need applies to industrialised countries that seek to curb high resource use as well as to middle and low income countries, which have much more potential to leapfrog high level resource use patterns without sacrificing economic development. Establishing a robust evidence base about the mechanics and dynamics of the material pathways through the human sphere is therefore of key importance for tackling environmental problems in a preventive policy approach, as highlighted in the sustainable development strategy of the government (DEFRA, 2005a). Hence, MFA approaches are at the heart of environmental decision-making.3

Environmental problems are generally related to resource use patterns in two different ways – in the level of resource use and the impact per unit of mass flow. In the MFA literature, therefore, two general strategies for tackling environmental problems have emerged in the policy context of MFA - dematerialisation and detoxification. The former focuses on the reduction of the environmental stress/pressure through cutting down the total amount of physical flows. It therefore often targets 3 The reader should keep in mind that we apply the most general notion of MFA covering all flows within

and between economy and environment. This does not say that the approach covered in this review can tackle all environmental problems. A significant amount of insight must also come from an understanding of the impact of flows in the environment, which is usually covered by the research efforts of ecologists.

SCP001 17/130

physical flows of high volume. The latter concentrates on reducing the impact per unit of physical flow. Both dematerialisation and detoxification are inherently interlinked and crucial for pursuing the goals set-out on the SCP agenda. However, most MFA methodologies have no in-built capacities to address environmental impacts. They can only prioritise physical flows according to their weight and achieve detoxification only indirectly through dematerialisation. However, in some later parts of the report it will be briefly outlined how environmental impacts can be linked to MFA methods and how detoxification policies can be prioritised using physical flow evidence.

In order to classify these various methodologies for the course of this project, it is useful to distinguish them in the material as well as the human sphere according to their level of aggregation. In we have attempted to map a selected number of MFA methodologies and other important physical evidence sources into a 3x3 matrix distinguishing a macro, meso and micro-level in both spheres. In the material sphere we distinguish between substances, bulk-materials and aggregate physical flows. (Aggregate physical flows add up the weights of all physical flows circulating in society/economy within a certain reporting period.) In the economic sphere we distinguish between enterprises sectors and aggregate economic activity. It should be noted that products are part of the bulk material category. Other similar classification systems of material flow methodologies have been proposed by Daniels (2002), Bringezu (2001) and Femia and Moll (2005).

Figure 2.3 - Classification of MFA Methodologies

Assigning the different MFA methods and other important physical evidence sources to a particular box in the matrix is not always unambiguous as they can often be applied at various aggregation levels in the human or material sphere. The assignment just reflects the area where a methodology is

SCP001 18/130

pre-dominantly used based on the authors’ judgement. Especially with respect to the material sphere, methodologies can theoretically be applied at any aggregation level. Pedersen (1999), for example, compiled Physical Input-Output Tables also for substances. Bulk MFA approaches can also widely differ in their aggregation level, depending on the purpose they are assigned. For some methodologies the assignement to one box does not seems to be justifiable at all as for environmental input-output analysis. In such cases the method has been assigned to more than one box.

In terms of the policy relevance of the various methodologies readily hints towards what might turn out to be a fundamental insight of this report. The various methodologies under consideration seem to rather complement than substitute each other as they refer to different aggregation levels in the human and material sphere and therefore might answer different sets of policy questions. Therefore, the main challenge might be to understand their individual policy potential and to learn how they can be used most effectively together to inform the SCP agenda.

1.4 Changing Patterns – MFA and the government’s framework for Sustainable Production and Consumption

Sustainable Consumption and Production (SCP) has developed into the over-riding priority area to tackle environmental degradation within the Government’s Sustainable Development Strategy (DEFRA, 2005a). Behind this promotion stands the insight that the production and consumption of goods and services in the UK (as in all other industrialised countries) is far too resource intensive. This results in a variety of environmental problems at home as well as abroad such as global warming, deforestation, biodiversity loss, soil erosion or desertification, which adversely affects human well-being today and in the future. The aspiration of SCP policies, therefore, is to foster lifestyles that are much less dependent on natural resources. This implies that goods and services need to be both produced more efficiently and consumed differently.

The UK government has been actively promoting SCP policies since the middle of the 1990s, when the agenda it featured prominently in the SD strategy ‘A Better Quality of Life’ (DEFRA, 1999). After the World Summit on Sustainable Development in Johannesburg, the government established its SCP policy framework ‘Changing Patterns’ (DEFRA, 2003). In this document the government laid out its SCP vision to “deliver continuous economic and social progress that respects the limits of the Earth’s ecosystems, and meets the needs and aspirations of everyone for a better quality of life now and for future generations to come” (DEFRA, 2003: 6). This vision was thought to accomplish:

1. “Decoupling economic growth and environmental degradation.

2. Focusing policy on the most important environmental impacts associated with the use of particular resources, rather than on the total level of resource use.

3. Increasing the productivity of material and energy use, as part of the broader Government commitment to increase the productivity of the nation.

4. Encouraging and enabling active and informed choices of individual and corporate consumers who practice more sustainable consumption” (DEFRA, 2003: 6).

Throughout the outline of this framework two interlinked themes – which are essential if the above objectives are to be achieved – recur. First, all life-cycle impacts of resource use need to be taken into account. Second, (and connected to the first) the impacts of domestic consumption patterns in the rest of the world need to be considered, when SCP policies are devised. The methodological question of how these impacts can be estimated robustly will be specifically addressed in a separate part of the report.

SCP001 19/130

The rather general guidelines and principles laid out in the ‘Changing Patterns’ SCP framework were taken forward in the recently revised Sustainable Development Strategy (DEFRA, 2005a). Chapter 3, called ‘One Planet Economy,’ provided what might be called a national action plan for SCP. These actions might be clustered in seven key areas identified to accelerate the shift towards a more sustainable system of consumption and production:

1. Products: Strengthening UK and international measures to improve the environmental performance of products and services including improved product design;

2. Production: Improve resource efficiency and reduce waste and harmful emissions across business sectors;

3. Consumption: Influence consumption patterns including proposals for new advice for consumers;

4. Procurement: Sustainable procurement in the public sector to make the UK a leader within the EU by 2009;

5. Innovation: Support for innovation to bring through new products, materials and services;

6. Business: Increase transparency, corporate responsibility and skills in business and other organisations.

7. Waste: Increased emphasis on reducing waste at source and making use of it as a resource.

Because policy relevance is of key concern for this project, it might be helpful to break the national SCP agenda further down. Table 2.1 gives an overview of some of the most important initiatives on the environmental policy agenda. Some of these initiatives have directly emerged out of the SCP debate and can be related to the seven key policy areas outlined above. The lower half of the table shows those initiatives, which have appeared elsewhere in the environmental discourse, whilst still being of relevance to the SCP discourse. The initiatives can be further assigned to different institutional levels (global, national, regional local, sectoral, company).

Note that this matrix only shows key elements on the SCP agenda based on the authors’ perception. There are other important programmes and initiatives not represented in the matrix which are addressed in later Sections of this report. It is one core concern of this review to assess to what extent the different MFA methodologies, in general, and the Biffaward studies, in particular, can inform decision-making on the SCP and more general environmental agenda.4 Table 2.1 will be used to locate the contribution on the SCP map in the assessment carried out in Sections 4 and 5 of this report.

4 Specific attention will be directed towards the methodological question of how to create a robust indicator for assessing the environmental pressure caused in other countries from the domestic consumption of imports.

SCP001 20/130

    Public Private

Sustainable Consumption and Production (SCP)

policy agenda (P1)

  Global European UK/DA National Regional Local Sector Company

ProductionDecoupling/'green

growth'/eco-efficiency/resource productivity

 

Sustainable business

Corporate reporting guidelines/regulations

Products Integrated product policy Market transformation

Innovation Green 'challenge' funds

Procurement Green public procurement Green public procurement

Consumption

Information provision, economic instruments

Waste Waste management planning

Waste management

planning

Other environmental policy agendas

(P2)

  Comparison of national SDSs

EU SDS/ 6EAP/ Thematic strategies

Sustainable development strategy (SDS) Regional SDS

Sectoral sustainability

strategy

Corporate sustainability/CSR

strategy

 Resource

accounting methodologies/

indicators

Resource accounting

methodologies/ indicators

Resource accounting methodologies/ indicators Benchmarking

Part of Environmental Management

System

  Benchmarking Benchmarking Modelling policy impacts (also by sector/region)

Regional Development

Agencies

Local authorities

Meeting sectoral targets (e.g. recycling)

 Sectoral analysis/

prioritisation (i.e. where are the biggest impacts)

Eco-industries/Indus-trial symbiosis

  Identification of environmental limits

Freight traffic planning

Table 2.1 - The SCP policy agenda and other relevant environmental agendas

SCP001 21/130

2 Approach to AssessmentA generic method for assessing resource flow studies has been developed which aims to determine systematically how studies perform against desired characteristics. The objective of the project is to determine the policy relevance and robustness of the evidence base provided by such studies. To be robust, studies need to be assessed in terms of the appropriateness of the boundary set around the analysis, the quality of the data that it uses, the credibility of the methods applied and the consistency and usefulness of the results. Depending on the type of study being assessed, the criteria used to assess these will differ. However, there are a number of generic criteria which need to be considered when assessing the robustness of a study and which are likely to be required in all assessments of the project. Hence, the policy relevance of studies will depend on whether:

Boundaries are clearly defined and are both relevant and valid for what is being recommended.

Data is of relevant age, updatable, reliable, available, documented and represents real observations or measurement (i.e. the robustness of the study will be improved if the data is not modelled from other data).

Methods used are robust (i.e. credible and reliable), objective (in terms of their selection and application) and they have been verified and reviewed by others; and

Results are consistent with recognised guidelines and systems of classification as well as being comparable with other data sets.

The distinction between criteria which increase the policy usefulness of a particular study and those criteria which are considered to be essential for the study to be used in policy is very important and something which is reflected in this assessment methodology. If an overly cautious distinction is made, useful studies may be excluded from use in policy making which would represent a lost opportunity to formulate better policy. This is of particular relevance if there is no practicable alternative to conducting a particular study without accepting compromises. Where this is the case, the study’s results represent the best evidence available to policy makers when deciding how to proceed. It may be argued that it is better to use a good study which has weaknesses than to leave the policy maker to rely entirely on ad-hoc informed judgement. This is of course assuming that the policy maker has been made aware of and is able to appreciate the implications of the study’s weaknesses.

If however an overly lenient distinction is made between essential and inessential criteria, poorly conducted studies may end up being used to inform policy. This could lead to policy being based on a misunderstanding of what a study shows, or in some cases incorrect evidence. This distinction is therefore central to the assessment methodology and has been made with the agreement of the project’s steering group whilst asking: does this criterion need to be satisfactorily achieved for the study’s conclusions and recommendations to be valid? Within all of the project’s assessment matrices, essential criteria are marked as orange and inessential criteria are marked as yellow.

The assessment methodology also sets out to identify excellence, and where possible, rank the studies being assessed. This has been achieved by having three possible outcomes for each criterion (good, satisfactory and unsatisfactory). The particular requirements for good and satisfactory assessments differ for each criterion in each part of the project. However, the following generic distinction is being sought when these requirements are being set:

Good – Best practice has been followed and the issue has been explicitly addressed in the report.

SCP001 22/130

Satisfactory – The minimum required for the report’s recommendations to be valid for policy purposes.

Unsatisfactory - By implication, studies which are assessed to have failed to achieve a good or satisfactory outcome will be considered to be unsatisfactory for policy purposes.

Within all of the project’s assessment matrices, good assessments are shown with a light green, satisfactory with light yellow and unsatisfactory assessments with a red colour.

When combined, the following overall assessment outcomes can be generated in green, yellow and red:

Green – (No red unsatisfactory assessments) - Can be used directly to inform policy.

Yellow – (Only unsatisfactory red outcomes in inessential criteria) – Can be used to inform policy with an appreciation of the study’s weaknesses.

Red – (Assessed as unsatisfactory in at least one essential criterion) - Should not be used to inform policy without further work.

Inessential

Criterion 1

Essential

Criterion 2

Inessential

Criterion 3

Overall

Assessment 

Study 1 Satisfactory Good Unsatisfactory

Study 2 Good Satisfactory Satisfactory

Study 3 Good Unsatisfactory Good

Study 4 Unsatisfactory Satisfactory Unsatisfactory

Figure 3.4 - Illustrative example of assessment matrix

How these overall assessments are arrived at is also shown by a simplified example. As shown by Figure 3.4, even though study 1 was assessed as being unsatisfactory for criterion 3, this has not excluded its use in policy as criterion 3 is considered as inessential. Study 1 has therefore been overall assessed as yellow. As a further example from Figure 3.4, even though study 3 was assessed as good for two of the three criteria, it should not be used to inform policy at present as it was assessed as unsatisfactory in criterion 2, a criterion considered to be essential to a robust study.

Overall scoring and ranking of studies is problematic. The contribution each criterion makes to a study’s overall robustness and policy relevance will not be the same. Indeed the relative weight one might place on the different criteria will be particular to each reader for each study, depending on their interest. This has been dealt with in part by providing detailed criteria which provide three categories of policy relevance. Beyond this, scoring and ranking has been dealt with differently for the three parts of the project, depending on what is thought to be the most effective way of communicating outcomes of the particular assessment.

SCP001 23/130

3 Review of Resource Flow Methodologies(by Jan Minx, John Barrett and Thomas Wiedmann)

3.1 IntroductionThis Section reviews the robustness and policy relevance of the different MFA methodologies and their role in the build-up of a physical evidence base to inform the government’s SCP policies. Initially the review solely concentrates on the capabilities of the different MFA methods to trace and analyse physical flows within a defined system as requested by DEFRA. This implies a focus on ex-post analysis on the one hand, and weight and dematerialisation on the other hand. However, restricting the review in such a way excludes two other crucial applications of MFA evidence – namely detoxification and the reduction of environmental impacts per unit of mass flow as well as (ex-ante) policy analysis.

Detoxification prioritises physical flows according to their environmental impacts. This is one of the key concerns of the government’s SCP agenda and crucial in the use and build-up of the physical evidence base. Ex-ante policy analysis reduces the uncertainties associated with future policies by testing their (economy-wide) economic, social and environmental implications and choosing the most effective policy instruments to achieve a certain SCP policy target. This might be seen as one of the most policy relevant application of physical evidence. In this context Proops (1991) has for example emphasised the need to move from national accounting to global modelling. Hence, excluding detoxification and policy analysis would impede a fair assessment of the policy relevance of the different MFA methodologies and the provision of valuable advice for the use and build-up of the physical SCP evidence base.

For matters of conciseness the main body of the report will mainly present the results of the various assessments, which have been carried out in the course of this project. The interested reader is referred to the comprehensive Appendix with the details of the different assessments and arguments. In the Section 4.2 the choice of MFA methods included in the review is justified. The assessment results are presented in Section 4.3. The issue of detoxification and policy analysis is brought into play in Sections 4.4 and 4.5 before Section 4.6 reviews some available resource flows models for the UK in the light of the findings.

In Section 4.7 general insights from the review will be summarised. These will be related to the potential of MFA methodologies to answer the policy questions DEFRA raised in the tender in Section 4.8. From there Section 4.9 moves on to explain how the government can apply the different methodologies to derive policies on different levels and Section 4.10 frames this discussion into the specific UK SCP context. The outcome is a framework that allows informing a strategic material flow policy approach, which tackles the physical flows with the most important environmental impacts first, appropriately. Section 4.11 demonstrates how such a strategic approach can be used to prioritise action on the SCP agenda by providing examples in three policy areas. Finally Section 4.12 responds to a particular request made by DEFRA in later stages of the project and prioritises materials according to their environmental impacts based on results of a study by Van der Vart et al. (2005). It then proposes a framework how these key flows might be best traced.

SCP001 24/130

3.2 Choice of MethodsThe methodologies included in this report have been chosen through a two-step process. First, after a comprehensive review of the literature, general MFA methodologies have been identified on the aggregation levels in the economic as well as in the material sphere. However, there is a vast variety of more specific MFA models available. In a second step, we have therefore picked some of these approaches based on their importance to the UK’s SCP Agenda, in general, and to DEFRA’s key SCP interests expressed in the course of the tender for this project. That is, can the approach in question:

1. Illuminate which materials/products/sectors should be targeted to achieve the greatest environmental benefits?

2. Show where is the best point in the supply chain to achieve maximum environmental benefit?

3. Identify, set and monitor targets for SCP policy? Does it identify whether greater reductions in environmental impacts are achievable from product design (and the subsequent changes in production) or from improving resource efficiency in the production process?

4. Assess the environmental impacts prevented by waste recycling, re-use and minimisation?

Under these considerations, methodologies in three groups have been identified to complement the general resource flow approaches. Firstly, due to the high importance assigned by DEFRA to specific information about products, sectors, supply-chains etc., and due to the key role of lifecycle considerations in the government’s SCP strategy, three hybrid input-output methodologies that tie together features of the more general approaches have been picked: these are Environmental Input-Output Lifecycle Assessment, Hybrid Input-Output Lifecycle Assessment, and Waste Input-Output Analysis.

Secondly, due to their popularity on the policy level and their controversial status in academia, a group of MFA approaches that directly refer to sustainability criterion have been included. These are Environmental Space and Ecological Footprint Analysis.

Finally, four available UK specific resource analysis methodologies are discussed: REAP, REEIO, Stepwise and Surrey’s Resource Flow Accounting Methodology. Table 4.2 provides an overview of the included methodologies.

In the course of the review practical knowledge was of great advantage for discussion and judgement. In areas where this practical experience was lacking the assessment was carried out to the best knowledge. Moreover, feedback from MFA experts was sought in a consultation process organised through a project workshop for which an Interim Report with preliminary results was distributed.

SCP001 25/130

Category Methods included Representation in Report

General Resource Flow Analysis Methods

Economy-wide Material Flow Analysis Biffaward Bulk Material Flow Analysis Physical Input-Output Analysis/ NAMEA/

Environmental Input-Output Analysis Substance Flow Analysis Lifecycle Inventories

Individual discussion;

detailed assessm

ent

Specific Methods with direct sustainability reference

Ecologicall Footprinting Environmental Space

Joint discussion;

generalassessm

ent

Specific Hybrid Methodologies Waste IO Hybrid LCA Environmental Input-Output LCA

Jointdiscussion;

generalassessm

ent

UK resource flow models Stockholm Environment Institute - REAP Cambridge Econometrics - REEIO Best Foot Forward – Stepwise University of Surrey – Regional Material

Flow Accounting Model

Joint discussion;

generalassessm

ent

Table 4.2 - Material Flow Methodologies included in this review

3.3 AssessmentThis Section presents the main results of the review of the different methodologies. The review of the general MFA methodologies was carried out using an adjusted version of the assessment matrix introduced in Section 3. Most importantly some of the criteria were tailored towards the more general context of this part of the review and a scorecard system was developed for the assessment of the policy relevance. A comprehensive description of the assessment procedure is presented in Appendix I.1. The specific MFA methodologies have only been assessed contextually with view to their contribution to the government’s SCP agenda for the following reasons:

In the case of MFA methodologies with sustainability reference it was sought to avoid interference with other review projects commissioned by DEFRA (see DEFRA, 2005c);

Hybrid input-output methodologies as well as the UK specific MFA models closely relate to the general MFA methodologies, which are reviewed in detail.

Throughout the assessment the terminology established in the Eurostat Guide for Economy-wide Material Flow Accounting (2001) is used for distinguishing material flows. This terminology is summarised in Appendix I.1 together with a first summary of major general characteristics of the different MFA methodologies.

Some general comments should be added concerning the result presentation:

The main reports only contains the summary of the methodological reviews. The comprehensive assessments can be found in the Appendix.

For summary Sections with an Appendix, referencing is avoided in the main report.

SCP001 26/130

3.3.1 Economy-wide Material Flow AnalysisIntroductionEconomy-wide material flow analysis (EMFA) can be seen as an attempt to systematically measure and understand the overall physical size of a country’s societal metabolism, in tons, for a given reporting period and within clearly defined geographical boundaries (usually nations, regions etc.). The measurement is carried out at the system boundaries between the environment and the economy:

EMFA includes all flows out of and into the natural environment as well as imports and exports.

The physical flows of goods and services within the economy remain a black-box.

EMFA accounts usually try to cover all direct and indirect used and unused physical flows (see Appendix I.2).5

The ultimate aim is usually the provision of a set of indicators at the highest aggregation level that allows the description and the analysis of the physical system at the input and output side of the economy (see Eurostat, 2001). Box 4.2 provides a selection of general policy questions, which can be informed via EMFA methodologies. The full assessment can be found in Appendix I.3.1.

What is the size of the physical system supporting the UK economy? What is the composition of physical inflows and outflows – what materials drive the total weight? How much of a particular resource input is used by the economy and how much of a particular waste/pollutant is generated? What share of the total material inputs into the UK economy stem from reuse and recycling? How much resource inputs can be saved?

Is the physical size of the UK economy increasing or decreasing? Is a decoupling of aggregate resource use and economic growth observable in the UK? Is this decoupling of relative or absolute nature? Could the amount of municipal waste in the UK be reduced over time? Is there an increase in reused and recycled material input in the UK?

How much does the UK rely on the resources provided by its own territory? Is the physical trade balance of the UK positive or negative?

Box 4.2 - Policy questions economy-wide material flow analysis

Summary – assessment resultsEMFA can play an important role in informing the government’s SCP agenda. The review of its methodological robustness and policy relevance shows that based on a sound and well-established methodological framework, the unique appeal lies in its comprehensive set of accounts, which provides an overview of the physical basis and requirements of all economic activities taking place within a system over a defined period. Only based on such a set of accounts can changes in the composition of physical flows induced by SCP policies be comprehensively monitored and undesirable side effects be picked up.

With Domestic Material Consumption (DMC) officially endorsed in the government’s basket of decoupling indicators, EMFA already directly informs the government’s SCP agenda by providing information about overall developments in system-wide resource use, resource productivity and the decoupling from economic activity. However, the assessment has shown that great care needs to be taken in the interpretation of such aggregate indicators derived from the EMFA account. This is where flow size and impact per unit flow tend to be inversely related. In this context a very careful 5 This is to say that the Eurostat guide (2001) recommends such comprehensive flow coverage. However, it

also acknowledges that many studies in practice, for example, do not cover indirect used and unused flows. For the EU-25 such comprehensive accounts have been established by the Wuppertal Institute (see Eurostat, 2002; Moll et al., 2003)

SCP001 27/130

interpretation in terms of potential environmental pressure6 might be justifiable in the sense that a bigger system size has more potential to create environmental stress – without saying whether or not this is actually the case.

However, a clear (positive) relationship between flow size and environmental pressures would be required for the indicators to inform SCP policies directly. With the government’s emphasis on the development of strategic SCP policies, which tackle the flows with the most significant environmental impacts first (see DEFRA, 2003), engaging directly in economy-wide material flow policies by setting (factor 2, 4 or 10) reduction targets for aggregate material flows (based on these aggregate indicators) does not seem to be a viable policy option.

Instead target setting for SCP policies should be problem oriented. Most often then indicators will need to focus not on aggregate, but specific material groups of relevance. Obvious candidates would be wastes or fossil fuels. EMFA accounts and indicators can then be used to pull out the relevant information and construct tailor-made indicators according to the government’s policy needs . The Japanese government provides a good example. Instead of just devising aggregate resource flow targets Japan firstly devised a material flow policy. Based on the vision of a sound material-cycle Japanese economy, priority areas for manufacturing and waste management were established. Indicators where then designed to monitor progress on a structural level and targets were defined. The “Cycling Use Rate” indicator, for example, measures the share of reused and recycled materials of the total material input and progress towards a 14% target by 2010 (2000 level < 5%). Applied in such a systematic way and tailored to policy needs EMFA accounts can play a vital role in informing SCP policy.

The role of aggregate EMFA indicators can then mainly be seen as a general one for monitoring overall trends in the physical system complementing more detailed indicators, which are more appropriate to inform, and tracing progress of more specific policies. The methodological assessment further discouraged the use of hidden flow estimates7 such as Total Material Requirement (TMR) or Total Material Input (TMI) due to the high level of uncertainty surrounding these estimates. Moreover, the relevance of information on hidden flows for informing UK SCP policy remains unclear. For DMC as a consumption indicator (which does not comprise hidden flows) the inclusion of indirect used foreign flows is recommended. However, to ensure international comparability they should be shown as a separate part of the indicator.

The biggest drawback in terms of the policy relevance of EMFA is the lack of waste data, as well as emissions to water. Filling this data gap will be crucial for comprehensively informing the UK’s sustainable waste policy agenda. This data gap is general in nature, and not only relevant in the context of EMFA.

In terms of future directions in using the available evidence there are two main development options relevant to SCP decision-making.

Firstly, there seem to be a lot of unexplored analytical options mainly associated with the use of more advanced statistical methods. There is enough longitudinal as well as cross-sectional data available, which can be further exploited for the benefit of SCP policy making. This also includes the analysis of the EMFA data together with economic and social variables.

6 The term “environmental pressures” is used as long as no relationship to particular environmental problems is established. In contrast, “environmental impacts” will always imply the relationship to at least one environmental problem.

7 “Hidden flows” are the unused materials associated with the extraction of raw materials, domestically and abroad, which never enter the economic system.

SCP001 28/130

Secondly, there is still the need to establish clearer links between the various material groups represented in the EMFA accounts and environmental problems. While these links are well established for climate change, these need to be established in many other areas. Encouraging first attempts have been made linking material flows and environmental impacts (see De Bruyn et al., 2004; Van der Voet et al., 2005), which should also be encouraged in a UK specific context. This is the most important development option: the closer the link between environmental problems and the flows of materials in the economy, the more readily the information can be applied in the policy process. The results are summarised in Table 4.3.

Unique Contribution to UK’s SCP toolkit

Size and composition of human-induced physical flows Comprehensive control for material implications of policy

Limitations Robustness of hidden flows Lack of physical output data Under-representation of most damaging substances Aggregation level of indicators – policy relevant interpretation difficult

Improvement Potential/Recommendations

Utilise rich EMFA database by devising tailor-made indicators in area of relevance such as waste, material cycling etc.

Inclusion of indirect used foreign flows in DMC estimates Waste data improvement Encourage attempts to link material flows closer to environmental

problems

Table 4.3 - Summary economy wide material flow analysis

3.3.2 Bulk Material Flow or Material Systems Analysis – Assessing the methodological framework of the Biffaward studies

IntroductionIn this Section bulk material flow (BMFA) or material system analysis (MSA) are reviewed. These are MFA methodologies, which are the most heterogenous among the ones reviewed here, and have not been standardised to any extent. “It is rather a composition of certain MFA applications and case studies that are not characterised by a unique methodology, but by a common meso-level of aggregation” (Femia and Moll, 2005: 34) in the human as well as in the material sphere. The majority of the Biffaward Mass-Balance studies can be grouped into this category. Due to the lack of a common methodology for BMFA/MSA approaches and as a preparation for the review of the robustness and policy relevance of the individual mass balance studies in Section 5 of this report, the methodological framework in the “data co-ordination guide” by Linstead and Ekins (2001) and its follow-up (Lindstead et al., 2003) will be assessed for its contribution to robust and comparable study outcomes. However, it needs to be emphasised in this context that the Mass Balance Programme started in 1997 and a number of studies were already on their way when the guidelines were published. For the comprehensive review of the methodological framework see Appendix I.3.2.

Summary – Assessment resultsThe methodological framework of the Biffaward programme provided by Lindstead and Ekins (2001) and Lindstead et al. (2003) was intended to ensure consistency and compatibility of the data collection efforts across the more than 50 different projects. It was hoped that through the

SCP001 29/130

establishment of a systematic, regulated approach for the data collection process the results could be integrated in an overarching data structure providing a comprehensive cross-sectoral and geographical picture of resource flows in the UK. This would increase acceptance, applicability and policy usefulness of the studies.

Despite the methodological freedom required within the Biffaward scheme, the data guide managed to set rules for a clear definition of system boundaries in the material as well as human sphere . On the project level methodological freedom was crucial, because policy relevance was reinforced within the Biffaward Porgramme through a particular tailoring of funding requirements – funding required an additional 10% from an outside body. Considering that the Biffaward Programme was set-up on an ad-hoc basis, this seems to have significantly contributed to the policy relevance of the various studies. Moreover, clear recommendations for the establishment of material balances, and how to avoid double-counting, should have positively contributed to the quality of the individual studies.

On the downside, the methodological framework gave little consideration for other practical problems in the data compilation process, from the definition of a set of standard data sources to the suggestion of helpful accounting conventions. Admittedly this was complicated by the wide scope of the programme and the impossibility of foreseeing the nature of future projects. Finally, the material classification was not fully defined in the guides. The conclusion is that the methodological framework provided a workable, though not a detailed, platform from which studies could start.

Overall, it must be said that the Biffaward studies certainly had an impact. With respect to the major aim of the Biffaward Programme - demonstrating the need and value of the mass balance approach - the project has certainly been successful. However, in terms of the provision of compatible research outcomes, in has been much less successful. If the quality of a study is bad - even though it followed the instructions of the guide in terms of system boundary setting, classifying the data and general accounting procedure - the shortcomings might be rooted in the lack of attention to practical problems in the guide. However, if the studies did not follow the instructions, it might have been a problem of project management and enforcement in the guidelines. The establishment of rigid reporting guidelines (see Section 4.3.5 on SFA) is key, if the government aims to commission another series of studies similar to the Mass Balance Series.

Methodological Strength The methodological framework of the Biffaward studies provided the potential to provide tailor-made MFA studies on various levels in the material as well as the human sphere to provide answers to policy relevant questions.

Funding strategy designed to trigger policy relevant research.Methodological Limitation Lack of practical advise – in particular with respect to the exploitation of

data sources Incomplete material classification

Improvement Potential/ Recommended Action

More rigorous regulation to ensure consistency between studies Provision of more practical guidelines if similar projects are intended in

the future

Table 4.4 - Summary bulk material flow/ material systems analysis

3.3.3 Analysis of Material Flows by Sector: NAMEAs, Generalised Input-Output Models, and Physical Input-Output Analysis

Introduction

SCP001 30/130

This part of the review has focused on environmental input-output analysis, which can be based on two types of tables: monetary input-output tables (MIOTs) and physical input-output tables (PIOTs). They can be used in combination with NAMEA type environmental account data in generalised input-output analysis (GIOA) and physical input-output analysis (PIOA) to build models to estimate physical flows through the economic network from entry to release. Such environmental input-output analysis8 (EIOA) models usually show the following features:

high degree of sectoral detail - allows the detailed study of sectoral supply-chains and resource productivity;

production and consumption entities unified in a systemic approach - goods and services delivered to final demand entities can be traced back the production-consumption chain in a lifecycle perspective;

physical data typically used on various aggregation ranging from individual substances to aggregate material flows.

A set of relevant policy questions, which can be answered with such models is summarised in Box4.3. The comprehensive review results are outlined in Appendix I.3.3.

What are the absolute and relative direct contribution of the individual industries to environmental pressures? What are the technological, and behavioural driving forces behind changes in the use of resources and the

generation of wastes? How resource efficient are the different industries? Which resources, and in what quantities, are consumed by an industry? How much emission and waste are

released by different institutions? Where do the most significant resource flows occur in the supply chain of a certain product (group)? Which product groups drive waste flows associated with household and government consumption and where might

be the best intervention points in the supply chain? What are the total CO2 emissions associated with services? How does inequality in a society contribute to resource use patterns? Is the UK economy a throughflow, or a cyclic, material system? Is the flow behavior of materials in the British

economy increasingly cyclic?Box 4.3 - Policy questions environmental input-output analysis

Summary – Assessment ResultsEIOA is a valuable tool for informing SCP policy. Its capacity to model the pathways of materials (substances, bulk materials, aggregate material flows) through the complex economic supply chain from production to consumption with great detail, and to assign these physical flows to the sector (or product groups) of final use based on a consistent accounting framework, allows for a high level of responsiveness to the government’s SCP agenda.

EIOA models have been frequently used

to calculate the total resource use triggered by different final demand categories (e.g. household consumption, government consumption, capital investment, exports),

8 In the context of this review EIOA refers to both GIOA and PIOA models. If specific reference to one individual model is made, the terms GIOA and PIOA will be used.

SCP001 31/130

to assess the material use patterns associated with different lifestyles in the UK, to benchmark sectoral environmental (as well as social and economic) performance,

to identify key product groups in terms of resource use and waste generation,

to unravel the physical supply chain in the search for appropriate policy intervention points for reductions in carbon emissions,

to analyse the driving forces behind changes in emissions between two reporting periods, and,

to assess the material implications of UK consumption patterns abroad or to compare the efficiency of industrial sectors in the conversion of natural resources over time and across industries.

For building up the evidence base, and for informing the government’s SCP agenda, EIOA methodologies have four particular attractions:

They provide a complete picture of the economy’s resource flow under a detailed coverage of the supply chain on a meso-level. This makes EIOAs generally suitable for identifying the environmentally most relevant economic activities and therefore prioritising key policy areas, as well as building up SCP evidence;

They allow an estimate of, not only the direct resource use on-site, but also the indirect resource use higher-up in the economic supply chain. Therefore, they also have the particular strength of taking an international perspective and quantifying the full resource use implications of domestic final demand patterns abroad (see Section 6);

MIOTs are a standard economic accounting tool, which can be linked to a wealth of other socio-economic statistics (e.g. Keuning, 2000). Moreover, as an economic statistic, which serves as a data inputs for many macro-economic models, they are a good starting point for an integrated inquiry, which does not only take into account environmental, but also social and economic issues. This, for example, allows for the informing of the SCP agenda about resource use in the context of economic growth, employment and/or inequality, and for an analysis of the relationship between the issues.

EIOA models are very flexible at integrating other (bottom-up) data sources in the most appropriate measurement units. As it will be shown later, this makes them the key for overcoming some of the weaknesses of their own and other methodologies, through methodological integration and strengthening the robustness of SCP evidence.

A limitation of input-output models is the sheer amount of information, which is often generated and provided to policy makers. Good practice needs to be established to ensure a clear focus and a better packaging of results. (For best practice see Appendix I.3.3).

The underlying methodology is well-established and sufficiently robust for policy advice. While PIOTs are (ideally) compiled in a bottom-up data collection process, 9 GIOAs use the monetary production structure to impute the pathways of physical flows from production to consumption assuming that the deliveries of products to other sectors always trigger the same amount of resource

9 In practice the use structure is usually derived from monetary data even though an extremely low level of aggregation is used (see Stahmer et al., 1997; Pedersen, 1998) – in the case of Germany more than 5000 product groups are mapped into more than 100 industries. On such a low level of aggregation the relationship between the value of goods and their weight is very close. Moreover, price differentiation is largely controlled for.

SCP001 32/130

use and wastes per unit of output. In this sense they do not observe, but model the intersectoral flows in the economy.

This assumption is the more reasonable, the more homogenous the sectoral output and the lower the aggregation level is. Even though a more detailed breakdown would be desirable, the 76 sectors available for most environmental account data can be seen as sufficient for providing policy advice. From a methodological perspective PIOAs should therefore provide more reliable results than GIOAs.

However, the vast amount of GIOA literature indicates a sufficient level of confidence in the research community into the reliability and robustness of the method for such ex-post analysis. In the light of the unavailability of PIOTs in the UK and the relatively high costs associated with their compilation, the question of choice between GIOA and PIOA models for informing policy, therefore, comes down to whether or not they lead to significantly different policy implications.

However, this question has not been comprehensively answered in the literature. The debate surrounding PIOTs is still very new, and has so far mainly focussed on methodological issues surrounding the treatment of wastes in physical input-output models. In the light of the general lack of resources for the compilation of the MIOTs the recommendation is to postpone the compilation of tables until their importance for informing policy has been established.10

This points towards the major limitations of EIOA applications in the UK: the lack of current data. The most recent analytical MIOT published in the UK is from 1995. This means that some of the results generated from GIOA models are based on the production structure from more than 10 years ago. Only the Stockholm Environment Institute and Cambridge Econometrics (see Wiedmann et al., 2006) have managed to update input-output data to 2003, even though the resulting data will not be of the same quality as that produced by ONS. Due to the large number of policy relevant applications available and the potential to provide cost-effective research, it is of the greatest importance for the government to improve this situation. Different solutions with different amount of financial resources required are possible:

Encourage the provision of sufficient data to easily update input-output tables for every year since 1992;

Encourage the over-due provision of the input-output table for the year 2000 and the automatic updating of the table;

Encourage the establishment of a time series of input-output tables (as is common in many other Eurpean countries) as done for SUT publication;

If further resources are available, or the policy need for physical tables is demonstrated, promote the construction of PIOTs;

The extension of the NAMEA type environmental account data should be considered.

10 There are sufficient tables available in the international community to establish the policy relevance of using PIOAs instead of GIOAs.

SCP001 33/130

Unique Contribution to UK’s SCP toolkit/ Strength

Lifecycle view; direct and indirect resource use Can take into account the full supply chain interactions with a reporting

period Allows the assessment of full burden shifting Integration potential with other methods, models and data sources Can help prioritising evidence-build-up

Limitations Data for the UK very outdated Physical flows in the supply chain imputed Amount of results generated sometimes detrimental for clear analysis

Recommended Action Promote more timely and time series MIOT publication Promote integration with other MFA tools (see Section 4.3.6) Encourage provision of more comprehensive environmental data Use for prioritising SCP policies (identification key sectors, key product

groups etc.)

Table 3.4 - Summary Environmental Input-Output Analysis

3.3.4 Lifecycle InventoriesIntroductionIn the context of the focus on methodologies for modelling and analysing physical flow, this part focuses on life cycle inventories (LCI). These form the second of four stages of a full lifecycle analysis (LCA) and have been standardised together with goal and scope definition through the ISO 14041 standard. In LCIs the product system is modelled, data gathered, and then used to compile the data for each part of the system. Based on functional system boundaries, LCIs aim to establish a systematic and comprehensive inventory of all physical flows in a bottom-up approach. They therefore involve the identification and quantification of physical input and outputs during each lifecycle stage. The holistic nature of the lifecycle approach requires us to not only consider the direct flows associated with the various lifecycle stages, but ideally also the indirect flows occurring higher up in the supply chain. A selection of policy question that can be answered with the help of LCIs can be found in Box 4.4. The full assessment can be found in Appendix I.3.4.

Where are (important) physical flows triggered along the lifecycle of a product? Which phase – extraction, processing, distribution, use, wasting – is most relevant for policy intervention? Where are the most relevant policy intervention points?

Which of two similar products is environmentally preferable in terms of the physical flows triggered across the lifecycle?

Which materials should be used to maximise the environmental performance of a product? In which stages of the lifecycle of a product are the largest potentials for maximsing the environmental performance

of a product? How can the life cycle impacts of a product be reduced by design?

Box 4.4 - Policy questions LCI/LCA11

Summary – Assessment resultsLCIs can contribute to informing SCP policies in various relevant areas such as sustainable waste management, integrated product policies or as input for consumer information systems. In these capacities the information provided might inform the setting of product standards, contribute to waste minimisation strategies, lead to more informed consumption decisions through their contribution to labelling schemes, or even motivate bans of materials for use within the lifecycle of product systems. There is a whole array of additional applications on the company level, where LCIs can contribute to

11 While environmental impact assessments and their limitations have not been introduced yet, some of the questions are phrased towards full LCAs.

SCP001 34/130

improve the environmental performance of a specific product system by informing the selection of process technologies, product design as well as process optimisation. Due to the project specific focus and the detailed information provided by LCIs on the micro-level it is relatively straightforward to establish ex-post links with policy implementations.

Their close relationship to LCAs allows LCI evidence to be readily expressed in terms of their contribution to particular environmental problems and to be further used in risk assessments. Moreover, as products in certain parts of the world (Western Europe) can be assumed to trigger a very similar amounts of physical flows, and consumers in the UK buy an “international basket of goods”, available LCI evidence from many different parts of the world can be used to feed into the government’s SCP evidence base.

The review of the methodological robustness shows that LCIs have been successfully standardised during the last decade and careful procedures have been established for example in the appropriate definition of the study system, the associated system boundary choice and permanent feedback processes including a terminal sensitivity analysis. Ideally, LCIs would collect a complete set of physical flow data associated with a particular product system. In practice this is an unmanageable task due to the system complexity caused by the (global) interconnectedness of activities of such a cradle-to-grave- perspective. Hence, as most of the other bottom-up methodologies LCIs suffer from an incomplete system coverage and they must carefully choose a set of activities covering the majority of physical flows associated with a product system. This has three major implications:

Due to this truncation error LCIs consistently underestimate the „true“ physical flows“ of a product system and must therefore be seen as lower bound estimates;

This underestimation seems to be significant12 and tends to be higher for product systems with complex supply chains and high proportions of indirect flows of materials and energy;

Even though the ISO standards provide detailed guidance in establishing system boundaries, the cut-off points will differ from study-to-study. This limits the comparability of results across studies. This is particularly relevant for the government’s collection of evidence from LCA/LCI studies for SCP consumer information services.

As good practice, this problem of comparability through the choice of cut-off points can be partially overcome if EIOA type models are used for Structural Path Analysis to identify the supply-chain nodes in the life cycle with the widest system coverage. To fully overcome this problem an environmental input-output lifecycle assessment, or a hybrid input-output lifecycle assessment model, needs to be carried out. These methodologies will be more closely examined later and the associated problems will be outlined.

With data collection being one essential part of the methodology, LCIs usually use most recent data. While the data exchange has already been standardised in ISO 14048 through a common data documentation format, the actual data collection methods are not fully detailed by ISO. However, a protocol is already under discussion, which tackles these problems based on a common ISO 14041/49 standard. Finally comparability of studies is further aggravated through the reliance of the data compilation processes on LCI databases. These databases all differ in their coverage of products and materials, in the origin of the source data, and the time they cover, and therefore ultimately in their quality.

12 Lenzen (2001) suggests that for most commodities, direct energy requirements account for less than a quarter of total energy requirements. Process analyses include 132 first-order inputs that carry truncation errors that are up to 50%, and 17424 second-order input paths that 30% truncation error.

SCP001 35/130

With its cradle to grave perspective, and detailed information provided on resource use and waste generation throughout a product system‘s lifecycle, LCI/LCA can provide unique information to the government‘s SCP agenda. However it should be noted that,

a minimum requirement, for the government‘s efforts to pull together LCI/LCA data for consumer information services, would be the meticulous recording of which database is used. Even though this information is not relevant to the user of the information service, it allows for a better control of data quality and comparability;

if the government intends to make intensive use of life cycle assessment data, a review of the various available LCI/LCA database should be commissioned. This could exclude the ones, which provide insufficiently robust data, and champion the ones which are most reliable and appropriate for use in the UK. The result could be a shortlist, provided by the government, of which LCI databases can be used for feeding into the government‘s SCP evidence base.

Unique Contribution to UK’s SCP toolkit/ Strength

Cradle to grave perspective Direct links to environmental impact and risk assessments Many direct policy applications for policy makers and companies Often directly used by policy makers Close links to comprehensive impact assessment and risk analysis Only MFA methodology which can comprehensively inform eco-design

processes Usually uses/collects most recent bottom-up data

Limitations System boundary choice arbitrary Comparability of results limited Identification of environmentally preferable products limited without

performance of environmental impact assessment Recommended Action Promote the use of hybrid LCI/A in future studies to overcome system

boundary problem (see Section X) Consider review of LCI databases Record the database used in studies for building up information services

based on LCI data

Table 3.5 Summary life cycle inventories

3.3.5 Substance Flow AnalysisIntroductionSubstance Flow Analysis (SFA) is a bottom-up methodology, which sets out to follow a single or a group of substances in a predefined system – usually a geographic region. SFAs trace the pathways of these substances through this system from origin to destination and identify where they accumulate. The information generated in this process is usually used to provide an overall management strategy for these substance flows (Femia and Moll. 2005: 42). Some typical features are listed below:

SFAs usually start from a particular “environmental” (or human health) problem and typically focus on highly toxic or other hazardous substance flows. Good examples are lead, chromium and heavy metals.

SCP001 36/130

SFAs are usually not restricted to the human sphere. They often also trace flows back into the environment and account for accumulations and resulting biological transformations in the environment.

SFAs are often carried out at the meso (sector) level in the human economic sphere.

A list of typical policy questions, which can be answered by SFAs are found in Box 4.5. The full complete assessment can be found in Appendix I.3.5.

Through which economic activities does a particular substance enter the human sphere? Where is it released back to the environment?

What are the pathways of a substance through the defined system? What are the most relevant production processes and in which products do harmful substances end-up?

Where are the most effective policy intervention points for eliminating, reducing or minimizing a particular substance in the supply-chain? Which are the most important economic sectors in terms of substance use?

How does a specific substance contribute to a particular environmental problem?

Box 4.5 - Policy questions substance flow analysis

Summary – Assessment ResultsSubstance Flow Analysis (SFA) is a tool of direct relevance for the government’s SCP agenda and can help to answer a variety of key policy questions. It is a useful and flexible tool for tackling specific substance related environmental problems and can provide valuable policy advice for the elimination, reduction, minimisation and recycling of specific substances in the course of comprehensive substance management strategies.

Because SFAs frequently – even though not necessarily13 - focus on low volume-high impact flows, they are often better seen in the context of detoxfication rather than dematerialisation policies. The information provided by SFAs is of direct relevance for informing SCP product policies, where the reduction of environmental impacts from hazardous substances is a major concern (see DEFRA, 2005). SFAs cannot only tell in which products a certain harmful substance is embodied, but also identify where the most effective policy intervention points in the supply chain are. In this course it is a particular strength that SFAs usually establish links between the flows of materials and substances. Further application of SFAs are on the de-coupling agenda. While the substance specific accounts by ONS and the Environmental Agency might be able to monitor whether or not decoupling is taking place, SFAs can help to inform how de-coupling can be brought most effectively through a set of policies.

Much of the appeal of SFAs is rooted in the fact that the flows of substances are closely related to environmental problems. As soon as particular environmental problems associated with a limited number of substances become of interest on the SCP agenda, SFAs are an appropriate tool for informing SCP policy. They can therefore be applied very much in line with the government’s emphasis to focus on the most significant environmental impacts first.

SFAs are built on a flexible bottom-up methodology. On the one hand, this is a strength as it allows the practitioner to respond adequately to particular problems by tailoring the system boundaries that are most appropriate and using the most recent data sources can be used. On the other hand, this

13 They can be applied to any substance /substance specific problem. The two substance specific studies in the Biffaward series, for example, focussed on nitrogen and carbon.

SCP001 37/130

means that the quality of individual studies very much depend on the good practice and experience of the practitioner in choosing appropriate system boundaries as important physical flows can be easily neglected.14

The various studies available in the literature are very difficult to compare as they ultimately all vary in the particular choice of system boundaries (in the human and material sphere) and quality. This problem is reinforced through the lack of methodological standardisation even though current leading researchers are pushing towards greater standardisation (e.g. Brunner and Rechberger, 2004; De Haes et al., 1997). However, the ex-post usage of individual SFA studies for SCP policies might still be appropriate as long as the system boundaries of the study correspond with the ones required for an appropriate assessment of policy problem under consideration. This must be decided on a case by case basis.

If the government wants to monitor and control specific problem substances in the course of its SCP policies, SFAs provide a valuable information tool. In the course of the build-up of the SCP evidence it will ultimately remain one of the challenges for the government to establish good practice in the way a set of relevant target substances might be monitored to ensure uniformity and comparability in results. This will also improve the chances of integrating data with other physical flow evidence.

Denmark provides an example of a good national substance management strategy. The Environmental Protection Agency (EPA) commissions SFAs, which have been identified as actual or potential hazardous to humans and/or the environment.15 The results of SFA are used for by the EPA and related to the need for instruments, risk minimisation for the substance in question and for the monitoring of policy success in controlling the substance. For the latter some SFAs are updated regularly. During the 1990s the Danish EPA prepared its own reporting guidelines to ensure uniformity and comparability of SFAs undertaken in Denmark. These guidelines inform methodology, data sources, reliability and confidentiality and present a standard format for the reports.

A further limitation is the focus of SFAs on a small selection of substances. This means undesirable substitution effects on other harmful substances and environmental problems cannot be fully controlled and implys the need for the integrated use of SFAs with other physical accounts and MFA methodolgies.

Unique Contribution to UK’s SCP toolkit/ Strength

Trace specific (or group of) substance(s) through a defined system to tackle specific environmental problem

Establishes links between substance and material flows Detailed focus on a specific environmental pollution problem Very likely to trigger direct policy action

Limitations System boundary choice arbitrary Lack of comparability of results across studies – problems for ex-post

usage Narrow range of material flows covered gauges the net effects of

substitution by other substances and related shifts in flows and accumulation in the human economy and through environmental media.

Indirect links with monetary data

14 Moreover, the bigger the system size, the more difficult it often is for such bottom-up methodologies to cover and allocate all relevant physical flows appropriately. Quality of the studies is further heavily influenced by the comprehensivness of the data collection efforts being very much dependent on the available funding.

15 Existing substance flow accounts held by ONS and the Environment Agency usually monitoring the inflows or outflows of substance from or into the Environment can further inform this identification process.

SCP001 38/130

Usually uses very recent up-to-date dataRecommended Action Use when substance related environmental problem needs to be

effectively tackled Promote integration with other methodological frameworks were suitable Establish good practice: Develop standard reporting and accounting

approach for SFAs to ensure uniformity and comparability of results

Table 3.3 Summary Substance Flow Analysis

3.3.6 Integrating approaches – Hybrid MethodologiesThe different methodologies have all shown particular strength and weaknesses. Bottom-up approaches such as LCI, BMFA/MSA or SFA often work with very accurate and detailed information, but have problems in covering all relevant physical flows throughout the supply chain. In contrast, EIOA as a top-down approach provides a very complete picture of the domestic supply chain at the meso level, but models physical flows on a rather aggregate level based on the assumption that for each sector the amount of pollution/resource use triggered is the same per unit of output produced. Linking bottom-up16 and top-down approaches therefore might allow the combination of the strengths of both. The potentials for integrating the different methods are summarised here. First, an explicit focus will be given to alternative approaches for compiling LCIs using input-output data. Second, these insights will be generalised for integration with other methods. A full discussion can be found in Appendix I.3.6.

Integrating input-output and LCI approachesTo overcome the system boundary problem of conventional process based LCIs and to increase their comparability, environmental IO lifecycle inventories (EIOLCI) have been proposed as a cheap and time-efficient short cut to LCI data. Methodologically they generally do not differ from conventional EIOAs. However, they usually apply much more detailed input-output tables and much more comprehensive physical flow/pollution data inventories. Even though they can overcome the truncation error problem associated with the domestic supply chain17 through the use of top-down input-output data, they suffer from a variety of other limitations such as

their aggregation level;

their lack of any comprehensive coverage of use and end-of-life stage of product systems;

the delayed publication of input-output data (see Appendix I.3.3 for details).

Therefore, EIOLCIs are incomplete in their coverage of the lifecycle and cannot address individual products, but only larger product groups. Even though it will be shown later that they might have their applications to inform specific SCP agenda items, they do not seem to be an appropriate alternative to process based LCIs. This is particularly true in the light of the relatively high aggregation level of the UK’s input-output data, its outdated reference period (most recently 1995),18

and the sparse availability of environmental account data in the UK for LCI purposes. Therefore, it 16 Bottom-up data and process data will be used interchangeably in the course of the discussion of LCIs.17 For physical flows occurring in other countries, it depends on the model chosen to what extent EIOLCA

models can cope with the problem. The interested reader is referred to Section 6 for a comprehensive discussion of the problems associated with the quantification of import related physical flows.

18 It has been highlighted previously that only Cambridge Econometrics and SEI York have updated the IO data for more recent periods. Even though this seems to be preferable to the use of older data, it introduces new error sources. The need for the overdue update of the IO publication, which would resolve such problems, is therefore re-emphasised.

SCP001 39/130

might be said that as a minimum requirement hybrid IO LCI approach must overcome the aggregation problem and must fully cover the life cycle stages of product systems . In the literature three hybrid input-output life cycle inventory19 approaches can be found:

Tiered input-output LCI (TIOLCI): TIOLCIs combine bottom-up data from traditional LCIs for the end-of-life stages to cover the full lifecycle of a product. For key upstream flows further physical data is often entered directly into IO framework. However, they are still prone to the aggregation problem and do not seem to be a robust alternative to traditional LCIs in the UK SCP context.

Input-output based hybrid LCI (IOHLCI): IOHLCIs use bottom-up data to disaggregate the sectors/product groups in input-output tables according to project needs. The remaining stages in the life cycle need to be added manually as for TIOLCI. Such a methodology seems to be a robust alternative to conventional LCIs.

Integrated input-output hybrid LCI (IIOHLCI): IIOHLCI can fully overcome the truncation error problem associated with conventional process based LCIs without adding new sources of error by fully integrating conventional process based LCI and IO data. In particular, by using a matrix representation for compiling the process data instead of a process flow diagram approach, the bottom-up LCA data can be readily combined with the input-output matrix in a fully integrated model. They are therefore the only hybrid method, which deals with all life cycle stages in a systemic way and can fully combine the strength of top-down and bottom-up approaches. They must therefore be seen as the best way of compiling LCI evidence for SCP policy in the UK.

Therefore, it can be concluded that hybrid IO LCI approaches can provide a feasible way to increase the robustness and/or comparability of conventional LCIs. IIOHLCIs can inform SCP policies in the most robust way, but represent also the most labour intensive alternative among the different hybrid IO LCI methods. IOHLCIs might therefore provide a practical alternative. The choice will depend on time availability and project budget. For IOHLCIs it cannot be fully established that they necessarily provide better results as this procedure introduces new sources of error associated with the sectoral disaggregation. However, Lenzen (2001) discusses these various error sources and shows that they are likely to be smaller and more controllable than the large truncation error found for conventional LCIs. It might therefore be concluded that they are at least as robust as conventional LCIs. In such a case they still seem preferable as an SCP evidence source as they provide more comparable results due to their roots in a consistent integrated national accounting framework.

The only current restrictions for hybrid LCI methods are the outdated input-output data (see Section 3.3.3) and the limited amount of environmental account data available. This particularly applies to IOHLCIs, which sometimes only enter a very limited amount of physical bottom-up data into the input-output framework. Therefore, good practice needs to be established in the bottom-up data collection process, if applied in the UK context. Moreover, UK practitioner so far have little experience with these rather new LCI approaches. This means that there might be a limited capacity to apply these methods even though a first UK specific software tool will soon enter the market (see Lenzen, 2006).

Under this reservation the increased use of IOHLCIs and IIOHLCIs for compiling life cycle inventories can still be recommended. However, as not every hybrid IO LCI approach is more 19 In the course of the report it will be referred to hybrid IO LCIs when all three different methods are referred

to, while for specific reference the individual names of the different methodologies will be used.

SCP001 40/130

suitable to inform the SCP agenda with LCI evidence, the government needs to be careful about what people “sell” as such. This Section might provide helpful guidance for distinguishing useful from less useful approaches for particular SCP applications.

EIOLCIs or (better) TIOLCIs should not be used for detailed LCIs. However, they could play an important role for informing government as they allow the identification of the most environmentally important product groups. This can help to identify key policy areas and direct the further build-up of more detailed LCI/LCA evidence.

Other hybrid approachesMethodological integration is not only beneficial in the case of LCIs and EIOA. There are further options to increase robustness and/or policy applicability for other MFA methodologies. However, LCIs and EIOAs are usually involved in these integration efforts due to some unique features, of which the most important are:

LCIs are the only methodology with functional system boundaries.

EIOAs provide a full coverage of the supply chain on the meso level, combine data in monetary and physical units and can link bottom-up data into a standard accounting framework.

This can radically increase analytical options and policy applicability, and contribute to increased comparability and robustness of the methods. BMFA/MSA bottom-up data can, for example, be integrated into IO tables. A detailed description of such an integrated approach can be found in Section 3.12. De Haes et al. (1997) show how the policy applicability of SFAs can be expanded by adopting functional system boundaries. Van der Voet et al. (2005) add a new dimension to EMFA accounts by linking via LCI databases to environmental impacts.

Methodological integration is therefore generally seen as a welcome addition to the standard MFA toolkit. The relevance of applying such hybrid methods will often depend on the policy needs. While hybrid IO LCIs are generally recommended for use in the build-up of new LCI evidence, the application of these methods depends much more on the specific policy context. Therefore, in later sections some other key hybrid methods for informing the SCP agenda will be highlighted.

3.3.7 Approaches with sustainability reference pointsAs demanded this discussion of MFA methodologies with sustainability reference will remain very brief and descriptive to avoid any interference with other projects commissioned by the government (DEFRA, 2005a; DEFRA, 2005b). No direct recommendations will therefore be derived. Two methodologies are briefly summarised below:

Ecological Footprint Analysis (EF)

Environmental Space Analysis (ES)

Since the beginning of the sustainability debate there has been a continuous interest in research and on the policy level to answering questions such as: “Are we living within the carrying capacity of the planet?“ or “Are the levels of resource use sustainable?”. These questions are of direct relevance for the government‘s SCP agenda with its continuous emphasis on the need for the UK‘s economy to operate within the means of nature. The title of the SCP Chapter in Securing our Future, A Open Planet Economy, further emphasises the significant of these debates and the direct relevance for the SCP agenda. EF and ES set out to provide evidence for answering these questions. Both

SCP001 41/130

methodologies provide measures of carrying capacity and incorporate equity concerns by expressing resource consumption in terms of global shares.

The Ecolgical Footprint is a measure of the ecological load imposed by a given population. It measures the land area necessary to sustain current levels of resource consumption and waste (incorporating carbon dioxide emissions) discharged by that population. Results are usually expressed in per capita terms and compared with the available area of bioproductive land per global citizen. This allows assessing whether a society lives beyond the natural means under global equity considerations. It therefore provides a measure for one of the government’s key policy questions.

Ecological Footprinting(Wackernagel and Rees, 1996)

EFA was designed as a readily comprehended indicator of the sustainability of the human economy vis-a`-vis the Earth’s remaining “natural” capacity to supply resources sometimes considered equivalent to the planet’s terrestrial “carrying capacity”). It groups and calculates material and energy requirements of nations (or regions) for a limited number of consumption functions, converts these metabolic flows into the ecologically productive land area required to produce the resources used in these activities, and compares the required areas to available regional, national, and global ecologically productive areas. Existing studies have typically been restricted to the ecological resource output potential of terrestrial areas.

Environmental Space (Opschoor and Reijnders, 1991)

The primary function of environmental space (ES) is to quantify or track sustainable development by comparing resource demands with available “environmental space” (closely linked to notions of “carrying capacity”) or the upper and lower physical boundaries of the Earth’s supply of environmental services that are available and can be appropriated sustainably by humans. A primary yardstick is the quantity of various categories of natural resource services that can be exploited by humans over a given time period (usually a year) without compromising the quantity and quality that can be accessed by future generations, that is, without exceeding the “ceilings” or maximum limits based on renewable resource regeneration and the waste assimilation capabilities of the environment. The ES methodology does not attempt to convert all environmental impacts into a single unit or statistic but separately assesses sustainability for energy and raw materials, biological resources, waste assimilation and storage, and fundamental life support functions. A fairly comprehensive coverage of the major human environmental impacts is undertaken by independently analyzing resource service function capabilities or sustainability limits (ideally, the global or appropriate regional average per capita maximum consumption levels consistent with sustainable development). These limits are compared with national per capita use levels for categories such as energy, forests, natural or agricultural areas, water, and major nonrenewable minerals groups.

Table 4.5 - Overview ecological footprinting and environmental space (Daniels and Moore, 2002a)The robustness of the method has been controversially discussed in the literature and recently been assessed for the government elsewhere (DEFRA 2005c). Like all aggregate measures critique is mainly triggered by the fact that the ecological load is aggregated into a single index (see discussion in Section 3.4). Therefore, most disputed is the conversion of MFA inventories into „hypothetical“ energy land, which is required to absorb the emissions occuring throughout the life cycle of the goods and services consumed in a country. The reply to many of the criticisms is that living within the carrying capacity of the planet is an essential pre-condition for living a sustainable lifestyle and at present, the footprint is the only indicator that attempts to quantify this.

While this conflict cannot be easily resolved, it should be noted that the underlying CO2 account used for the calculation is equivalent to what will be introduced as the consumer emission account in Section 6. Moreover, beside the ongoing standardisation process on the one hand20, there has been a lively academic debate taken place, which has proposed alternative methods, which can partially overcome these problems. Most interesting for the government might be an alternative EF indicator proposed by Lenzen and Murray (2001), which is only based on „real land“ and also accounts for land disturbance. Despite the criticism there is little disagreement that the EF is a valuable educational tool. While it should never be used as a stand-alone, it might prove valuable for building capacity in the SCP context when integrated into a set of robust MFA indicators.

20 The Global Footprint Network was formed in 2004 to set about standardising the Ecological Footprint methodology.

SCP001 42/130

ES does not convert the ecological load imposed by various types of resource use into a single unit. Instead it tries to measure the quantity of various categories of natural resoure services and compares them with sustainable levels, which can be exploited by humans over a given time period without compromising the quantity and quality that can be accessed by future generations (DEPA, 1998). The method therefore usually maps current per capita resource use levels of a given population in different units against sustainability threshold levels (ceilings) for different physical flows such as energy and raw materials or biological resources (Table 4.6)

Table 4.6 - Examples: results ES analysis (Spangenberg, 1994)

The gaps between the ceilings and current consumption enables government’s to identify sustainability targets and appropriate strategic policy action. This is very much in the spirit of distance-to-target type approaches and has an immediate appeal for the SCP agenda. However, it is very difficult to establish robust sustainable resource use thresholds. In fact, the ecological literature has discussed the estimations of such thresholds in the context of complex-dynamic systems for a long time. If there has been any consensus, then probably that such thresholds cannot be easily estimated and only be derived in a safe-minimum standard approach. Hence, while the various indicators can be easily measured and provide a robust basis for informing policy, it is the „sustainable threshold“ for the various resources, which is of concern and provides the main appeal for policy formulation.

Finally some words should be directed towards the availability of evidence in the UK. The EF methodology has been heavily applied in the UK and there is a wealth of evidence and models available on the national, regional, local and other institutional level(s), which could directly feed into the SCP evidence base. On the other hand there has been hardly any evidence produced in the ES literature (for an exception see McLaren, 2001). Also in the international literature much more interest has been stimulated by the EF method.

This review has attempted to avoid major judgements regarding the methodological problems associated with EF and ES. Both certainly have an immediate appeal for informing the SCP policy agenda as they bring natural limits into the scope of policy and therefore allow answering unique policy questions. At the same time both suffer from concerns about their methodological robustness and can certainly only used with great care, for informing SCP policy embedded in larger sustainability indicator sets.

SCP001 43/130

3.4 From Weight to Impact – Changing Perspective on Material Flows

As requested, this report has so far almost exclusively focussed on the description and analysis of the economy’s “physical inventory” on the input and output side in terms of physical weight. The derived information is particularly helpful when it comes to the design of policies that ultimately aim to effectively reduce or control particular physical flows. With the exception of SFAs, these policies might all be grouped into the wider field of de-materialisation. However, they do not allow for the assessment of environmental flows in terms of their environmental impacts and the prioritisation of material flows in terms of their contribution to particular environmental problems. In this sense the reviewed methodologies do not allow for a comprehensive treatment of the issue of de-toxification (reduction of impact per unit of average material flow) as an integral part of a material flow policy and the associated need for material substitution. A full version of this discussion can be found in Appendix I.4.

By considering environmental impacts and detoxification we are able to address not only the physical size of flows through preventive measures that increase resource productivity and recycling, but also the composition of the flows through material substitution efforts. In this sense impact assessments provide complementary pieces of information, which increase the capabilities and policy relevance of MFA methodologies and provides information to the full menu of (general) policy options for tackling material flow patterns.

There are two general ways of taking environmental impacts into account:

Full environmental impact assessments;

Impact potential approaches (IPA).

Even though full environmental impact assessment approaches carry great potential for the formulation of specific policies, these approaches are still surrounded by high levels of uncertainty and can only be applied to small spatial entities and specific substances. Their broader application in the SCP policy context seems unfeasible.

IPAs – as commonly used within full fledged LCAs - can be more easily linked to the different MFA methodologies and readily applied to the government’s SCP framework. However, as indicated by their name they do not deal with actual, but potential impacts. In particular, based on some impact equivalence coefficients, material flows can be weighted according to their potential contribution to a specific environmental problem theme such as global warming or acidification. This allows the assessment and comparison of the contribution of a list of materials to a certain impact category. As IPAs extend the capabilities of MFA methods, the prioritisation of materials as well as material substitution could be informed through the SCP evidence base.

Method CommentGeneral IPA Commission study for identification of robust environmental impact

categories; create “government shortlist”EMFA Linkage to environmental impacts most difficult; currently comprehensively

achievable only via integration with LCA (see Appendix I.6)BMFA/MSA Linkage to environmental impacts via integration with LCA data or

integration with EIOA/EIOLCAEIOA Linkage to environmental impacts partially achievable with available data,

partially via integration with LCA data (EIOLCA)LCA/ EIOLCA/ hybrid LCA Linkage to environmental impacts inherent part of LCA; establish good

practice in use of impact assessment

SCP001 44/130

SFA Linkage to environmental impacts at substance level; usually straightforward

Table 4.7 - Comments on linkage of MFA methodologies to environmental impact categories

It should be noted that the reliability and robustness of impact coefficients (which are at the heart of IPAs) vary across environmental themes. Within the scope of this report it could not be established, which ones are suitable to inform policies and which are not.21 The commissioning of a study is recommended, which provides the required information to shortlist environmental impact categories for informing SCP policy. Ideally, these would form the environmental problem themes, which should be applied in all future LCAs carried out in the course of the government’s SCP/SD programme. This review should be carried out by an inter-disciplinary team of researchers including LCA practitioner, ecologists and other MFA experts.

A further aggregation of potential environmental impacts across environmental themes could not be recommended for three reasons, which are outlined in more detail in the Appendix I.4:

Aggregate (potential) environmental impact cannot be appropriately conceptualised (everybody likes to talk about it, but nobody knows what it is);

Aggregate (potential) environmental impact cannot be observed in the real world and therefore cannot be measured in a controlled way (nobody will ever know what is measured);

Despite these issues, the aggregation level might lead to a situation where results are overly condensed. This might hide important information required for well-informed decision support for MFA policies and encourage choices between existing rather than innovative new solutions (great danger of discouraging eco-innovation).

Therefore, it is concluded that the aggregation of the various environmental impacts into a single index is neither statistically robust nor meaningful. This also means that such aggregation should not be carried out within LCA analyses. The results of the assessment of the policy relevance of environmental impact approaches is summarised in Box 4.6.

Dematerialisation and detoxification should be jointly considered through linking MFA methodologies with impact potential approaches as commonly done for LCAs;

This allows prioritising physical flows as required on the SCP agenda and talking meaningfully about material substitution;

Environmental impacts cannot be meaningfully and robustly aggregated into a single index; Reliability and robustness of impact coefficients across environmental themes is unclear and should be

reviewed in a inter-disciplinary research effort commissioned by the government.

Box 4.6 – Key recommendations environmental impact assessment

3.5 Policy AnalysisSo far this part of the report has exclusively been concerned with ex-post analysis. Even though important insights for shaping future policies can be derived from the analysis of “historical” data, policies are strictly forward directed. Further value can therefore be added by using MFA evidence to advise policy makers on the choice and effectiveness of different policies and to reduce the large uncertainties attached to unknown future states. This can be done through the realm of

21 A list suggested within the LCA literature can be found in Appendix I.4.

SCP001 45/130

policy analysis and scenario models.22 These models establish relationships between external factors (which are not under the control of policy makers), policy instruments (variables which are assumed to be under full control of decision makers like taxed or subsidies) and policy targets through the introduction of mathematical functions. While the complete review of policy analysis can be found in Appendix I.5, only some of the most important findings will be briefly summarised here:

As human well-being is the ultimate concern of policy makers based on an anthropocentric conception of SD, it is crucial that human welfare is appropriately represented in policy/senario analysis models. This review follows the arguments of the national accounting literature and suggests that welfare cannot be measures in a single index. Instead different welfare attributes should be measured in the most appropriate units and collated in a comprehensive indicator set. In this sense negative externalities triggered by physical flows should be represented in physical units. In turn a very critical view should be taken towards approaches, which provide policy decision in a single unit such as environmental valuation or other stated preference approaches.

Because physical flow variables only represent a small selection of relevant welfare attributes, future policies can only be adequately analysed, if environmental, economic and social variables are represented.

The requirement of the SD and SCP concept as adopted by the government, to analyse the social, economic and environmental dimension jointly, implies that comprehensive decision support can only be provided in large scale models. The development of existing models into this direction should be encouraged.

Among the MFA methodologies EIOA seem to provide the best (if not only) starting point for such developments as they can link (quantitative) social, economic and environmental information in one integrated and consistent data framework in multiple units and multiple classifications (see Keuning, 2000; Stahmer, 2002).

While EIOA models can be useful in providing “what-if” scenarios for strategic planning, they are very limited in modelling structural policies such as taxes and subsidies, which have economy-wide re-shuffling effects. However, the underlying accounting data of EIOA can also be used in many other types of integrated environmental economic models from linear programming, to macro-econometric to dynamic computable general equilibrium models. Structural macro-economic models seem to be an especially appropriate model choice for comprehensive policy analysis in the spirit of the government’s SD concept.

While physical data has been comprehensively used for policy/scenario analysis in the climate change debate, other MFA evidence – most prominently represented in the collection of Biffaward studies - has often not been sufficiently exploited. In these situations a lack of integration of social and environmental aspects was found. When scenarios are carried out, often clear links to policy instruments available to policy makers are missing. This section concludes by highlighting the relevance of applying MFA evidence in policy analysis models to inform the SCP agenda and reduce uncertainty associated with the economic, social and environmental implications of policies. The scenario capabilities of some of the comprehensive UK resource flow models will be discussed in the next section.

22 Even though the term policy analysis has a more general meaning following Tinbergen (1957), it will be used interchangeably here with scenario analysis in the sense of ex-ante analysis in the most general sense.

SCP001 46/130

Increased use should be made from the application of MFA evidence in scenario models to inform SCP policy and reduce uncertainties surrounding the choice and effectiveness of different policy instruments;

Encourage the joint treatment of economic, social and environmental issues in policy analysis models based on integrated accounting frameworks in multiple units and multiple classifications;

Encourage the clear identification of policy instruments in policy analysis to allow for unambiguous policy advice.

Box 4.7 - Key recommendations policy analysis

3.6 Resource Flow Models in the UKSeveral resource flow models have been developed for the UK, which provide an available and continuously updated stream of evidence for SCP policy. These models have been developed over the past five or ten years and applied in various projects to support policy. Four of these models will be briefly discussed here for their ability to inform SCP policy with MFA evidence:

Bestfoot Forward - “Stepwise”;

Stockholm Environment Institute York – “REAP”;

Cambridge Econometrics – “REEIO”;

University of Surrey – “Regional Material Flow Accounting model” (RMFA).23

It needs to be emphasised that this review is undertaken fully in the light of the policy needs defined in the government’s SCP agenda. If a model is not recommended, this does not mean that it cannot be valuably employed in a different context. All models have successfully undergone the rigid process of peer-review (see Chambers et al., 2000; Simmons et al., 2000; Bell, 2003; Wiedmann et al., 2006; Ekins and Etheridge, 2006) and can therefore be seen as methodologically sound. Moreover, the authors’ deep involvement in the build-up of the REAP model need to be highlighted. Even though we attempt to avoid any biased view, it is certainly an advantage that full information on the REAP development is available. All models except REEIO have been previously discussed in a review commissioned by the „Sustainable Consumption Roundtable“ (see MacGilvray and Levett-Therivel, 2004), which aimed on the identification of the most promising models to understand the environmental impacts associated with lifestyles in the UK.

The „Stepwise“ model is based on an ecological footprint (EF) methodology. The only alternative tool, which provides such an indicator, is REAP. Rooted in a bottom-up data collection procedure, „Stepwise“ can provide results in terms of carbon dioxide emissions, material flows and the ecological footprints broken down into different footprint components. The model has been used to bring the notion of ecological limits through the EF indicator into the realm of decision making on various levels. In general it therefore responds to a core theme on the SCP agenda. Policy advice is usually derived from an analysis of the different footprint components and the assessment of the options as to how these footprint components can be reduced. Much of the general usefulness of the „Stepwise“ model therefore depends on the outcome of the government’s review process on the applicability of Ecological Footprint as an indicator to inform SCP policy (DEFRA, 2005c).

REEIO, REAP and RMFA are all rooted in an EIOA methodology. Therefore, they provide a much wider coverage of environmental indicators and can answer a much wider range of policy questions based on a meso-level view of the economy (see Section 4.3.3, Appendix I.3.3), which

23 There is also a Scotland specific model, which is not of immediate relevance here, due to the focus on the UK as a whole. However, it might provide the best foundation for SCP decision support in a Scotland specific context (see McGregor et al., 2004).

SCP001 47/130

covers the full supply chain and connects production and consumption in a systemic way. Moreover, they readily allow for an integrated treatment of economic and social issues and can be used for comprehensive policy analysis. Viewed from these general methodological capabilities the models are therefore very similar. However, they differ quite significantly in terms of their data usage and implementation.

While REAP and REEIO can draw on updated input-output data up to 2003, the RMFA model still uses the analytical input-output data from 1995 (see Section 4.3.3). All models are based on, or, can readily integrate24 the complete set of NAMEA type environmental account data (see Appendix I.3.3). REEIO is the only tool based on „regionalised“ input-output tables, which are derived from national ones. This makes it the only suitable tool for modelling regional differences in the way that resources are used in production processes, even though the regional data not have the same quality as the national data in the model.

REAP and RMFA only use national production data.25 However, their consumption data used is more detailed going down to the local level (RMFA – county level; REAP – local authority area). Using these detailed data in the EIOA models allows informing about the differences in resource flows and their environmental impacts (in the categories covered by the environmental accounts) triggered by the final demand patterns in various parts of the country from a consumer responsibility perspective. RMFA distinguishes all 73 counties in the UK using micro-data from the Family Expenditure Survey and some additional transport data. A flexible database structure allows compiling expenditure patterns of socio-economic groups according to policy needs. REAP uses CACI/ACORN data, which readily distinguishes the expenditure patterns of 55 socio-economic groups in all (~500) different local authority areas. In this context the model has provided the first comprehensive, consistent and comparable set of resource and energy accounts for local authorities.

Further differences can be found in the treatment of import related emissions and resource flows. RMFA and REEIO estimate embodied emissions purely based on national data (see Section 6). In fact, REEIO further limits its interest in embodied emissions to the contribution of electricity to consumption to primary fuel consumption and air emissions. REAP is the only model, which attempts to relax the single region assumption associated with the calculation of embodied emissions (and resource flows) by using resource efficiency data from ten different world regions. Therefore, it is currently best suited for dealing with such trade related emissions.

REEIO has the most developed scenario capabilities. Rooted in a macro-econometric model it can envision structural policies such as taxes or subsidies and their economy-wide re-shuffling effects. Moreover, it provides a wide range of specific socio-economic variables and policy options. This allows assessing not only the effects of environmental policies on e.g. employment or economic activity, but also the resource flow implications of different economic or social policies such as changes in tax rates or any other fiscal policy. In this sense it seems to be the most integrated scenario tool available at the moment for the national and regional level (see Section 3.5).

REAP provides a good alternative for “what-if” type scenarios. In this context it provides quantitative evidence in key policy areas such as housing, transport or energy. The different variables are provided in relevant functional units for the user. Moreover, REAP is the first scenario tool

24 The RMFA model is documented as a regional material flow model with material flows and CO2 being the only two indicators. However, as the environmental accounts can be easily added, no problem occurs informing on a wider range of pollutants.

25 For both models material flows are the exception, where also (some) regional bottom-up data is used.

SCP001 48/130

available to (all) local decision makers in the UK, which allows for strategic MFA policy planning based on sound quantitative evidence. However, the REAP scenario functions are still under development. The model has remained static in nature so far and has not yet relaxed the restrictive assumptions of input-output models in the scenario context.

In terms of the RMFA model there was a lack of information about its scenario capabilities . However, the authors know that the developers at the University of Surrey have developed scenarios in the past. While it is believed that these were conducted in a similar “what-if” way to the ones in REAP, the government might want to inquire further on this issue.

Finally, REEIO and REAP are designed as software tools (see Figure 4.5) which can be readily used by policy makers (supported by the distributing research institutes). This is another important aspect in bridging the gap between research and decision-making and encouraging increased evidence-based policy decision on the SCP agenda. In this context the „update data“ function in REAP should be highlighted, which allows policy makers to input better bottom-up data where available. This ensures that the most recent and robust data can be used and should be seen as good practice.

Figure 4.5 - Software interfaces of REAP (left) and REEIO (right)

Overall it might be said that there is a set of more comprehensive resource flow models available in the UK, which can be valuably applied for ex-post and ex-ante analysis on the SCP agenda. The applicability of the “Stepwise” model with its particular focus on EF depends on the government’s decision on the use of the indicator. However, it was requested in the course of this project not to make recommendations on the issue to avoid interference with other on-going consultation processes.

The three EIOA based models can all be applied to a wide range of issues on the SCP agenda. Their shared methodological roots provide them with very similar capabilities in tracing resource flows. Differences are caused by the input-output data used on the one hand, and the additional data linked into the input-output framework by the different research institutes on the other hand. REEIO is the only tool, which can provide estimates of differences in resource use in regional production, RMFA can trace resource flows associated with consumption down to the county level and REAP further down to local authority areas. None of these tools seem to provide algorithms to disentangle the domestic/regional supply chain. The development of such analytical facilities is recommended. As a scenario tool REEIO certainly provides the most developed methodology in particular for modeling

SCP001 49/130

structural policies..The scenario capabilities of the RMFA model should be inquired into by the government as there was a lack of information on the issue in the course of this project.

3.7 Synthesis – Painting the Bigger PictureThe review of the policy relevance of the different MFA methodologies has shown that they can all provide relevant information to various SCP agenda items. This is mainly because they address physical flows on different levels in the material as well as in the human sphere, and can therefore answer a range of different policy questions.

Doubts about the immediate relevance to the UK SCP policy were only raised for aggregate EMFA indicators such as DMI, DMC or TMC26. At the same time the underlying physical EMFA accounts were identified as a crucial component of the SCP evidence base as they provide the only comprehensive and consistent description of the physical flows passing through the economic system (for details see Section 3.3.1).

Overall the various MFA methodologies broadly complement each other. The main challenge is to orchestrate their usage for the optimum achievement of the various SCP policy targets. With this in mind it was recognised that there are strengths and limitations associated with individual methodologies. There are a number of ways in which these could be overcome:

1. Ensuring good practice: Some of the limitations can be overcome through the establishment of methodological guidelines and standardised reporting in line with the government’s policy needs. This is particularly important for BMFA/MSA and SFA approaches, where methodological standardisation processes are under-developed. Establishing good practice allows the government to:

ensure robustness and high quality standards for individual studies;

improve comparability of results across studies;

integrate results in a higher level data structure, and contribute to the development of an SCP evidence base;

readily access relevant results.

The development of methodological guidelines and reporting practises is of particular importance, if a second wave of Biffaward type mass-balance studies is to be commissioned in the future by the government itself.

2. Integrating methodologies: The robustness and applicability of MFA evidence can be further improved through the integration of methodologies in hybrid approaches. The integration of the EIOA method as a top down approach with the other bottom-up methodologies (EMFA, BMFA/MSA, LCA, SFA) should be a high priority as it combines the best of two worlds: the complete coverage of the economic system in general and the economic supply chain in particular provided by EIOA approaches and the detailed and problem specific information compiled in bottom-up approaches.

The review has also established that the policy relevance of MFA evidence can be further improved if weight and dematerialisation is considered alongside environmental impacts and detoxification.

26 as described most comprehensively in Eurostat (2001)

SCP001 50/130

Impact potential approaches (IPAs) provide a suitable methodology. They can help to inform policies focusing on the most damaging physical flows first.

Finally, scenarios and integrated policy making are seen as crucial to reducing uncertainty and avoiding undesirable side effects associated with future policies. The quantitative testing of economic, social and environmental effects associated with different policy instruments is a major area in which physical flow evidence can help inform the government’s SCP agenda. The methodologies are mapped against these three core themes in figure 4.2.

Figure 4.6 - MFA methodologies in relation to dematerialisaton, detoxification and policy analysis

3.8 Answering DEFRA’s specific tender questionsBased on the previous discussion it is now also possible to comprehensively show how the different methodologies can help to answer the policy questions raised by DEFRA:

DEFRA1: Which materials/products/sectors can be targeted to achieve the greatest environmental benefits?

DEFRA2: Where is the best point in the supply chain, in relation to the different resources, for intervention to achieve maximum environmental benefit?

DEFRA3: How can the approaches be used to identify, set and monitor targets for SCP policy?

SCP001 51/130

DEFRA4: Is it possible to identify from the studies whether greater reductions in environmental impacts are achieved from product design (and the subsequent changes in production) or from improving resource efficiency in the production process ?

DEFRA5: What is the potential of the approaches for assessing the environmental impacts “foregone” from waste recycling, re-use and minimisation?

Because the emphasis of the questions is on “impacts”, to answer these questions the methodologies need to be linked to IPAs. Therefore, the ratings shown in Table 4.8 take into account the potential of the different methodologies to link with IPAs directly. It should be noted that the ratings in the tables only reflect which method can answer this sort of question not which is the best. Advice on the choice of methodologies is provided in the next Section.

Table 4.8 - Overview assessment results DEFRA tender questions

EMFA and BMFA cannot answer any question directly as they cannot be easily linked to environmental impacts.27 Question 1 can be answered by all methodologies. SFA only gets a satisfactory rating for question 2 as the supply chain coverage is usually very limited. Answering question 4 requires very detailed process information, which is not provided by most methodologies. In fact only LCAs and hybrid LCAs can fully answer this question. SFA is very limited due to its narrow focus on very few substances. Therefore, trade-offs in impacts are not well captured. The same holds for question 5. IO based methods have problems answering this question directly as waste treatment stages in a product/materials lifecycle are not well represented. They can however be easily computed using some additional bottom-up data. As soon as IO tables are extended in this way they can fully answer question 5 as indicated by the good rating for the waste IO LCA and the hybrid IO LCA method.

27 However, a methodology hybrid EMFA-LCA method will be introduced later, which allows one to identify materials in these terms. The same is then also true for BMFA.

SCP001 52/130

3.9 A Brief Guide to Policy Makers: on the choice of MFA tools - Choosing MFA methods for SCP policy use – from general to specific

The choice of methodologies first of all depends on the choice of policy questions under consideration. These have been outlined previously for each methodology and discussed in the individual assessments (see Appendix I.3). For an orchestrated use of the methodologies it is helpful to outline their nature in relationship to information demands on the policy level. There seems to be a general link between the level on which policies are required and the level of detail provided by the individual MFA methodologies (see Femia and Moll, 2005).28

Figure 4.7 - MFA methodologies in relation to policy demands on different level (adapted from Femia and Moll, 2005)

This is shown in the information pyramid in Figure 4.7, which also provides a general idea about how information can be organised in the government‘s physical SCP evidence base. The more general the policy the more aggregate the information required to inform policy. In turn, the more specific a policy, the more specific the physical information required. Different types of policies can be broadly grouped accordingly and it becomes much clearer how the various methodologies can be used in the decision-making process.

LCI/As provide specific information about physical flows associated with the lifecycle of a certain product. They try to take all relevant production processes into account – from the cradle to the grave. This information can be used to establish product standards, to change a certain production process or to ban a certain product, if its lifecycle impacts are perceived as detrimental to the well-being of society.

EIO depicts production and consumption processes on the sectoral level. While its lack of detail makes it unsuitable for devising such specific policies, it covers all economic activities taking place within a certain region and not only a subset like LCI/As. This framework is much more suited for the devision of structural policies such as taxes, subsidies etc., which address larger entities and have economy-wide implications like re-shuffling effects from new green taxation schemes.

28 The ideas in this Section closely correspond with the ones provided by Femia and Moll (2005).

SCP001 53/130

On the highest level only a small selection of highly aggregated physical indicators is used by decision-makers. They are designed to measure the progress of policy strategies and programs and to evaluate their success. The derivation of actual policies is beyond their scope. The SCP de-coupling indicator set is a good example. From this perspective it is not surprising that specific data provides much clearer ex-post links between LCAs and policy than less specific physical data. However, this does not mean that highly aggregated data is any less policy relevant – it just has a different purpose.

Figure 4.7 also highlights the key role of hybrid methodologies, which establish a link between specific process data and higher aggregated economy-wide data. Their ability to integrate specific bottom-up with aggregated top-down data pulls down the boundaries between methodologies and makes way for a more integrated and consistent approach to addressing policy questions at the right level. EMFA comprehensively accounts for overall changes in physical flows. In this sense it can be seen as the general framework through which SCP policies and their material implications take place.

There are other factors that might be of importance for the correct choice of a method depending on the purpose of the inquiry. The nature of a pollutant or the spatial scale could make certain methodologies more and others less appropriate. Supply chain policies in regional or local economies might not be well represented in economy-wide frameworks, for example. Moreover, policies for highly local pollutants such as PM10 can also only be addressed at this spatial level as the impacts elsewhere of the same amount of flow might be irrelevant.

3.10 Establishing strategic evidence sources for informing the prioritisation of SCP policies

The previous paragraphs clarified how the various MFA methodologies can be chosen according to the type of policy intended. Together with the detailed assessments of the various methods this should provide valuable information on how available and new MFA evidence can be related to policy types. However, the UK’s SCP agenda continuously stresses the need to focus on the most environmentally damaging material flows first. The remainder of this section discusses the methodologies usefulness for informing a strategic approach to SCP material flow policies, which combines dematerialisation and detoxification and allow for a prioritisation of physical flows according to their environmental impacts (see DEFRA, 2003; DEFRA, 2005b). Such an approach is crucial for the use of existing and build-up of new SCP evidence.

This can be described as a strategic material flow policy approach and is geared towards the actual use of materials – from raw materials to products to wastes. Such a clear focus on materials allows material flow policies to complement rather than interfere with existing environmental policies which usually target substance releases/emissions directly.

It could be said that the ultimate aim of a strategic material flow policy approach is the reduction of the life cycle impacts of material use from the cradle to the grave irrespective where the impacts occur (see also De Bruyn et al., 2004; Van der Voet et al., 2005). Such a strategic approach should be built around three pillars, which provide the main policy levers available for implementing:

Dematerialisation/ Prevention – reducing the amount of physical input required per unit of output;

SCP001 54/130

Re-use and recycling – closing the loop and extending the life-span of materials through longer use of materials and increased use of secondary materials;

Material Substitution – using materials with smaller environmental impact and replacing the environmentally most damaging materials.

Each policy derived from this strategic material flow policy approach can then be expressed in one of these three categories.

3.10.1 A general framework for tackling the most important material flows first

Figure 4.8 - Framework for prioritising SCP material flow policies

summarises in a step-by-step procedure how material flows can be prioritised in such a strategic approach according to the environmental impact and the requirement of physical evidence:

1. Prioritise environmental problems: Because there is no meaningful way to talk about aggregate environmental impacts across environmental themes, it is not possible to prioritise physical flows solely based on MFA evidence. It remains therefore the responsibility of decision-makers to prioritise the different environmental problems on the agenda. Currently climate change, for example, is right at the top.

2. Identify key substances: Once environmental problems are prioritised, physical evidence can be used to prioritise substance flows accordingly.

3. Prioritise materials (natural resources, goods, wastes): For a strategic division of SCP material flow policies, it is important to link the substance releases ultimately causing the various environmental problems to the flows of materials and goods in the economy (for argument see Appendix I.4). These are often the direct target of policy intervention.

SCP001 55/130

Based on the prioritisation of environmental problems and substance flows, the flows of materials and products can be prioritised according to their contribution to different environmental problems based on MFA evidence. However, material flows – in particular the flow of goods – also have an important economic and social dimension. Therefore, further social and economic indicators such as the total economic activity triggered or the employment generated throughout the economy should be added to allow for integrated SCP decision-making. These will need to be considered jointly with material flow evidence, in order to come up with an ultimate shortlist of key materials. The potential trade-offs among these different economic, social and environmental issues will remain the responsibility of decision-makers and cannot be avoided through the application of quantitative methods.29

4. Identify key policy intervention points/ key activities: different MFA methodologies can be used to help identify where the most important flows of materials and goods (and therefore also embedded problem substance flows) occur throughout the economy and therefore where they might be best tackled throughout the domestic economic system.

5. Identify most effective policy instruments: MFA evidence can be applied in policy analysis models to identify the most efficient and effective policy instruments for tackling key environmental flows for the government’s SCP agenda.

3.10.2 Using MFA methodologies for informing prioritised MFA policies Even though each individual MFA methodology can inform the government’s SCP agenda not all are suitable for prioritising physical flows – BMFA/MSA, for example, are much better suited for tracing prioritised materials, while SFAs30 can help in establishing crucial links between the flows of materials (and goods) and substances. There are some key requirements that should be met by a methodology, if it is comprehensively to inform the prioritisation process outlined above:

All evidence used in the prioritisation process should be integrated in one comprehensive methodological approach;

All economic activities and associated physical flows of the system under consideration need to be covered. For DEFRA this usually means the national economy. Under the condition of full system coverage, an increasing amount of detail (i.e. sectoral disaggregation) is beneficial for the analysis;

Dematerialisation and detoxification can only be addressed comprehensively, if clear links between the different aggregation levels of physical flows are established as shown and further explained in Figure 4.9;

The methodology should not only focus on environmental, but also economic and social information;

29 There are quantitative methods to do things such as multi-criteria analysis. However, these methods do not seem to provide robust results and their application is not recommended. Due to the lack of detailed insights into this field, the government might want to ask for further consultation from experts.

30 In fact, SFAs are generally in a very special position in this context as they have usually been triggered by a specific environmental or human health problem caused by a particular substance. In this sense, prioritisation has been a characteristic feature of SFA applications.

SCP001 56/130

The methodology should facilitate the performance of comprehensive scenario analysis in order to identify the most suitable policy instruments to suffice the government’s ambition jointly to consider these issues together in the course of SD and SCP policies (DEFRA, 2005).

Environmental problems are usually caused by the flows/release of certain substances. Therefore, there is a direct link between the size of the substance flow and its (potential) environmental impact. Such a link cannot be established for physical flows on higher aggregation levels. Material as well as aggregate flows can only be linked to environmental impacts by linking them to the substance releases causing the problem. Hence, to prioritise MFA policies in a strategic policy approach physical flows need to be addressed on the various aggregation levels within one analytical framework.31 However, the links between bulk materials and substances are only well established for particular environmental problems such as global warming or acidification. The previously recommended review of IPAs therefore should also identify where these links between substance and bulk materials flows need to be strengthened in the future for moving towards a fully integrated material flow evidence base.

Figure 4.9 - MFA methodologies in an integrated material flow approachThe development of a general or several more specific models for prioritising material flow policies based on robust MFA evidence on the national level, is a key recommendation of this review. Only such an approach allows the strategic use of available MFA evidence and the build-up of new MFA evidence according to the government’s SCP policy needs.

Figure 4.9 highlights that only LCA and IO based methodologies are capable of addressing physical flows on the various aggregation levels. However, LCAs by their nature do not attempt to address the whole economic system, but only economic activities relevant to their functional system boundary setting. Recently it has been shown that this shortcoming can be partially overcome through

31 By extending this argument also the complementary relationship between conventional environmental policies focussing on substance releases directly and MFA policies can be clarified. Substances are often only directly measurable at certain stages of their journey from the environment through the economy and back - usually just at the point when they leave the human sphere again as emissions to the various environmental media.? This is where traditional environmental policies usually address these substance flows. At most stages throughout the production-consumption chain they are embodied in materials. Even though the environmental impacts are the ultimate interest for policy makers, materials and products are the direct object of strategic material flow policies. Therefore, they address undesirable substance flows indirectly in a “preventive” approach through dematerialisation, material substitution and recycling/reuse.

SCP001 57/130

integration and generalisation, as will be explained later. Therefore IO and LCA based methodologies are the key for providing strategic policy advice.

What is the contribution of the other methodologies to such a strategic SCP policy agenda? While certain methodological frameworks might facilitate prioritisation of policies, the other methodologies remain crucial for informing government policy:

The methods can still be applied in their “natural” capacity for tracing physical flows of strategic importance. Once a key material is identified, BMFA/MSA, for example, can be used to trace it (see Section 4.12). In the same way the build-up of LCI evidence can be used to cover product groups with particularly high environmental impacts (see Tukker et al., 2005).This can be summarised as follows: “Essentially nothing changes – only that the method is applied to the most important flows on the SCP policy agenda and therefore informs the development of the most policy relevant SCP evidence”.

The methods can contribute to the prioritisation process, if integrated into hybrid models. Later a hybrid LCA-EMFA will be introduced for the identification of strategic materials (see Section 4.12).

The methodologies can help to strengthen the links between the different aggregation levels of physical flows. While the link between aggregate and (bulk) material flows are well covered by EMFA, there are still many missing links between the flow of substances in the economy and materials. For important environmental problems SFAs can help to establish these missing links and to collect the required data. This is because SFAs are usually triggered by a substance specific problem, which is then traced throughout the economy (mostly embodied in materials/goods).

This complementarity of MFA methodologies is reconfirmed in Figure 4.9.

3.10.3 Towards a comprehensive and strategic evidence-based material flow approach for informing SCP policies

Having identified the need for a strategic material flow approach, proposed a framework to approach the challenge, located the various MFA methodologies within this framework, and developed criteria that should be fulfilled for informing the prioritisation process with robust evidence, the remaining task is to discuss different ways of providing such evidence in the light of the UK data availability.

EIOAs immediately fulfil all requirements as a methodological framework on which a comprehensive model for the prioritisation of physical flows could be built on to provide evidence for a strategic prioritisation of SCP (material flow) policies. Not only can flows on all different aggregation levels be dealt with simultaneously in one consistent economy wide data framework, but socio-economic as well as environmental issues can be jointly addressed and different sets of policies can be readily tested in policy analysis models. However, the data situation limits the present ability comprehensively to implement such an approach. Only for climate change can direct links between material and substance flows be established. All other physical flows can only be related to the monetary flows of goods in the economy for a limited amount of pollutants.

LCA approaches on the other hand only cover the economic processes relevant to their product system boundaries and LCA databases only cover a subset of goods circulating in the economy. However, more recently LCA data has been increasingly generalised to cover the whole economy (see

SCP001 58/130

Tukker et al., 2005). This generalised LCA approach begins with an individual product and conducts a Life-Cycle Assessment (LCA) of it. The results for this particular product are then assumed to be representative for a wider range of products and so are extrapolated to a much larger group of products. Combined with other LCAs for representative products it is possible to put together a picture of the whole economy. In this course, De Bruyn et al. (2004) and Van der Voet et al. (2005) have linked EMFA and LCAs to identify materials of strategic importance. Such approaches are limited in different ways:

Data cannot be easily linked to socio-economic data in such an integral and comprehensive ways as possible for EIOAs.

Data is not framed in a format that it is readily compatible with comprehensive integrated scenario approaches as discussed in Section 3.5.

Bottom-up nature of the approach makes results less comparable. This is mainly due to the bottom-up nature of the data, but aggravated by the weak links between EMFA and LCA data.

Due to the generalisation, data might contain considerable error and it is unclear where physical flow impacts can be best addressed throughout the economy.

On the other hand, the detailed information provided by the generalised LCA approach can be used to advise how the environmental impacts of materials can be most readily reduced through dematerialisation, material substitution and recycling/re-use. Hence, especially in the end-of-life stages, there is a clear superiority of LCA based approaches over EIOA ones.

Overall it must be concluded that it is currently not possible to support prioritised SCP material flow policies with consistent evidence from one comprehensive model. However, in a more modest approach prioritisation can still be achieved. In this context the greatest benefits might be achieved if the above models are used side-by-side:

EIOA models are readily available (see Section 3.6) and are much better suited to prioritise SCP policies in the economy-wide context. They allow for a much more robust identification of key sectors or key products under joint consideration of economic and social key variables and allow the identification of the best policy options in comprehensive scenario models. Some or all of these models could be progressively developed into more comprehensive hybrid. On the one hand the physical data underlying these models must be improved, and facilities to comprehensively integrate bottom-up data need to be put into place. On the other hand comprehensive modules could be developed in areas of key interest for the government’s SCP agenda such as waste.

Generalised LCAs could be used for prioritising material flows more comprehensively and inform government how the environmental impacts of these key physical flows can be best reduced. Such a model does not exist for the UK (even though some data is available) and would need to be developed. In the course of a scoping study it should be established whether robust results can be derived from such an approach and whether or not the required data is available.

In the remainder the value of both model approaches is demonstrated. In the next Section example applications are given for EIOA/hybrid IO applications, while the last Section demonstrates how generalised LCA type models can be used to identify strategic materials.

SCP001 59/130

3.11 Examples – Applying a strategic evidence-based material flow approach

This Section discusses how existing EIOA models can be used and extended to provide quantitative policy advice for the government’s SCP agenda. In the course of this we will also briefly outline how other MFA methods can be used in this context. Examples in three policy areas are provided:

Sectoral Benchmarking

Integrated Product Policies

Sustainable Waste Management

Sectoral Benchmarking and IPPs are provided as examples for immediate applications of such strategic evidence sources. Sustainable Waste Management will serve as an example for the need to extend the capabilities of existing models in key policy areas.

3.11.1 Example 1: Sector Benchmarking – Monitoring SuccessSector benchmarking plays a prominent role on the government’s agenda. It is the key to monitoring success and identifying good practice in the business community as well as in developing policy action for establishing greener and more competitive business practices. Moreover, sector benchmarks can also indicate where there might be the greatest potential to increase resource efficiencies throughout the supply chain.

It is important for prioritised decision making in this context that all economic sectors are first benchmarked. This requires a consistent methodological framework, which provides stable and comparable results. This can only be achieved in EIO frameworks (see Lenzen, 2005). Alternative approaches, where a full range of sector studies is performed (see Biffaward studies Section 5) following an MSA or BMFA methodology, would not be able to provide the same level of detail and comparability even if good practice through a strict definition and control of reporting guidelines was established by the Government. The prioritisation process itself would start from the question ‘what are the relevant environmental problems and how do the various industrial sectors contribute to these?’. The juxtaposition of social and economic information in this context is key for informed decisions as well. Important physical variables not covered by the standard environmental account data in the UK such as water consumption could be added using bottom-up data.

In a second step a detailed supply chain analysis of key sectors could be carried out in order to identify the most promising policy intervention points. Different policy interventions could then be tested for their economic, social and environmental implications in a scenario analysis. Alternatively, within each sector key production processes within the supply chain could be identified and tackled. Moreover, in correspondence with the business part of the SCP agenda, the most important players could be identified in key sectors and benchmarked against the sectoral average to identify good practice.

Manfred Lenzen (2005) and his colleagues from the University of Sydney carried out a benchmarking for all 135 sectors of the Australian economy following a triple bottom accounting approach. The difference to conventional triple bottom line assessments is their use of a variety of socio-economic and environmental performance indicators estimated from an input-output methodology. The positive side effect are not only an increased comparability due to the use of such a standard framework, but also that it allows them to assess the total impact in the various dimensions taking into account all upstream and down stream impacts.

SCP001 60/130

The study is not only a good lesson in terms of how to enrich sector reporting through an input-output framework, but also in how the large amount of information created by the model can be effectively condensed for effectively inform policy makers. The spider diagram above shows the normalised figures of the headline indicator set. It shows, for example, that the aluminium industry uses 7 times more primary energy than the average Australian sector, but only causes a tenth of the average land disturbance. Further details were provided for experts covering the identification of the most important supply chain nodes. These provided the basis for in depth consultation where resource flows could be tackled in the supply chain. Such an in depth, comparable sector benchmark for the UK economy would fit very well into the Government’s agenda (e.g. BREW programme) and contribute to strengthen the SCP evidence base.

Box 4.8 - Good practive sector benchmarking (Lenzen, 2005)

In such cases hybrid IO tools seem to be the most promising way forward, if consistent results are to be achieved even though conventional benchmarks based on a MFA/MSA methodology might be seen as an alternative under the promise of the establishment of good practise. Alternatively the government might want to encourage or even enforce the establishment of environmental management or other CSR type reporting systems in key sectors. As soon as the spatial dimension plays a role on a subnational level, IO based approaches reach their limits as the national IO tables available for the UK might not adequately represent the regional or local production structure of a sector. In such a case, again, the more flexible bottom-up methods seems to be the preferable choice. Overall sectoral benchmarking is an area of interest, in which strategic evidence can be immediately applied.

3.11.2 Example 2: Integrated Product PoliciesAll products cause environmental degradation in some way, whether from their manufacturing, use or disposal. Integrated Product Policy (IPP) seeks to minimise these by looking at all phases of a products' life-cycle and taking action where it is most effective. MFA evidence for IPP is relevant on various menue items of the SCP agenda – most importantly the “product”, “consumption” and “procurement” agendas. Various methodologies such as LCAs, EIO, hybrid IO LCAs or functional SFAs can provide valuable information in this context.

For example, the evidence can be used to provide information to private, public and corporate consumers in support of their choice processes or to help identifying where in the supply chain the lifecycle impact of a certain product or product groups can be most effectively reduced. As mentioned

SCP001 61/130

above a systematic use of the various MFA tools in the spirit of the government’s agenda requires a move from the general to the specific:

Identify those product groups among all the goods (and services) circulating in the economy, which have the highest impact in a particular, or across a variety of, environmental problem themes: This can either be achieved on the basis of generalised LCA or through hybrid IO methods. The latter has the advantage that it covers all products and offers a consistent framework for allocating environmental impacts consistently to the supply chain of certain product groups without facing the previously discussed requirement to cut-off any process in the various supply chain layers. This circumvents the truncation error of the generalised LCA approach, but also avoids the problem of generalising the environmental impact of individual products over a product group. However, such an approach only comprehensively covers pre-consumptive stages and considerable efforts are required to extend the tables for the environmental effects of the use phase of products as well as the management of wastes emerging after the use of products at the end of their lifecycle following a hybrid IO LCA approach. (see Tukker et al., 2005: 38; Joshi, 2000).

Identify where further evidence is required in key areas and comission specific LCA studies: The complete assessment of product performance allows the identification of key product groups. These might be the ones which have a high environmental impact across most of the environmental problem themes or ones which score in areas of key concern for the government. In the context of consumption these might for example be goods with a high toxic content or which produce a lot of hazardous waste throughout their life cycle. This will depend on the government’s priorities within a certain policy field. If further evidence for the completion of product information databases (such as environment direct), in the identified key product groups is sought, traditional or hybrid IO LCAs might be comissioned with the previous discussion of their pros and cons in mind (see Section 3.3.6). Hybrid IO LCAs might be preferable, if the government seeks a consistent set of studies on an array of products in the identified key product group. Moreover, if toxic substances contained in products are of concern, a functional SFA, for example, might be a suitable choice of tool. The SFA literature shows (see Brunner and Rechberger, 2004) how the most relevant products could be indentified within the wider product group.

Overall, IPP seems to be an area where information for prioritising the government’s agenda can be provided without major alterations in existing models even though end-of-life stages would need to be added to the EIOA models. Guidance is provided in Tukker et al. (2005).

SCP001 62/130

Tukker et al. (2005) use a hybrid IO with LCA type extension for end-of life product stages for the identification of products with the greatest environmental impacts across 10 environmental problem themes. Such information can be used to prioritise product policies outlined above according to product groupings, consumption domains and functional areas of consumption.

ConsumptionDomain

Abi

otic

depl

et.

Glo

bal

War

m.

Ozo

ne la

yer

Dep

l.

Hum

anto

xici

ty

Eco-

toxi

city

Phot

o-ch

emic

alox

idat

.

Aci

di-

ficat

ion

Eutr

ophi

ca-

tion

Agg

reg.

Scor

e

Priv

ate

and

publ

icex

pend

iture

.

CP01 Food and non-alcoholicbeverages

0.206 0.293 0.236 0.236 0.316 0.255 0.297 0.581 0.318 0.166

CP07 Transport 0.199 0.185 0.140 0.248 0.147 0.204 0.138 0.061 0.171 0.141

CP05 Furnishings, householdequipment and routine maintenance ofthe house

0.278 0.159 0.124 0.117 0.125 0.131 0.183 0.070 0.139 0.120

CP11 Restaurants and hotels 0.070 0.091 0.090 0.084 0.090 0.088 0.096 0.126 0.094 0.096CP04 Housing, w ater, electricity, gasand other fuels

0.070 0.077 0.082 0.094 0.079 0.088 0.074 0.029 0.073 0.131

CP09 Recreation and culture 0.053 0.060 0.107 0.066 0.068 0.067 0.071 0.035 0.061 0.091CP12 Miscellaneous goods andservices

0.047 0.052 0.098 0.063 0.055 0.065 0.055 0.021 0.053 0.103

CP03 Clothing and footwear 0.022 0.024 0.035 0.027 0.057 0.032 0.024 0.045 0.031 0.031CP08 Communications 0.019 0.021 0.026 0.024 0.021 0.023 0.023 0.007 0.020 0.040

CP02 Alcoholic beverages, tobacco andnarcotics

0.016 0.017 0.018 0.019 0.022 0.019 0.015 0.016 0.018 0.027

CP06 Health 0.015 0.016 0.037 0.017 0.014 0.020 0.015 0.007 0.016 0.039CP10 Education 0.004 0.005 0.007 0.006 0.005 0.006 0.006 0.002 0.005 0.014

Total fraction: 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

In this context the analysis showed that high ranking products vary considerably in their scores per Euro on specific impact categories. For DEFRA such results would imply that priorities on specific environmental problem themes would lead to different priorities in product policies. The results further showed that some services belong to the top 35 (among 255) products and that services cannot be ruled out from product policies. As the impacts of services often almost entirely stem from off-site resource use higher up in the supply chain, the importance of covering the whole economic supply chain and avoiding cut-off points like in traditional product LCAs is reconfirmed.

Box 4.9 - Good practice prioritising product groups according to environmental themes (Tukker et al., 2005)

3.11.3 Example 3: Sustainable Waste ManagementFinally one example is provided to demonstrate how building more specific models would enable comprehensive prioritisation in areas of key interest on the government’s SCP agenda. Sustainable Waste Management has been chosen as much more evidence for the main priority “climate change” has already been compiled.

Sustainable Waste Management ranks high on the government’s SCP agenda constituting one of the seven specific SCP objectives. Three immediate key concerns of the sustainable waste management agenda spring to mind, which have direct implications for the build-up of appropriate physical MFA evidence to inform the government’s waste agenda:

Closed-loop-economy: Throughout the waste agenda the paradigm of a closed-loop-economy is highlighted. A precondition for a move into this direction is information on who produces how much waste, of what type, and where in the supply chain. Hence comprehensive information on the waste flows between sectors as well as final consumers is of great importance to the sustainable waste management agenda.

SCP001 63/130

Waste treatment: The efficiency of different waste treatment options are controversially discussed in the sustainable waste management debate. In this context it is important to consider all ripple-down effects caused by a solution. This comprises the cost structure, the scale of a solution, the induced activities (e.g. transport and impacts on other associated industries). Therefore, comprehensive information about the interlinkages of different waste treatment options with other economic activities are required for a full assessment of the efficiency of waste treatment options.

Policy Analysis: Another issue in the debate on sustainable waste management is the use of fiscal policy instruments better to control/ reduce the waste arising in the UK and to encourage the re-use and recycling of materials. As fiscal policies always have structural impacts throughout the economy, it is of great importance to test possible economic, social and environmental consequences in a scenario approach. Such an analysis will only satisfy the requirements imposed by the holistic nature of the SCP agenda, if an integrated data framework in multiple units and multiple classifications is used (see Section 3.5).

A framework, which facilitates this types of analysis and provides many other relevant analytical options is the Japanese Hybrid Waste IO LCA model (see Takase et al., 2005a). The model is briefly described in Box 4.10. The build-up of such a comprehensive model would allow the government to provide the national strategic guidance currently needed in waste policies. It combines capabilities to identify the key product and material flows associated with waste generation and the identification of policy intervention points throughout the economy. The best choice of policy instruments for waste treatment options and waste prevention strategies can be assessed in a scenario framework taking into account social and economic effects. The LCA methodologies enable a focus on particular waste streams and waste generation within a product system. This would enable the government to inform the sustainable waste agenda from one comprehensive and consistent evidence source. The model could be set-up as “data modules”, which can be readily linked to existing models.

Other methodologies such as BMFA/MSA might also be commissioned during the build-up, when important data gaps are discovered. They could be designed in a way that they directly feed into the data structure of a waste IO model. In later stages such a model could then be further developed to account for regional differences. SFAs would be applied in the waste context to trace the origin of certain hazardous substances leading for example to problems in their long-term storage or treatment.

However, the lack of good waste statistics in the UK needs to be re-emphasised. The lack of reliable statistics for commercial and municipal waste has just very recently been pointed out by members of ONS (e.g. Gazley and Francis, 2005). This structural data gap needs to be overcome by concerted and swift data collection efforts (if necessary by allocating further resources towards these institutions), by the responsible institutions – mainly ONS and the Environment Agency.

SCP001 64/130

The Japanese waste input-output model is an excellent example for models, which can be used to prioritise waste policies in the SCP context (see Takase et al., 2005a). As indicated in the table structure below it provides information on material flows including the flows of goods and services in the economy, substances (added according to needs) and other socio-economic variables such as income and employment in an integrated accounting system in multiple units and multiple classifications. The distinguishing feature is the detailed description of waste streams and their treatment including comprehensive information associated with recycling.

The WIO model has been used to assess the waste streams triggered by Japanese consumption patterns in general or applied in a more specific context such as the waste flows and management options associated with metal flows in the Japanese economic including costs (see Nakamura and Nakajima, 2005). However, as a hybrid-IO-LCA methodology the WIO model is capable to take into account all the phases of the life cycle – production, use and end-of-life. In this context it provides facilities to integrate the technology matrix of a traditional product system LCA in the technical coefficient matrix of an economy provided by the IO tables. This allows not only the prioritisation of waste policies on higher aggregation levels, but also detailed inquiries such as the assessment of the recycling options and their associated costs for of end-of-life electrical home appliances (see Kondo and Nakamura, 2004; Nakamura and Kondo, 2006). Recently it has also been linked to environmental economic models, which have modelled the rebound effect associated with energy or material use (Takase et al., 2005b). These behavioural adjustments in consumer behaviour caused by changes in relative prices have received much attention in the SCP debate, but have been rarely comprehensively quantified. (see Bingswanger, 2001; Jalas, 2002)

Box 4.10 - Good practice: The Japanese waste input-output model

SCP001 65/130

3.12 Towards a Research Agenda for Tracking Strategic Materials in the UK

This last Section responds to the Government’s request to identify strategic materials and to propose a methodological framework for tracing them. To do this an example is provided of how generalised hybrid EMFA-LCA methodologies can form a second strategic evidence pillar for the government’s SCP policy agenda (covering the first three steps of the proposed strategic material flow policy framework (see )). To make it useful outside the context of the report, this Section has been written in such a way that it can stand on its own.

3.12.1 Policy Background and MotivationIn December 2005 the European Commission introduced its new “Thematic Strategy on the Sustainable Use of Natural Resources” (EC, 2005a; EC, 2005b). The overall objective is to reduce the negative environmental impacts generated by the use of natural resources in a growing economy through:

improvements in resource productivity;

reductions in resource-specific environmental impacts.

As most resource or material flows do not fall under exclusive Community competence the strategy acknowledges that success is dependent on the national policies in the member states. To reach its objective the Commission therefore proposes “that each EU Member State develop national measures and programmes on the sustainable use of resources to achieve the strategy’s objectives. The measures and programmes should focus on resource use which has the most significant environmental impact” (EC, 2005a: 9). In this context it is of key importance that each Member State develops a “strategic approach” (EU, 2005a: 6). To do this:

resource use and environmental impacts should be approached from a lifecycle perspective;

information on resource use and environmental impacts should be provided in a coherent database;

a global perspective should be taken on resource use and environmental impacts;

resource use policies should be integrated with other environmental, social and economic ones.

As a direct response DEFRA requested the proposition of an approach for the identification and tracing of key materials in the economy in the spirit of the EU strategy. This coincides closely with the strategic evidence-based material flow policy approach proposed earlier. The remainder of this Section will provide suggestions on:

which materials might be of particular interest for a strategic resource management approach;

and in what methodological framework these flows might be best traced and analysed.

3.12.2 Identifying key materials based on environmental impactsIn order to make the UK’s material use patterns more sustainable, the “thematic strategy on the sustainable use of natural resources” suggests tackling those materials flows first, which have a high environmental impact. This coincides with the government’s SCP agenda as outlined earlier. However, it has been previously established that environmental impacts cannot be compared across environmental (problem) themes. The adoption of measures of aggregate environmental

SCP001 66/130

impact for the identification of key materials based on quantitative evidence has therefore been rejected. Because different groups of materials might dominate the impacts in different impact categories, an unambiguous identification of key materials with high environment impacts might therefore not be straightforward anymore.

It will ultimately be the responsibility of decision makers to prioritise environmental problems on their agenda (see ). Supported by quantitative evidence, key materials can then be identified in this process. An approach such as “Distance to Target” might be appropriate then; exploring the proportion that a particular material contributes to a particular pollution problem and identifying how a material management strategy can help meet a particular target. For example, if the production of concrete produced 25% of all UK carbon dioxide emissions then it could be defined as an important product that required immediate attention and strategies could be devised on how its impact can be reduced through dematerialisation, material substitution and recycling (re-development) strategies.

In the absence of direct information from the policy level on environmental priority themes, key materials are recommended based on a two stage process in this report. First, key documents in the SD and SCP debate are reviewed for the identification of key environmental problems or impact categories. Second, UK results from a study by Van der Voet el al. (2005) for the UK are taken to analyse the environmental impact of 31 (finished) materials in 13 impact categories in the light of the Government’s SCP and SD agendas. Based on these two components a material is proposed to be of strategic importance for the development of more sustainable resource use patterns in the UK, if

a material has a particular high impact for a key environmental theme;

a material has a relatively high impact in most impact categories.

It should be noted that this prioritisation process does not take other economic and social issues into considerations. This would ideally be the case for SD and SCP policies.

3.12.3 Establishing Policy Priority AreasIn the absence of a robust quantitative method for prioritising environmental problems the government’s SD strategy was reviewed (DEFRA, 2005) with a particular focus on SCP to obtain an idea about the relevance of the various environmental impact themes. Throughout the document there is a strong emphasis on the priority status of climate change and the link with carbon dioxide and other greenhouse gas emissions. This is not only expressed by the fact that it is one of the four priority areas in the strategy together with Sustainable Consumption and Production, Natural Resource Protection and Sustainable Communities, but also through the identification of climate change by the government as “the greatest (environmental) threat” (DEFRA, 2005:72).

The role of the remaining 12 impact categories on the government’s agenda covered by the study of Van der Voet et al. (2005) is much more difficult to assess. Most of the problems such as land competition, acidification, euthrophication or ozone layer depletion are mentioned. However, it is not easy to identify any other environmental problem with a clear priority status. Within the SCP framework “waste” as a key objective receives a lot of attention and might be seen as a priority problem. A way to identify further environmental priority fields might be to count the occurrences of individual pollutants in the government’s SCP de-coupling indicator set. In such a case SO2, NOX and PM10 should be prioritised beside CO2. The importance of climate change, waste and these other air

SCP001 67/130

pollutants as DEFRA priority areas is confirmed by the Public Service Agreement with the Treasury (DEFRA, 2006).32

Pollutant Agent Impact/ EffectParticles (e.g. PM10) Human Health; Nuisance Dust; Soiling; Ecosystem

Degradation; Reduced Crop Quality and yield; Impacts on Visibility

Carbon Monoxide Human Health; Climate ModificationNOx Human Health; Acidification; Eutrophication;

Photochemical Oxidant FormationLead Human Health; Food Chain Effects; Impacts on

VegetationSulphur Dioxide Human Health; Acidification; Impacts on crops,

forests and natural vegetation; corrosionBenzene; Buta-1,3-diene Human Health (carcinogen)

Table 4.9 - Pollutant agents and their environmental impacts (Kuylenstierna et al. , 2002)Connecting pollutants to impact categories and identifying key environmental themes is a difficult task. As shown in Table 4.9 the air pollutants considered by the public service agreement, for example, have a variety of impacts across environmental themes. Concluding that all associated impacts are of key concern to government would be misleading. Indicators like PM10 or NOX might have been chosen just for their broad bandwidth of impacts across the various themes. This, does not mean that these individual impact categories themselves are of greater importance for the government. Furthermore prioritising environmental impact themes would be inappropriate because not all impacts that are important to the government are covered by the study of Van der Voet et al. (2005). It is, for example, reasonable to assume that human health considerations might have played an important role in the government’s decision to commit to reduction targets in PM10 emissions.33 However, it is not covered within impact categories. A clear link from substances to a particular environmental problem theme and then back to materials can therefore not easily be established. From this brief discussion it is therefore concluded that only climate change can be unambiguously identified as an environmental priority theme from the government’s SD/SCP agenda.

3.12.4 Prioritising materials based on environmental impactsHaving established that only climate change can be unambiguously established as a key environmental theme, the results of the study by Van der Voet et al al. (2005) as shown in Table 4.11 and Table 4.12 are analysed for the identification of key materials. Even though several limitations will be discussed later and are further summarised in Appendix I.6, this study provides the most comprehensive readily available evidence on a wider range of environmental impacts. It is important to note in the course of the interpretation of results that the 31 material groups under consideration are neither raw materials nor products, but finished materials. Finished materials are “materials just one step away from being applied in a product. Wheat and cotton are then materials, not bread or textile. Glass is the material, not windows or bottles, nor sand […]” (Van der Voet et al., 2005:34). This was seen by the authors’ as the most viable way to link EMFA and LCA data and avoid double counting problems. Even though this might be a bit more awkward to think about it does not add any substantial complications. In correspondence with the required global perspective and lifecycle view of the “Thematic Strategy on the Sustainable Use of Natural Resources” as well as the Government’s

32 Biodoverity is another environmental impact theme, which is covered by the Public Service Agreement. However, even though biodiversity features prominently in Chapter 5 of the SD strategy, it does not seem to be of any bigger importance in the course of the SCP agenda. Biodiversity is therefore not considered as a key environmental theme.

33 In fact, all air pollutants have detrimental effects on human health.

SCP001 68/130

SCP agenda, all impacts are then assigned to these finished materials from the cradle-to-the-grave in an LCI/A tradition.

Abbreviation Environmental Theme Unit of MeasurementADP Abiotic Depletion Potential kg antimony equivalentsLC Land competition m2/yearGWP Global Warming Potential kg CO2 equivalentsODP Ozone Depletion Potential kg CFC-11 equivalentsHTP Human Toxicity Potential kg 1.4 dichlorbenzene equivalentsTETP Terrestrial Ecosystem Toxicity

Potentialkg 1.4 dichlorbenzene equivalents

FAETP Freshwater Aquatic Toxicity Potential

kg 1.4 dichlorbenzene equivalents

MAETP Marine Aquatic Ecosystem Toxicity Potential

kg 1.4 dichlorbenzene equivalents

POCP Photochemical Oxidant Creation Potential

kg ethylene equivalents

AP Acidification Potential kg SO2 equivalentsEP Eutrophication Potential kg PO4 equivalents

radiation DALYfinal solid waste kg/kg

Table 4.10 - Environmental impact categories applied by Van der Voet et al. (2005)Table 4.11 ranks the results for the environmental impacts per kg of physical flows in the various categories, while Table 4.12 shows the results for aggregate impacts. The impact categories included with their abbreviations are listed in Table 4.10. As explained earlier a material might be considered of strategic importance here, if it either has a particularly high impact in the field of global warming (policy priority area) or if it has a high impact across most of the 13 environmental themes. By colouring the TOP10 impacts for each impact category in both tables, a fairly obvious pattern occurs:

SCP001 69/130

Table 4.11 - Ranks of environmental impacts per kg of materials across environmental problem themes derived from Van der Voet et al. (2005)

Impact per kg of material flow (see Table 4.11): According to the study by Van der Vart et al. (2005) throughout almost all impact categories ores and metals show the highest impact per kg of material flow consistently across environmental themes. Even in terms of global warming potential they rank higher than most fossil fuel based materials (4 of 5 materials in “TOP 5” fall into the category metals and ores). Copper, Nickel, Aluminium and Zinc therefore would be obvious candidates if materials were chosen that have a relatively high per unit flow environmental impact across environmental themes. Plastics34 also rank highly in most environmental impact categories. Biomass materials only have high impacts in particular environmental fields such as land competition or eutrophication.

Table 4.12 - Ranks of total environmental impacts of materials across environmental problem themes derived from Van der Voet et al. (2005)

Impact of aggregate material flows (see Table 4.12): When the flow volumes are brought into the picture and aggregate environmental impacts are compared within each impact category the picture changes quite radically. Across environmental impact themes plastics and fossil fuels seem to rank high consistently. The same is true for “paper and board” as well as for “animal fats”. Even though “iron and steel” rank consistently low in the various impact categories per unit of flow, the weight of its flows boosts the impact on an aggregate level. Again, all high impact materials in the key policy area of global warming are covered by the shortlist.

34 Even though plastics belong strictly speaking to the group of fossil-fuel based materials they will be treated separately in the course of the assessment here.

SCP001 70/130

Shortlist strategic environmental materialsImpact per unit of flow Aggregate environmental impact

Aluminium Animal FatsCopper Hard CoalNickel Iron and Steel

Plastics Natural GasZinc Oil

Paper and BoardPlastics

Table 4.13 - Shortlist of strategic materialsThe shortlists, which can be derived from looking at the aggregate and per unit impact of the various materials, is shown in Table 4.13. Instead of engaging in a long discussion about whether the focus should be on the materials with high aggregate or high per unit impacts, a much more pragmatic approach was taken. With the belief that this transparent approach will easily enable the government to review and rethink the suggestions made in the course of this report, the list was further reduced through reasoning:

Fossil Fuels: Fossil fuel based materials all rank comparatively highly across the environmental impact themes with the exception of brown coal. Thus they must be considered as strategic materials. However, it is the aim of the government to focus on these materials, which have not been traced on a regular basis so far. For fossil fuels large databases like the Environmental Accounts have already been established as the discussions on energy use (e.g. oil crisis) and climate change have been on the top of the political agenda for the last 40 years. Even though they are of strategic importance, fossil fuels therefore do not need to be considered in the remaining part of this discussion.

Ores and Metals: Amongst “the ores and metals” a focus on “aluminium” as well as “iron and steel” is recommended. Even though copper, nickel and zinc rank high on a per unit flow level their aggregate impacts tend to be below average in the various impact categories. However, this decision is arbitrary and can be revised by the government depending on its objectives.

Final Shortlist of materials of strategic environmental importanceAluminium

Iron and SteelPaper and Board

PlasticsAnimal Fats

Table 4.14 – Final Shortlist of strategic materials

3.12.5 Towards a modelling framework for tracing strategic materialsAs a result, a shortlist of five materials are derived, which might be of strategic importance for achieving more sustainable patterns of resource use. These materials are listed in Table 4.13. The next step is to propose a modelling framework that traces these key flows through the economy. In consistency with previous recommendations a hybrid framework, which combines bottom-up and top-down data, is proposed. However, in contrast to other hybrid approaches a two step procedure is recommended:

SCP001 71/130

In a first step bottom-up data could be compiled and presented as conventionally done in Bulk Material or Material System Studies. Within this report the Biffaward studies were taken as a representative for such an approach (see Section 3.3.2). An example for iron and steel production is shown in Figure 4.10. However, the data collection efforts should be organised in a way that the data is directly compatible with the sectoral classification of the national accounts in general, and the input-output publication in particular.

This bottom-up data could be the starting point for a second stage of analysis where the physical data would be framed into a hybrid input-output model. In this context industrial sectors would be further broken down according to project demands. For example, different types of steel could be distinguished within the steel sector and main raw materials inputs like iron ore could be separated out of the mining industry. This would be achieved through the combined use of available monetary and physical data as well as statistical optimisation methods.

Figure 4.10 - Studying the physical flows associated with aluminium and steel production (Dahlstroem at al., 2004)

This additional analytical stage provides an array of new opportunities for dealing with these flows through material prevention, material substitution and recycling:

SCP001 72/130

Framing the physical data in the input-output framework accounts for (a selection of) environmental impacts triggered by these flows of materials throughout the economy. This allows the development of a material management strategy, which takes into account dematerialisation, material substitution and reuse/recycling options and enables further prioritisation of policies.

Depicting the physical flows of key materials in the input-output context provides much more detailed information about their use throughout the economy. This helps policy-makers obtain a much clearer picture of why (for what) materials are cycling in the economy as well as where and how the associated environmental impacts might be best tackled. This is facilitated through detailed information about:

which sector or final demand category uses these materials (directly);

which final goods and services ultimately trigger the demand for these materials throughout the supply-chain;

where various (potential) environmental impacts created throughout the lifecycle of these materials occur in the supply chain. In this context a variety of analytical options such as Structural Path Analysis (see Defounrey and Throbecke, 1984) can be exploited. This incorporates the lifecycle perspective highlighted by government throughout the SD/SCP strategies.

The combination of monetary and physical data within one framework also provides a whole new array of analytical options as outlined previously. On the one hand, monetary and physical data can be analysed simultaneously. This means environmental indicators can be complemented with a whole array of socio-economic ones such as employment or GDP. On the other hand, the analysis of physical flows is brought into the scope of policy analysis and the efficiency and effectiveness of various policy instruments can be assessed a priori for example in scenario models.

Up-stream effects of imported materials associated with iron and steel production and consumption in the UK can also be accounted for. The first stage of data collection provides a coherent database. The second stage of the data collection and analysis framework addresses the remaining three key principles of the EC’s Thematic Strategy on the sustainable use of natural resources – a lifecycle view, a global perspective, an interdisciplinary approach35. These are also highlighted throughout the government’s SCP and SD strategy (DEFRA, 2005a). The general limitations of such an approach have already been discussed in Section 3.3.3; the remainder will discuss specific problems associated with data collection and modelling efforts.

3.12.6 Challenges for Data Collection and ModellingThe discussion associated with limitations imposed on data collection and modelling will be held first in general terms and then specifically for the five key materials. In general, it might be said that

bottom-up data is the more reliable and the more comprehensive in coverage, the less complex the industrial supply chain through which the material is transmitted.

35 Multi-disciplinary quantitative inquiries are best facilitated in data framework in multiple classifications and multiple units, which allows to represent all monetary and non-montary phenomena in the most appropriate way (see Keuning, 1994; Keuning, 2000).

SCP001 73/130

bottom-up data collection is more difficult, the more heterogenous the output of the producing entity. For example, it may be easier to compile data on the input requirements for the production of steel than for the production of a particular kind of plastic.

the hybrid model will be the easier to establish and the more reliable

the more homogenous the output of the sector producing the material under consideration;

the less disaggregation efforts need to be undertaken in the course of the monitoring effort;

the better the main suppliers are represented by the sectoral breakdown.

Even though these general considerations are by no means complete, they might serve as a good starting point for considerations of other materials not discussed here. For “paper and board”, bottom-up data can be obtained from annual production statistics. More detailed monetary and physical data has also been collected in the course of Paper Federation Surveys and from Integrated Pollution Prevention and Control Applications (see, SWEET and Enviros, 2003). No additional data needs arise from the sectoral aggregation level of the input-output table as the “paper and board” industry represents one industry in the UK’s IO publication and the main suppliers are well represented.

A similarly good data situation can be found for the aluminium as well as the iron and steel industry (see Dahlstroem et al., 2004; Michaelis and Jackson, 2000), where for example additional data can be obtained from the Iron and Steel Statistics Bureau. However, aluminium is intermingled in the production of all “non-ferrous” metals in the IO publication. Therefore, additional data needs to be collected and used to separate aluminium out. Production and trade statistics should provide a good starting point for doing so, though the data requirements will need to be looked at in the initial stage of the study.36

All the materials discussed so far share the characteristic that they are relatively homogenous and require only a limited amount of raw material inputs. This is not the case for plastics. This is of lesser concern as long as aggregate plastic flows are traced, but in the course of material flow policy - where prevention, substitution and recycling are constituting elements - it might be important to follow broader groups of available plastics. It can be expected that these plastics will have very different raw material inputs and pollution outputs throughout their lifecycle. The required disaggregation of the plastic industry according to broader groups of plastics might still be possible but the heterogeneity of the output in the plastic sector means this might be more difficult and less reliable than for more homogenous sectors. At the same time the data situation might also turn out to make these disaggregation efforts more difficult. To make a clear-sighted judgement the data situation needs to be explored further.

The most challenging judgment is regarding “animal fats”. In a first instance this is rooted in a definition problem – most importantly how “animal fats” are different from “animal protein”. In SIC coding “animal fats” are part of the activity “animal by-product processing”, which is subsumed under “production, processing and preserving of meat and meat products” in the input-output publication. Until it is clarified what falls into each category a clear evaluation of the situation is not possible. The required information should be obtainable from the authors of the Dutch study (see Van der Voet et al., 2005).

Once the definition problem is resolved the bottom-up data can be collected and the meat production sector can be disaggregated into “animal fats” and “non-animal fats”. It must be expected that a

36 For example, additional data from small scale surveys within the industry might need to be collected. Partial survey techniques can then be used to dis-aggregate the sectors (see Miller and Blair, 1985).

SCP001 74/130

substantial share of the physical flows in the supply chain is associated with agricultural activities. Despite the broad variety of agricultural activities associated with pastoral to arable farming, they are all lumped together within the available input-output table. This means that further disaggregation of the agricultural sector is required if the supply chain is to be appropriately represented. As well as proving a laborious undertaking, this may hit boundaries in terms of data availability but a good example of how the agricultural sector can be disaggregated can be found in Dimaranan and McDougall (2002). In the present context it can be concluded that the tracing of “animal fats” within the proposed framework (bottom-up and hybrid!) may not be straightforward.

3.12.7 Further considerations and recommendations for tracing strategic materials in the SCP policy context

The study by van der Voet et al. (2005) used to identify five key materials is the most comprehensive attempt to link material flows with environmental impacts across a wide range of impact categories. As previously emphasised it might be seen as the best available evidence for the identification of strategic materials based on environmental impacts. However the approach also has its limitations. Especially as the establishment of links between the material flow database of the Zero study (see Moll et al., 2003) and finished materials could only be undertaken in a rough manner.37 Further limitations are imposed by an array of uncertainties, data gaps and methodological issues as outlined in some detail in Appendix I.6.

Moreover, the lifecycle database used in the study dates back to 1997 and is therefore rather outdated.38 The database also uses West-European averages rather than UK-specific data to estimate the impact of different materials and does not cover all relevant materials (see van der Voet et al., 2005).

The five materials recommended in this section can therefore only be seen as a first set of suggestions and the government should consider commissioning a study which identifies strategic materials based on the best available data for the UK. The proposed shortlist can, and should, be further complemented with other materials, which are seen to be of great importance for the Government’s SCP agenda. This might comprise high impact materials covered in this assessment, but not considered in the shortlist, such as copper, nickel and zinc. The government might also decide that waste is a key environmental problem. In such a case high impact materials in this area such as concrete or plastics might be included. The transparent selection procedure and tables provided should enable the Government thoroughly to re-think the suggestions made. Moreover, materials, which have not been covered in the study by Van der Voet et al. (2005), might be seen as of strategic importance to the government. Although it is a low impact material, water is a good example as it provides such an important service to households and industry and its use may have implications for the environment in the South East of England especially.

The proposed data collection and analysis framework has been designed following the key principles of the EU Thematic Strategy as well as the Government’s SCP agenda. In particular, it allows policy makers to trace the materials and their associated environmental impacts throughout the supply chain. It also opens up a large variety of analytical options.

For three of the five proposed materials the data (as it could be for the other “ores and metals” zinc, copper and nickel) is available at the required levels. The same is true for plastics as long as aggregate

37 For the approach it would be more convenient to use the system boundaries of apparent consumption (for details see Van der Voet et al., 2005)

38 However, a new LCA database that could be used has become available in the meantime.

SCP001 75/130

plastic flows are concerned. For a more disaggregated treatment of plastics the availability of monetary and physical data would need to be further explored. The same is true for animal fats, which would require a more disaggregated treatment of some supply chain flows. Therefore, it is recommended that a framework for two or three unproblematic materials is developed first. In later stages materials with more complex supply-chains could be added.

SCP001 76/130

3.13 RecommendationsThis project has set out to assess the robustness and policy relevance of MFA methodologies for informing the government’s SCP agenda and their potential to contribute in the build-up of an evidence base. The assessment shows that all methodologies can contribute valuable evidence to inform SCP policy and they all have their individual strengths and weaknesses. To overcome these it is important for the government to take a leading role in

Establishing good practice for methodologies, where standardisation is not sufficiently developed;

Encouraging the integration of methodologies where possible to increase their robustness and applicability to the SCP agenda;

Orchestrating methodologies in an integrated material policy approach, which makes best used of the individual strengths of the methods and closely responds to the needs on the SCP agenda.

Moreover, MFA approaches should also

integrate other relevant social and economic information to allow for informed decisions;

make increasing use of scenario approaches, which can help to reduce uncertainties associated with future policies and help to identify the most appropriate policy instruments.

This leads to a wider set of key recommendations, which can be summarised as follows:

The government should aim to implement a strategic material flow approach to inform SCP policy, which focuses on materials with the most detrimental environmental impacts first. This would encourage the most effective use of existing MFA evidence and build-up of new MFA evidence. Such an approach aims to minimise the life cycle impacts of materials through dematerialisation including reuse/recycling as well as detoxification (material substitution). A general framework for prioritising material flows according to their environmental impacts and where physical evidence is required in this process has been proposed.

In order to prioritise material flows, MFA methodologies need to be linked with (potential) environmental impacts as commonly done in the LCA literature. Because the robustness of impact potential approaches could not be fully established within the scope of this report, the government should commission a review and shortlist those impact categories, which are robust enough to inform government policy. The review should further outline practical ways of how closer links between individual materials and environmental impacts can be drawn and should be carried out in an inter-disciplinary research team comprising ecologists, MFA as well as LCA experts.

A comprehensive model should be built-up to inform these prioritisation processes. Two complementary approaches have been outlined, which can achieve this. The EIOA based approach can be built-up in a modular, step by step process and should start in a key SCP policy area – waste can be suggested. Further, a generalised LCA based approach can be used for prioritising materials.

Results of this latter approach provided by Van der Voet et al. (2005) have been used to follow the request of the government to identify a selection of key materials with particularly high environmental impacts and devise a methodological framework in which they can be traced. A list of 5 strategic materials has been proposed. However, concerns about the robustness of the data used for the identification are highlighted.

SCP001 77/130

Finally, in the light of the importance of input-output tables for delivering robust and policy relevant evidence to inform the SCP agenda, an update of the UK’s monetary input-output table is urged.

Beside these key recommendations there is a whole array of others. They are summarised in the tables below, divided into methodology-specific recommendations and general recommendations.

Methodology RecommendationEconomy-Wide Material Flow Accounting

Use for controlling aggregate flow composition Utilise rich EMFA database by devising tailor-made indicators in area of relevance such as

waste, material cycling etc. Include indirect used foreign flows in DMC estimates Encourage improvement of waste data Encourage attempts to link material flows closer to environmental problems

BMFA/ MSA Use for tracing strategic materials Ensure good practice through reporting guidelines: “UK standard” Integrate with EIOA for tracing material flows where possible

EIOA Use for policy advise on structural level (when detailed meso view is required) Use for integration with other methods when complete and detailed supply-chain coverage

is needed Encourage compilation of 2000 IO tables Promote more timely publication of IO data and establishment of consistent time-series;

review where better environmental data is needed Make increased use of hybrid IO LCIs Use for prioritization policies – consider integration with data from LCA databases

LCI/A Minimum Requirement: Ensure that in the compilation of LCA data evidence, clear reference is made to the underlying LCA database as results are not easily comparable across databases

Consider review lifecycle databases and assess quality Review robustness of various environmental impact categories Promote use of appropriate hybrid IO LCA methodologies to increase comparability and

robustness of LCI resultsSFA Use when tracing of particular hazardous or toxic substance is required

Use for establishing links between material and substance flows Establish good practice through reporting guidelines: „UK standard“

Table 4.15 - Specific Recommendations derived from the review of the individual methodologies

Concern RecommendationDevelopment of SCP Evidence Base

Aim to establish strategic build-up of SCP evidence base: a model is required for the identification of key physical flows (substances, raw material inputs, products and services, waste outputs)

Consider tracing of proposed strategic materialsEstablishing Good Practice

Data Collection Guidelines Reporting Guidelines (how results should be presented that they are most useful for

policy-makers and Environmental Impacts Commission study reviewing impact potential approaches as commonly used in LCAsIntegrative Policy Approach

Encourage integration with economic and social issues in ex-ante and ex-post analysis Encourage use of quantitative scenarios as decision-support tool

Table 4.16 - General Recommendation

SCP001 78/130

3.14 ReferencesAdriaanse, A., Bringezu, S., Hammond, A., Moriguchi, Y., Rodenburg, E., and Rogich, D., 1997 Resource

flows - The Material Basis of Industrial Economies, World Resources Institute (WRI), Washington.

Albritton, D., Derwent, R., Isaken, I., Lal, M., Wuebbels, D., 1996, Trace Gas Radioactive Forcing Indices, in: Houghton, J.T., Meihra Filio, L.G., Callandar, B.A., Harris, N., Knattenberg, A., Maskell, K. (eds), Climate Change 1995, Cambridge University Press, Cambridge.

Audit Commission, 2005, Local quality of life indicators: supporting local communities to become sustainable, Audit Commission, London.

Ayres, R.U. and Simonis, U.E., 1994, Industrial Metabolism. Restructuring for Sustainable Development, Tokio/ New York/ Paris, United Nations University Press.

Azapagic, A. and Clift, R., 1999, Life Cycle Assessment and Multiobjective Optimisation, Journal of Cleaner Production 7: 135-143.

Bancini, P. and Brunner, P., 1991, Metabolism of the Anthroposphere, Springer, Berlin.

Bell, D.N.F., 2003, Regional Econometric Modelling in the UK: A Review, Regional Studies 27(8): 777-782.

Binswanger, M., 2001, Technological Progress and Sustainable Development. What About the Rebound Effect?, Ecological Economics 36: 119-132.

Bringezu, S., 1997, From Quantity to Quality: Material Flow Analysis, in: Bringezu, S. (ed.): Regional and National Material Flow Accounting, Proceedings of the ConAccount Workshop, 21-23 January, Leiden.

Bringezu, S., 2000, History and Overview of Material Flow Analysis, Paper presented to the OECD Special Session on Material Flow Accounting, 24 October, OECD, Paris.

Brunner, P.H. and Rechberger, H., 2004, Material Flow Analysis, Lewis Publishers, London.

Bringezu,S., Schutz,H., Steger,S., and Baudisch,J., 2004. International comparison of resource use and its relation to economic growth: The development of total material requirement, direct material inputs and hidden flows and the structure of TMR. Ecological Economics, 51:97-124 pp.

Bullard, C.W., Penner, P.S., Pilati, D.A, 1978, Net Energy Analysis – Handbook for Combining Process and Iput-Output Analysis, Resources and Energy 1: 267-313.

Bullard, C.W. and Sebald, A.V., 1988, Monte Carlo Sensitivity Analysis of Input-Output Models, Review of Economics and Statistics 70(4): 708-721.

Brunner, P.H. and Rechberger, H., Practical Handbook of Material Flow Analysis, Lewis Publishers, New York.

Carlson, R. and Palsson, A.-C., 2003 Guideline for Collection, Treatment and Quality Documentation of LCA Data, Paper presented at the International Workshop on Quality of LCI Data, October, Karlsruhe.

Chambers, N., Simmons, C. and Wackernagel, M., 2001, Sharing Nature’s Interest, Earthscan, London.

Chen, P., Frohn, J. and Lemke, W., 2005, Linear and Nonlinear Identities in Simultaneous Structural Econometric Models, Applied Economics Quarterly 55(2): 169-180.

Cooper, J., Vigon, B., Curran, M.A., Franklin, B., 2005, Life Cycle Assessment in Managemant, Product and Process Design, and Policy Decision Making. A Conference Report, Integrated Environmental Assessment and Management 1(1): 60-65.

Dahlstroem, K., Ekins, P., He, Juanchun, Davis J. and Clift, R., 2004, Iron, Steel and Aluminium in the UK: Material Flows and Their Economic Dimension, Executive Summary Report, University of Surrey.

Daly, H., 1994, For the Common Good, Redirecting the Society Toward Community, the Environment, and a Sustainable Future, 2nd updated and expanded edition, Beacon Press, New York.

Daniels,P.L., 2002. Approaches for Quantifying the Metabolism of Physical Economies: A Comparative Survey, Part II: Review of Individual Approaches. Journal of Industrial Ecology, 6:65-88 pp.

Daniels,P.L. and Moore,S., 2002a. Approaches for Quantifying the Metabolism of Physical Economies: Part I: Methodological Overview. Journal of Industrial Ecology, 5:69-93 pp.

SCP001 79/130

Daniels,P.L. and Moore,S., 2002b. e-supplement to "Approaches for Quantifying the Metabolism of Physical Economies: Part I: Methodological Overview". Journal of Industrial Ecology, 5:69-93 pp.

Defourny, J. and Thorbecke, E., 1984, Structural path analysis and multiplier decomposition within a social accounting matrix framework, Economic Journal 94: 111-/136.

De Bruyn, S., Sevenster, M.N., Warringa, G.E.A., van der Voet, E. and van Oers, L., 2004, Economy Wide Material Flows and Environmental Policy, Report for the Netherlands Ministry of Housing, Spatial Planning and the Environment, CE, Delft.

De Haes, U., Van der Voet, E., Kleijn, R., 1997, Substance Flow Analysis – An Analytical Tool for Integrated Chain Management, in: Bringezu, S. (ed.): Regional and National Material Flow Accounting, Proceedings of the ConAccount Workshop, 21-23 January, Leiden.

De Haes, U., Heijungs, R., Suh, S. and Huppes, G., 2004, Three Strategies to Overcome the Limitations of Life-Cycle Assessment, Journal of Industrial Ecology 8(3): 19-32.

DEFRA, 1999, A Better Quality of Life: A Strategy for Sustainable Development for the UK, DEFRA Publications, London.

DEFRA, 2001, Total Material Resource Flows of the UK Economy. A Report by the Wuppertal Institute, London.

DEFRA, 2002, Ressource Use and Efficiency of the UK Economy. A Report by the Wuppertal Institute, London.

DEFRA, 2003, Changing Patterns. UK Government Framework for Sustainable Consumption and Production, DEFRA Publications, London.

DEFRA, 2005a, Securing the Future. Delivering UK Sustainable Development Strategy, TSO Stationary Office, Norwhich.

DEFRA, 2005b, Sustainable Consumption and Production – Development of an Evidence Base. Study of Ecological Footprinting, London.

DEFRA, 2005c, An International Peer Review of a DEFRA Funded Report Sustainable Consumption and Production – Development of an Evidence Base Study of Ecological Footprinting, London.

DEFRA, 2006, Public Service Agreement 2004-2008, http://www.defra.gov.uk/corporate/busplan/psa2004.htm (last accessed 14/05/2006).

Destatis, 2005, Umweltnutzung und Wirtschaft. Bericht zu den Umweltoekonomische Gesamtrechnungen 2005, Statistisches Bundesamt, Wiesbaden.

Dimaranan,B.V. and McDougall, R.A., 2002, Global Trade, Assistance, and Production: The GTAP 5 Data Base, Chapter 14.5, Center for Global Trade Analysis, Purdue University.

DTI, 2000, DTI Sustainable Development Strategy. A Strategy for Simultaneously Securing Economic Growth and Environmental Protection, DTI Publications, London.

Dietzenbacher, E., 2005, Waste Treatment in Physical Input-Output Analysis, Ecological Economics 55(1): 11-23.

Dorfman, R., Samuelson, P.A. and Solow, R.M., 1958, Linear Programming and Economic Analysis, McGraw-Hill, New York.

Ekins, P. and Etheridge, B., 2006, The Environmental and Economic Impatcs of the UK Climate Change Requirements, Energy Policy, forthcoming.

Ekvall, T., Person, L, Ryberg, A., Widheden, J., Frees, N., Nielsen, P.H., Weidema, B. and Wesnaes M., 1998, Lifecycle Assessment of Packaging Systems for Beer and Soft Drink, Miljoprojekt No. 399, Ministry of Environment and Energy, Copenhagen, Denmark.

ERM, 2005, Scoping of an Environment Direct Service, ERM Final Report, London.

European Commission (EC), 2005a, Thematic Strategy on the Sustainable Use of Natural Resources, European Commission, Brussels.

European Commission (EC), 2005b, Thematic Strategy on the Sustainable Use of Natural Resources – Impact Assessment, European Commission, Brussels.

SCP001 80/130

European Environmental Agency (EEA), 1997, Life Cycle Assessment, A Guide to Approaches, Experiences and Information Sources, Environmental Issues Series 6, Copenhagen.

Europen, 1999, Use of Life Cycle Assesssment (LCA) as a Tool in the Field of Sustainable Packaging and Waste Management, Europen Discussion Paper, Brussels.

Eurostat, 1999, The Policy Relevance of Material Flow Accounts, Eurostat Working Papers 2/1999/B/1.

Eurostat, 2001, Economy-wide material flow accounts and derived indicators. A methodological guide, , Office for Official Publications of the European Communities, Luxembourg.

Eurostat. Material use in the European Union 1980-2000: Indicators and Analysis. 2002. Luxembourg, Office for Official Publications of the European Communities, Statistical Office of the European Communities (Ed.). Working Papers and Studies Series.

Federal Government of Germany, 2002, Perspectives for Germany – Our Strategy for Sustainable Development, Berlin.

Federal Government of Germany, 2004, Progress Report 2004 Perspectives for Germany – Our Strategy for Sustainable Development, Berlin. (Bundesregierung, 2004, Fortschrittsbericht 2004 Perspektiven fuer Deutschland – Unsere Strategie fuer eine Nachhaltige Entwicklung, Berlin.

Femia, A. and Moll, S., 2005, Use of MFA-Related Family of Tools In Environmental Policy-Making. Overview of Possibilities, Limitations and Existing Examples of Application In Practice, European Environment Agency, European Topic Centre on Waste and Material Flows, Copenhagen.

Fischer-Kowalski, M., 1997, Society’s Metabolism: On the Childhood and Adolescence of a Rising Conceptual Star, in: Redclift, M. and Woodgate, G., The International Handbook of Environmental Sociology, Edward Elgar, Cheltenham.

Frischknecht, R., Oekoinventare fuer Energiesysteme, Parts 1-3 and Dataset, Institut fuer Energietechnik, ETH Zuerich.

Frohn, J., 1998, Zum Nutzen makroökonomischer Modelle (On the benefits of macro-economic models), in: Ifo Studien, No. 44, Bielefeld.

Gazley, I. and Francis, P., 2005, UK Material Flow Review, Office for National Statistics, London.

Grubb, M. (ed.), 1999, The Kyoto Protocol – A Guide and Assessment, Royal Institute of International Affairs, London.

Guinee, J., Udo de Haes, H. and Huppes, G., 1993 Quantitative Life Cycle Assessment of Products, Journal of Cleaner Production 1(1): 3-13.

Hansen, E., 1997, Experiences with SFAs on the National Level for Hazardous Substances in Denmark, in: Bringezu, S. (ed.): Regional and National Material Flow Accounting, Proceedings of the ConAccount Workshop, 21-23 January, Leiden.

Heijungs, R., 1994, A Generic Method for the Identification of Options for Cleaner Products, Ecological Economics 10: 69-81.

Heijungs, R., 1996, Identification of Key Issues for Further Investigation in Improving the Reliability of Life-Cycle Assessments, Journal of Cleaner Production 4(3-4): 159-166.

Heinjungs, R. and Suh, S., 2002, The Computational Structure of Life Cycle Assessment, Kluwer Academic Publisher, Dordrecht, Netherlands.

Heijungs, R. and Suh, S., 2006, Reformulation of Matrix Based LCI: From Product Balance to Process Balance, Journal of Cleaner Production 14: 47-51.

Hertwich, E., 2005, Consumption and the Rebound Effect: An Industrial Ecology Perspective, Journal of Industrial Ecology 9(1-2): 85-98.

Hubacek, K. and Giljum, S., 2003. Applying physical input–output analysis to estimate land appropriation (ecological footprints) of international trade activities. Ecological Economics 44(1): 137–151.

Hashimoto, S. and Moriguchi, Y., 2004, Proposal of Six Indicators of Material Cycles for Describing Society’s Metabolism: From the Viewpoint of Material Flow Analysis, Resources, Conservation and Recycling 40: 185-200

SCP001 81/130

Hendrickson, C., Horvath, A., Joshi, S. and Lave, L.B., 1998, Economic Input-Output Models for Environmental Lifecycle Assessment, Environmental Science and Technology 32: 184A-191A.

Jalas, M., 2002, A Time Use Perspective of the Material Intensity of Consumption, Ecological Economics 41: 109-123.

Joshi, S., 1998, Comprehesive Product Life-Cycle Analysis Using Input-Output Techniques, PhD Dissertation, John H. Heinz School of Public Policy and Management, Carnegie Mellon University, Pittsburgh, USA.

Joshi,S., 2000, Product Environmental Life-Cycle Assessment Using Input-Output Techniques. Journal of Industrial Ecology, 3:95-100 pp.

Juster, F.T., Courant, P.N., and Dow, G.K., 1981, A Theoretical Framework for the Measurement of Well-Being, Review of Income and Wealth 27(1):1-32.

Keuning, S.J., 1994, The SAM and Beyond: Open SESAME!, Economic Systems Research 6(1): 21-50.

Keuning, S.J., 2000, Accounting for Welfare with SESAME, in: United Nations, Handbook of National Accounting. Household Accounting, Experience in Concepts and Compilation, Vol. 2, Studies in Methods F, No. 75, New York: 273-307.

Kondo, Y. and Nakamura, S., 2004, Evaluating Alternative Life-Cycle Strategies for Electrical Appliances by the Waste Input-Output Model, International Journal of Life Cycle Assessment 9(4): 236-246.

Kuylenstierna, J.C., Hicks, K.W., Chadwick, M.J., 2002, A Perspective on Global Air Pollution, Global Environment Change 17: 21-56.

Lave, L.B., Cobas-Flores, E., Hendrickson. C.T. and McMichael, F.C., Using Input-Output Analysis to Estimate Economy-wide Discharge, Environmental Science and Technology 29(9): 420A-426A.

Lawson, N., Waghorn, D., Ravetz, J. and Douglas, I., 2003, UK Material Flow Accounts – Review of Indirect Flow Coefficients, Report to ONS, University of Manchester.

LCI Consultants, 2006, Current Issues in Life Cycle Assessment, http://www.lca-net.com/current/ (last accessed: 13/04/2006).

Lenzen, M., 2001, Uncertainty in IO-Based LCI, in: Nielsen, A.M. and Weidema, B.P. (eds.), Input-Output Analysis – Shortcuts to Life Cycle Data?, Project Report, Project No. 581, Danish Environmental Protection Agency, Copenhagen.

Lenzen,M., 2002, A guide for Compiling Inventories in Hybrid Life-Cycle Assessments: some Australian Results. Journal of Cleaner Production 10: 545-572.

Lenzen, M., 2003, Environmentally Important Paths, Linkages and Key Sectors in the Australian Economy, Structural Change and Economic Dynamics 14: 1-34.

Lenzen, M., Pade, L.-L., and Munksgaard, J., 2004, CO2 Multipliers in Multi-region Input-Output Models, Economic Systems Research 16: 391-412.

Lenzen, M., 2005, Balancing Act. A Triple Bottom Line Approach, Report for the Ministry of Environment Australia, University of Sydney.

Lenzen, M., 2006, Triple-Bottom-Line Accounting - a new software with a case study of Australian mining, Workshop on Material Flows and Environmental Impacts behind International Trade of Japan, Akihabara Convention Hall, Sotokanda, Chiyoda-ku, Tokyo, February 27, http://www.nies.go.jp/event/kaigi/topics/20060227/3Lenzen%20.pdf (last aaccessed: 15/05/2006).

Linstead, C. and Ekins, P., 2001, Mass Balance UK, Mapping UK Resource and Material Flows, Forum for the Future, London.

Linstead, C., Gervais, C. and Ekins, P., 2003, Mass Balance. An Essential Tool for Understanding Resource Flows, Forum for the Future, London.

MacGillivray, A. and Levett-Therivel, 2004, Consumer Expenditure and Environmental Impact: Finding the Right UK Model, Report for the Sustainable Consumption Roundtable, London.

Matthews, E., Amann, C., Bringezu, S., Fischer-Kowalski, M., Huettler, H., Kleijn, R., Moriguchi, Y., Ottke, C., Rodenburg, E., Rogich, D., Schandl, H., Schuetz, H., van der Voet, E. and Weisz, H., 2000, The weight of nations: Material outfows from industrial economies, World Resources Institute, Washington, DC.

SCP001 82/130

McGregor, P.G., McLellan, D., Swales, J.K. and Turner, K., 2004, Attribution of Pollution Generation to Local Private and Public Demands in a Small Open Economy: Results from a SAM-based Neo-classical Linear Attribution System for Scotland, Strathclyde Discussion Papers in Economics no 8.

McLaren, D., 2001, Environmental space, equity and the ecological debt, in: Aygeman, J. et al. (eds), Just sustainabilities: Development in an unequal world. MIT Press, Cambridge.

Miller, R.E. and Blair, P.D., 1985, Input-Output Analysis. Foundations and Extensions, Prentice Hall, Englewood Cliffs, NJ.

Millet, D., Bistagnino, L, Lanzavecchia, C., Camous, R. and Poldma, T., 2006, Does the Potential of the Use of LCA Match the Design Team Needs?, Journal of Cleaner Production, forthcoming.

Ministry of the Environment Japan, 2003, Fundamental Plan for Establishing a Sound Material Cycle Society, see http://www.env.go.jp/en/rep/waste/plan_sound.pdf (last accessed: 05.02.2006)

Moll, S., Bringezu, S. and Schuetz, H., 2003, Resource Use in European Countries, Report for the European Topic Centre on Waste and Material Flows, Copenhagen.

Michaelis, P. and Jackson, T., 2000, Material and Energy Flow Through the UK Iron and Steel Sector, Part 1: 1952-1994, Resources, Conservation and Recycling 29: 131-156.

Moriguchi, Y., Kondo, Y., and Shimizu, H., 1993, Analyzing the life cycle impact of cars, Industrial Environment 16(1-2): 42-45.

Munksgaard, J., Minx, J.C., Christoffersen, L., and Pade, L.-L., 2005, Models for National CO2 Accounting, in: Suh, S. and Lenzen, M., Handbook of Input-Output Methods for Industrial Ecology, North Holland Publishing, forthcoming.

Nakamura, S. and Nakajima, K., 2005, Waste Input-Output Material Flow Analysis of Metals in the Japanese Economy, Materials Transactions 46(12): 2550-2553.

Nakamura, S. and Kondo, Y., 2006, A Waste-Input-Output Life Cycle Cost Analysis of the Recycling of End-of-Life Electrical Home Appliances, Ecological Economics, forthcoming.

Nielsen, P.H. and Wenzel, H., 2002, Integration of Environmental Aspects in Product Development: A Stepwise Proceudre Based on Quantitative Life Cycle Assessment, Journal of Cleaner Production 10: 247-255.

Nielson, A. and Weidema, B., 2001, Input-output analysis - Shortcuts to life cycle data?., Danish Environmental Protection Agency, Copenhagen.

Nordhaus, W.D. and Tobin, J., 1972, Is Growth obsolete?, in: Bureau of Economic Research (ed.), Economic Growth, 50th Anniversary Colloquium Series 96, National Bureau of Economic Research, New York.

ONS, 2002, United Kingdon Analytical Input-Output Tables, ONS Publications, London.

ONS, 2005, Environmental Accounts – Autumn 2005, Press Release, ONS Publications, London.

ONS and DEFRA, 2005a, Sustainable Development Indicators in Your Pocket 2005, DEFRA Publications, London.

ONS and DEFRA, 2005b, Sustainable Consumption and Production Indicators. Basket of ‘Decoupling’ Indicators, DEFRA Publications, London.

Opschoor, J. B. and Reijnders, 1991, Towards Sustainable Development Indicators, in: Verbruggen, H. and Kuik, M., In Search of Indicators for Sustainable Development, Kluwer Academic Press, Dordrecht, Netherlands.

Pedersen, O.G., 1998, Physical Input–Output Tables for Denmark, Products and Materials 1990, Air Emissions 1990–92. Statistics Denmark, Copenhagen.

Princen, T., Maniates, M. and Coca, K., 2002, Confronting Consumption, MIT Press, London.

Proops, J.L.R., 1991, National Accounting and the Environment, Report to the World Bank, Keele University.

Saouter; R. and Van Hoof, G., 2002, A Database for the Life-Cycle Assessment of Procter and Gamble Laundry Detregents, International Journal of Life Cycle Assessment (2): 103-114.

Schmidt-Bleeck, F. et al, Statement to Government and Business Leaders, Wuppertal Institute, Wuppertal.

SCP001 83/130

SEI, CURE and WWF, 2006, Counting Consumption – CO2 Emissions, Material Flows and Ecological Footprint of the UK by Region and Devolved Country, WWF UK, Surrey.

SETAC, 2002, Life-Cycle Impact Assessment: Striving towards best practise, SETAC- Press, Pensacola, Florida..

Simmons, C., Lewis, K. and Barrett, J., 2000, Two feet – two approaches: a component based model of ecological footprinting, Ecological Economics 32: 375-380.

Sinclair, P., Papathanasopoulou, E., Mellor, W., and Jackson, T., 2005, Towards an Integrated Regional Material Flow Accounting Model, Journal of Industrial Ecology 9(1-2): 69-84.

Sonis, M., Hewings, G.J.D., 1999, Economic landscapes: multiplier product matrix analysis for multiregional IO systems. Hitotsubashi Journal of Economics 40, 59-/74.

Spangenberg, J., 1994, Towards Sustainable Europe, Friends of the Earth Europe, Brussels.

Stahmer, C, Kuhn, M. and Braun, N., 1997, Physical Input-Output Tables for Germany 1990, Eurostat Working Papers 2/1998/B/1, Luxembourg.

Stahmer, C., 2000, The magic triangle of input-output, Paper presented at the XIII International Conference on Input-Output Techniques, 21-25 August, Macerata, Italy.

Stahmer, C. 2004. Social accounting matrices and extended input-output tables. In OECD (ed): Measuring sustainable development: Integrated economic, environmental and social frameworks, OECD, Paris: 313–344.

Suh, S. and Huppes, G., 2005, Methods for Life Cycle Inventory of a Product, Journal of Cleaner Production 13: 687-697.

Suh,S., Lenzen,M., Treloar,G.J., Hondo,H., Horvath,A., Huppes,G., Jolliet,O., Klann,U., Krewitt,W., Moriguchi,Y., Munksgaard,J., and Norris,G., 2004, System boundary selection in life-cycle inventories using hybrid approaches. Environmental Science & Technology 38: 657-664.

Suh, S., 2004. A note on the calculus for physical input–output analysis and its application to land appropriation of international trade activities. Ecological Economics 48(1): 9 –17.

Suh,S., 2004c, Functions, commodities and environmental impacts in an ecological-economic model, Ecological Economics, 48:451-467.

SWEET and Enviros, 2003, Mass Balance for the UK Paper and Board Industry, Project Report, SWEET, Bristol.

Takase, K., Kondo, Y., and Wishizu, A., 2005a, An Analysis of Sustainable Consumption by the Waste Input-Output Model, Journal of Industrial Ecology 9(1-2): 201-220.

Takase, K., Kondo, Y. and WASHIZU, A., 2005b, Consumers' Behavior and Environmental Impact of Time Use: An analysis by the Waste Input-Output Model and a Consumer Model, EcoDesign 2005 Conference, The Fourth International Symposium on Environmentally Conscious Design and Inverse Manufacturing, December, 12-14, Tokyo.

Tan, R.R. and Culaba, A.B., 2002, Environmental Life Cycle Assessment: A Tool for Public and Corporate Policy Development, http://www.lcacentre.org/library/pdf/PSME2002a.pdf (last accessed 10.04.2006)

Tinbergen, J., 1956, Economic Policy: Theory and Design, North Holland Publishing, Amsterdam.

Treloar, G., 1997. Extracting embodied energy paths from input-/output tables: towards an input /output based hybrid energy analysis method, Economic Systems Research 9: 375-/391.

Tukker, A., Huppes, G., Guinee, J., Heijungs, R., de Koning, A., van Oers, L., Suh, S., Geerken, T., van Holderbeke, M., Jansem, B. and Nielsen, P., 2005, Environmental Impacts of Products, Institute for Prospective Technology Studies, Sevilla, Spain.

UK Government, 2005, Sustainable Development - The Government’s Approach – Delivering UK Sustainable Development Together, http://www.sustainable-development.gov.uk/what/what-you.htm (last accessed: 09 February 2006).

UN, 1992, Agenda 21, United Nations Publications, New York.

UN, 1993, Systems of National Accounts, New York.

SCP001 84/130

UN, 2003, System for Integrated Environmental and Economic Accounting

Van Berkel, R., 2002, The Application of Life Cycle Assessment for Improving the Eco-Efficiency of Supply Chains, Proceedings of the Muresk 75th Anniversary Conference, 3-4 October, Perth, Australia.

Van der Voet, Ester, van Oers, Lauran, Moll, Stephan, Schuetz, Helmut, Bringezu, Stefan, de Bruyn, Sander, Sevenster, Maartje, and Warringa, Geert. Policy Review on Decoupling: Development of indicators to assess decoupling of economic development and environmental pressure in the EU-25 and AC-3 countries. CML report 166, Department Industrial Ecology. 2005. Leiden, NL, Institute of Environmental Sciences (CML), Leiden University.

Vaze, P., 1998, A NAMEA for the UK, Structural Change and Economic Dynamics 10: 99-121.

Vigon, B.W., Tolle, D.A., Cornaby, B.W., Latham, H.C., Harrison, C.L., Boguski, T.L., Life Cycle Assessment: Inventory Guidelines and Principles, Environenmental Protection Agency, EPA/6000/R-92/245, USA.

Von Weizsaecker, E.-U., Lovins, A.B and Lovins, L.H., 1997, Factour Four, Doubling Wealth – Halving Resource Use, Earthscan, London.

Wackernagel, M. and Rees,W., 1996, Our Ecological Footprint, New Society Publishers, Gabriola Island, BC.

Wenzel, H., Hauschild, M. and Alting, L., 1997, Environmental Assessments of Products, Vol 1: Methodologies, Tools and Case Studies in Product Development, Chapman & Hall, London.

Weisz, H. and Duchin, F., 2006, Physical and Monetary Input-Output Analysis. What Makes the Difference?, Ecological Economics, forthcoming.

Wiedmann, T., Minx, J.C., Barrett, J. and Wackernagel, M., 2006, Allocating Ecological Footprints to Household Consumption Activities by Using Input-Output Analysis, Ecological Economics 56(1): 28-48.

Wright, E., Azapagic, A., Stevens, G., Mellor, W. and Clift, R., 2005, Improving Recyclability by Design: A Case of Fibre Optic Cable, Recources, Conservation and Recycling 44: 37-50.

SCP001 85/130

4 Review of Biffaward Studies(by Robin Vanner and Paul Ekins)

4.1 IntroductionThe Biffaward series of studies was launched following the publication by Biffa of a number of overview publications (Biffa 1997) relating to resource use and waste management in the UK, which identified the general paucity of data relating to material flows at that time. The rationale for commissioning the series of studies was that information about resource flows though the UK economy is of fundamental importance to the cost effective management of those flows, especially at the point at which those flows become ‘wastes’.

The overall objective of the Biffaward studies was to enable there to be built up over time a comprehensive database of material flows for the UK, covering all (significant) materials, and broken down according to the different materials, different major products, different economic sectors and different geographic regions.

During the commissioning of the earlier studies it became clear that this grand design might not be realised without some coordination of the projects. In 2000 Forum for the Future were commissioned to undertake the Mass Balance UK project. This project was established to co-ordinate and give support to Biffaward funded projects, and in particular to ensure that the studies categorised and classified their data in the same way, so that the data could be placed in a single database, related to other sources and provide the basis for the realisation of the overall objective over time (see Linstead & Ekins 2001, Linstead et al. 2003).

The studies were commissioned and administered through the Royal Society for Wildlife Trusts (RSWT), and funded through the landfill tax credit scheme. This funding stream is due to come to an end at the end of March 2006, at which point a sum in excess of £10.6 million will have been spent (Biffaward 2006, p.109). A funding requirement of all of the studies was that 10% of funds had to come from third parties. This requirement will have shaped the nature of many of the studies. This funding requirement should have ensured that the project team had developed some kind of genuine relationship with the relevant sector. These relationships will also have led to greater and wider expectations of what purpose the studies should serve.

4.2 The Biffaward StudiesAs shown in Figure 5.11, the various projects attempt to quantify the movement of material through specific economic sectors as resources are extracted and/or imported and products are manufactured. Some of these resources will be consumed within the manufacturing process; others will be embedded into products. In the process of consumption these products will become wastes, part of which will be recycled back into production, and the rest will need to be managed or disposed of in some way. This is the way that the studies have been grouped for the purposes of this assessment, and how the conclusions will be organised.

SCP001 86/130

Figure 5.11 The groupings for the Biffaward studiesThe main place where the Biffaward studies are now available is the Mass Balance website (http://www.massbalance.org). This is the central source of the study reports which have been reviewed. 52 studies as listed on the website (plus an additional study not listed) are shown in Table II.4 of Appendix II. The project code as used in the mass balance website has been provided to assure consistent identification of the studies. Not all of the reports were available for download from the mass balance web site at the beginning of the assessment process. Contact was made with Jonathan Nobbs of The Royal Society of Wildlife Trusts to capture reports which were nearing completion but had not been posted on the mass balance website. The studies have been grouped in the categories as shown in Figure 5.11.

4.3 The Assessment4.3.1 The process of assessmentThe assessment methodology is described below and in detail in appendix B. The content assessed in the first instance was the report as provided on the mass balance website. Additional information was only sought where it was thought necessary to complete the assessment, and then only when further material was referred to in this initial report or on the Biffaward website. Sometimes extra information was sought from Jonathan Nobbs of the RSWT. In the event that clarification (for example, relating to the availability of data) was required that could not be attained from RSWT, contact was made with authors on this basis. Notes were made by the assessor providing reasons for particular assessments and are available along with other detailed information about each study in appendix B. Consultation of authors and users of the reports was held with the interim results. Responses were received relating to 18 of the studies, and changes made where it was successfully demonstrated that the interim assessment did not reflect the content available to the assessor. Where a response was received for a study, this is shown in a column on Figure 5.12. In addition to this consultation was had with Jonathan Nobbs of the RSWT and David Aeron-Thomas from Forum for the Future, as well as all those invited to the project workshop.

SCP001 87/130

Production

Primary &

secondary

sectoral studies

Service sector

studies

Products

Product studies

Consumption

Regional,

tourism & the

eco-budget

studies

Waste

Waste studies

Material studies

4.3.2 Assessment specificationsAs discussed in Section 3, the assessment methodology as used throughout this project sought to identify excellence, as well as to identify those studies which are sufficiently robust and useful to be used to inform future policy. Failure to achieve a satisfactory assessment in any one of the criteria considered to be essential has led to a study’s exclusion. The generic distinction between satisfactory and unsatisfactory which has been applied when setting the specifications for the criteria is ‘the minimum required for the reports recommendations to be valid’. The specific assessment specifications as used to assess the Biffaward studies are shown in Table II.1 of Appendix B.

For two of the assessed criteria (peer-review and endorsement), the assessment was made based on information provided during the consultation. There are therefore many gaps for studies who did not respond. It was decided not to downgrade studies based on an unknown for these two criteria. However, any score awarded to studies that did respond positively to the consultation, or who had already demonstrated compliance within their report still contributed towards the study’s total score. The different status of these two criteria is shown in Figure 5.12 by their heading being coloured white rather than yellow or orange.

Most of the criteria have been assessed solely by following these specifications. However, this is not the case for all of the criteria relating to policy. A number of questions arose in relation to the assessment of the relevance of the Biffaward studies for environmental policy:

1. Ex ante analysis: How did the studies themselves conceive their relevance to policy within the study’s objectives?

This has been assessed with reference to the objectives provided, in this case taken to be the objectives as stated on the mass balance website.

2. Ex post analysis: In reality, how did the studies relate to policy? Were their objectives achieved, how were the study’s outputs used and were they endorsed by other practitioners or policy making authority?

This has been assessed by consulting the authors and users of the data, inviting them to report how the report has been used within policy and who has endorsed the work.

3. Future-oriented analysis: Given the nature and outcome of the studies, are there emerging policy areas to which they might have been relevant? What new or different data, or changes in methodology, might have made them, or might make them, more relevant to emerging agendas? Are there particular policy questions which they or the methodologies they use seem able to address?

4. This has been achieved by first reflecting on whether the objectives were actually achieved based on a reading of the report, and then identifying where the study will have future policy use within one of the policy agenda areas as set out in Table B4 in appendix B. This assessment sets out to distinguish the most useful studies which could feed directly into and provide clear guidance to policy maker, from those studies which merely provide background information to policy makers.

The outcomes from these questions are reflected in the outcomes as shown in Figure 5.12 (criteria P1, 2 & 4) and in detail for each study in the assessment notes in Appendix II.

SCP001 88/130

4.4 The relationship with the Biffaward Movement reportAs part of the Biffaward programme, RSWT undertook to review and synthesise the whole programme of work (Biffaward 2006). The approach taken to the review was to assess how comprehensive each study was in terms of its inputs and outputs (inputs of water, energy, gases and other materials and the outputs of products and solid, gaseous and water based wastes). In addition to this, each report was assessed out of 10 for its treatment of the financial, technical, social aspects of its analysis, as well as the timeliness of its subject matter. The very real strength of the movement report is that it clearly sets out the data coverage of the reports and provides a useful synthesis of the overall programme. The report should therefore be read in conjunction with this review if a detailed understanding of the Biffaward studies is required. However, unlike the movement report, this report provides a quantitative assessment of the robustness of the data used by each study.

4.5 Assessment resultsOf the 48 studies which have been assessed:

7 studies can be used to inform policy directly.

29 studies can be used to inform policy, but with a note of caution about some aspect of the study.

12 studies were assessed not to be usable in relation to future policy without further work

The assessment outcomes are summarised in Figure 5.12 and set out in full for each study in the assessment notes as provided in appendix B.

SCP001 89/130

O Y Y O O O Y W O O W Y Y O Y Y O O W

B1 D1 D2 D3 D4 D5 D6 M0 M1 M2 M3 C1 C2 R1 R2 P1 P2 P3 P4

Primary production studiesB/1127 4Sight Rocks to Rubble 2 1999 1 Mass balance 2 1 1 1 1 1 2 1 14 RB/2034 PigB/1639 Poultry Industry

Secondary production studiesB/1601 Chemicals 2 1999 1 2 1 2 2 Mass balance 2 2 2 2 2 1 1 2 25 YB/2043 Electricity Generation 1 2000 1 1 2 1 1 Mass balance 2 2 1 2 2 2 2 1 2 24 YB/1406 Construction 2 1998 1 1 2 1 1 Mass balance 2 2 1 1 2 2 2 1 2 2 25 YB/1412 Paper and Board Industry 2 2000 1 1 2 1 Mass balance 2 2 2 2 2 2 1 2 23 YB/1505 Ceramics 2 1999 1 1 2 Sample site MB 1 2 2 2 2 2 1 1 20 YB/1489 Glass Manufacture in UK 2 1 1 1 Mass balance 2 2 1 2 2 1 2 1 18 RB/1584 Foundry 1 1 1 1 Sample sites MB 1 2 2 2 1 1 1 1 15 YB/1705 Iron, Steel and Aluminium (Economic) 2 2001 1 1 2 2 1 Value chain analysis 2 2 2 2 2 2 1 1 1 26 YB/1698 Iron, Steel and Aluminium (MFA) 2 2001 1 1 2 2 1 Time series MF 2 2 2 2 2 2 1 1 1 26 Y

Tertiary service sector studiesB/1635 National Health Service - Material Health 2 2001 2 2 2 1 MB & Foot print 2 2 1 2 1 1 2 2 2 2 28 GB/1524 Public Sector 2 2001 1 1 1 Sample case studies 2 2 1 2 1 2 17 RB/2068 Education 1 2003 1 1 1 Sample audits 1 2 2 1 1 1 1 1 16 RB/1224 Schools Various 1 1 1 Case studies 1 2 2 2 2 2 1 1 16 RB/1536 Financial Sector 1 2000 1 1 2 Case study sites 2 1 2 1 1 13 RB/1634 Exhibition Industry 2001 1 2 2 Sample interviews 1 2 2 1 1 2 1 2 19 R

Product studiesB/1663 Automotive 2 2000 1 1 2 2 1 Mass balance 2 2 1 2 2 2 2 2 1 26 YB/1776 Magazine Publishing 2 2001 1 1 1 2 1 Samples & MB 2 2 1 1 2 1 2 2 1 24 YB/1352 Tyres 2 1 1 2 2 Mass balance 2 2 2 2 2 2 2 1 2 26 GB/1640 Housing 1 2001? 1 1 2 2 BRE's Ecopoints 1 2 1 1 1 2 2 2 1 2 24 YB/2222 Clothing and TextilesB/1837 Newspapers UK 2 2001/2 1 1 2 2 Mass balance 2 2 1 1 2 2 2 1 2 1 26 YB/2182 Food and Drink Processing 1 1999 1 2 2 1 1 Mass balance 2 2 1 2 1 1 2 1 2 23 Y

Consumption studiesB/2230 Ecological Budget UK 2 2001 2 2 2 1 REAP 2 2 1 2 2 2 2 2 2 2 2 32 YB/1831 Scotland 2 2001 1 2 2 2 MB & Footprint 2 2 1 2 2 2 2 2 2 2 2 32 YB/1996 South West - Stepping Forward 2 2001 1 2 2 2 MB & Footprint 2 2 1 2 2 2 2 2 1 2 2 31 YB/1946 Northern Ireland - Northern Limits 2 2001 1 2 2 2 MB & Footprint 2 2 1 1 2 2 2 2 1 2 2 30 YB/1699 South East - Taking Stock 1 2000 1 1 1 1 MB & Footprint 2 2 1 2 1 2 2 2 2 2 2 26 YB/2011 Wales 2 2001 1 1 1 2 MB & Footprint 2 2 1 1 1 2 2 2 1 2 2 27 YB/1646 London - City Limits 2 2000 1 1 2 2 MB & Footprint 1 2 1 1 1 2 2 2 2 2 2 27 YB/1303 Isle of Wight - Island State 2 1998/9 1 1 2 1 Footprint 1 2 1 1 1 2 2 2 1 1 2 24 YB/1195 Tourism

B/1783 Waste Data Flow 2 2 2 2 2 2 Database 2 2 2 1 2 2 1 2 2 2 32 GB/1425 Producer responsibility (REMAT) 2 2002-2004 1 2 1 2 2 Consultation 2 2 1 1 1 2 2 2 2 2 2 31 GB/1716 Transport Options for Scottish Waste 2 1998 1 2 1 1 2 Cost & LCA models 2 1 1 1 1 2 1 2 2 1 2 27 YB/1850 EuroCharge 1 2003 1 2 1 2 Case studies 1 2 1 1 2 2 2 2 2 23 YB/1413 Civic Amenity Sites 2 2001 1 1 2 1 Survey of site users 2 2 2 1 1 2 2 1 22 YB/31 Waste from electrical and electronic equipment 2 1998 1 1 1 2 Modelled arisings 1 2 2 1 2 2 1 2 1 23 YB/1597 Agricultural Waste (Mass balance) 1 1999 1 2 2 1 2 Mass balance 1 2 1 1 1 1 2 2 2 23 GB/1265 Agricultural Waste (Strategy) 1 1998 1 1 1 1 Mass balance 1 2 1 1 1 1 1 2 2 2 20 YB/1271 Packaging 2 2 Mass balance 2 2 1 1 2 2 1 2 1 20 YB/1826 Thermal Methods of Waste Treatment 1 1994 1 1 2 1 1 2 2 1 1 2 16 YB/1815 Wood combustion in furniture manufacturing 1 2003 1 1 Sample site audits 1 2 2 2 1 1 13 RB/1410 Furniture Packaging 1 Not provided 1 2 1 Sample audits 1 2 1 2 1 2 1 2 18 RB/1424 Solvent Waste in Furniture Manufacturing 1 1 Sample site audits 1 1 2 1 2 1 1 1 13 RB/1448 Glass from Fluorescent Tubes and Lamps 1 1 2 2 SFA of mercury in UK 1 2 1 1 12 RB/1411 Timber Waste

Methane 2 2002 2 2 2 2 2 2 2 2 1 2 2 2 1 2 2 32 GB/1355 Carbon UK 1 1999/2000 2 2 1 2 2 Material accounting 2 2 2 1 2 2 2 2 2 28 GB/1445 Nitrogen UK 2 1998 1 2 2 1 1 Mass balance 2 2 2 1 2 2 2 2 1 25 YB/2173 Wood 2 2002 1 1 2 1 1 Mass balance 2 2 2 2 1 1 1 1 22 Y

Eval

uatio

n

Rel

iabl

e

Ava

ilabl

e

Doc

umen

ted

Rea

l/mod

elle

d

Method(s) used Rob

ust (

cred

ible

&

relia

ble)

Obj

ectiv

ity

Pee

r rev

iew

ed

Data Policy cycle

Bou

ndar

ies

Yea

r

Upd

atab

le

Targ

ets/

obje

ctiv

es

SC

P p

olic

y ob

ject

ives

Acc

essi

bilit

y

Method

Iden

tific

atio

n

End

orse

men

t

Com

patib

le w

ith S

IC

Asse

ssm

ent

Results Policy

Non

-dire

ct fl

ows

Com

para

bilit

y w

ith

othe

r stu

dies

/ da

ta

Wor

k re

quire

d

Not assessed - No quantification of arisings

Material studies

Con

sulta

tion

resp

onse

Report not available

Report not availableReport not available

Report not availableWaste studies

Ass

essm

ent

Gen

eral

pol

icy

agen

das

Notes:- Criteria codes: O = Orange (Essential criteria), Y = Yellow (Inessential criteria), W = White (Neutral Criteria)- Score codes: 1 = score for Satisfactory assessments, 2 = score for good assessments- Colour codes: G = Green (Directly useable study), Y = Yellow (Study useable with caution), R = Red (Study not useable)

Figure 5.12 - Assessment Matrix: the Biffaward studies

The studies are grouped in Tables 5.1, 5.2 and 5.3. The limitations which led to the yellow studies not achieving a green assessment is provided alongside the studies’ title in Table 5.18. In all cases, the score relates to a potential maximum score of 36. It was decided not to rank the studies purely based on this score. If ranked purely by score, some yellow assessments would have been ranked ahead of

SCP001 90/130

the green assessments, even though a limitation has been identified. The alternative approach was to rank the yellow studies after the green studies. This would have suggested that all of the green studies are some how more policy relevant and robust than the yellow assessed studies. This is not the case as some of the yellow studies will be greatly useful in the formulation of policy as long their limitations are understood. Therefore no further categorisation or ranking can be provided beyond the following groupings.

Code Title Score- Methane 32B/1783 Waste Data Flow 32B/1425 Producer responsibility (REMAT) 31B/1635 National Health Service - Material Health 28B/1355 Carbon UK 28B/1352 Tyres 26B/1597 Agricultural Waste (Mass balance) 23

Table 5.17 - Studies which van be used directly to inform policy

Code Title Score Limitation of studyB/1831 Scotland 32 Modelled dataB/2230 Ecological Budget UK 32 Modelled dataB/1996 South West - Stepping Forward 31 Modelled dataB/1946 Northern Ireland - Northern Limits 30 Modelled dataB/1646 London - City Limits 27 Modelled dataB/1716 Transport Options for Scottish Waste 27 Age of DataB/2011 Wales 27 Modelled dataB/1699 South East - Taking Stock 26 Modelled dataB/1837 Newspapers UK 26 Modelled dataB/1663 Automotive 26 Non-direct flowsB/1705 Iron, Steel and Aluminium (Economic) 26 Non-direct flowsB/1698 Iron, Steel and Aluminium (MFA) 26 Non-direct flowsB/1601 Chemicals 25 Non-direct flowsB/1445 Nitrogen UK 25 Age of dataB/1406 Construction 25 Age of DataB/1640 Housing 24 Not updateableB/2043 Electricity Generation 24 Non-direct flowsB/1776 Magazine Publishing 24 Incomplete coverageB/1303 Isle of Wight - Island State 24 Modelled dataB/1850 EuroCharge 23 Not updateableB/1412 Paper and Board Industry 23 Not updateableB/31 Waste from electrical and electronic equipment 23 Age and level of modelling of dataB/2182 Food and Drink Processing 23 Non-direct flowsB/1413 Civic Amenity Sites 22 Not updateableB/2173 Wood 22 Incomplete coverageB/1505 Ceramics 20 Not updateable& modelledB/1265 Agricultural Waste (Strategy) 20 Age of data and not updateableB/1826 Thermal Methods of Waste Treatment 16 Data & compatibilityB/1584 Foundry 15 Age of data, not updateable, modelled and only direct flows

Table 5.18 - Studies which can be used for policy with an appreciation of the study's limitations

SCP001 91/130

Code Title ScoreB/1634 Exhibition Industry 19B/1489 Glass Manufacture in UK 18B/1524 Public Sector 17B/2068 Education 16B/1224 Schools 16B/1410 Furniture Packaging 18B/1127 4Sight Rocks to Rubble 14B/1424 Solvent Waste in Furniture Manufacturing 13B/1815 Wood combustion in furniture manufacturing 13B/1536 Financial Sector 13B/1448 Glass from Fluorescent Tubes and Lamps 12B/1411 Timber Waste -

Table 5.19 - Studies which should not be used to inform policy without further work

An alternative way of viewing these results is how they relate to the chain of production and consumption as introduced in Figure 5.11 and now set out in Table 5.20.

Table 5.20 - The robustness of the evidence base provided by the Biffaward studies

4.6 Discussion of Results4.6.1 Sectoral Studies

Secondary production studies:Methodologically, almost all of these studies set out to provide a mass balance of their sector capturing the main material inputs, product outputs and wastes. In the absence of official data, many of the relevant sectors needed to look internally for data. This led in many cases to the data being based on a scaling up of a sample of sites. In some cases there was a relatively small sample size (perhaps about 20% or less of the sector by value of output) combined with a lack of appreciation of the possibility of an un-representative sample. All but one of these studies are sufficiently robust to be used to inform future policy. However, none of these studies achieved a ‘green’ assessment and

SCP001 92/130

Products

Product studiesMining Poultry Pigs Auto Scotland South west South east Data flow Furniture wood

Magazines NI Isle of Wight REMAT Furniture packagingChemicals Construction Paper Ceramics Houses Wales London Furniture solvent

Glass Foundry Iron & steel Power Newspapers Packaging TimberClothes Glass tube lampsTyres

Health Public Education Tourism Civic sites EuroCharge WEEESchools Finance Exhibition

Methane

Nitrogen

CarbonWood

Material studies

Regional studies Waste studies

Scottish waste transportAgricultural waste (MB)

Agricultural waste strategy)Thermal methods

Primary production studies

Secondary production studies

Waste

Tertiary (service) sector studies

Production Consumption

Eco-Budget UK

therefore an appreciation of these studies’ limitations is required throughout. The main limitations relate to the data in some way. Specifically these limitations include not being updateable, using old data, and the use of modelled data (from the scaling up of small samples of the sector’s sites). In addition, all but the construction study failed to account for the sectors’ indirect material consumption.

In contrast to the other sector studies, the studies relating to the Iron, Steel and Aluminium sectors took time series and value chain analysis approaches which permit an understanding of the stocks throughout the UK economy, as well as possible efficiencies and savings available to the sector.

Tertiary (service) sector studiesThese sectors do not produce material outputs and often have diverse consumption patterns of their material inputs, in many cases similar in composition to the household sector. These studies have therefore faced greater challenges when collecting data and have often needed to take quite different approaches. In many cases, appropriate data has been unavailable at a sector level and the studies have therefore needed to base their data on samples of sites. These studies have been assessed for whether they capture indirect consumption (i.e. life-cycle flows related to their material inputs), and the assessment outcomes have highlighted when they have failed to do this. The only study to attempt this is the one relating to the NHS.

Of the six studies which have been assessed, only the study relating to the NHS can be used to inform policy, in this case without an appreciation of any limitations. The other five studies have all been assessed not to be sufficiently robust to inform policy, due in all cases to the lack of reliability of the data used.

4.6.2 Product StudiesMethodologically, some of these studies attempted to track the flow of a particular product or group of products through their value chain from production to disposal and possible recycling. The area of focus varied but in the case of tyres this focus was on disposal (as is the policy interest). The study looking at magazines considered the whole life cycle and highlighted the potential for resource efficiencies. The food and drink study largely focused on the production process and the direct material inputs and the study looking at housing (BedZed) used BRE’s Ecopoints methodology to consider the impacts of using different materials in their attempt to build sustainable housing.

Overall the product studies performed well and all of the six studies assessed are sufficiently robust to be used to inform future policy, with the tyres study achieving a ‘green’ assessment. However, in respect of the food & drink and housing studies it should be understood that the former does not fully consider indirect flows and the latter was assessed not to be updateable.

4.6.3 Consumption StudiesAll of the seven regional and the Ecological Budget UK studies use an ecological footprint methodology to estimate the material consumption which the residents of the region are responsible for, as well as a mass balance of materials flowing within the region. The use of the ecological footprint and the sources of data used to generate it have been improving throughout the time these studies have been undertaken. This evolution reflects the dedication of the relatively small group of practitioners who have worked on these studies (i.e. Best Foot Forward and SEI). This has led to a group of policy-relevant studies which have gained greatly from the experience of these researchers.

SCP001 93/130

The findings are largely consistent with one another and provide a powerful communication tool for engaging the public.

There have been a number of reviews criticising the eco-footprint methodology (see for example Defra 2005). Much of this criticism relates to the expression of all impacts in terms of land requirements (assessment of this criticism was outside this project’s terms of reference). In addition, the Defra’s review (2005) of UK’s eco-footprint studies which were completed by November 2004 identified the bottom-up component approach taken in sub-national studies as leading to variations in data sources, affecting the comparability of the results (Defra 2005, p. ii). However, the more recent studies undertaken by Best Foot Forward use the StepwiseTM approach and the studies undertaken by SEI use an Input Output (I-O) methodology, both of which take the compound UK footprint as their starting point, which is based on data collected to international standards, thereby providing a more consistent basis for the calculations (Defra 2005, p. iii). The technical quality of the more recent studies was considered to be very good which is reflected in the high scores.

The outcomes of the assessment show that all of these studies were assessed as sufficiently robust to be used to inform future policy with an explicit appreciation of their limitations, in this case the same limitation for all of the studies. All of the regional studies were assessed to be based on modelled datasets. For example, in the case of the StepwiseTM approach, regional product consumption data was based on economic data combined with data from the national level ProdCom database. This was the subject of much discussion at the project workshop and through the consultation with authors. It has been successfully argued that the modelling was expertly and transparently undertaken using satisfactorily robust data in all cases, which is reflected in the assessment of other criteria. The ecological footprint results of these studies have had a considerable impact with some policy makers. However, to achieve an ecological footprint at a sub-UK level presently requires the use of proxy data, not all of which directly relate to the measurement of mass, and this needs to be appreciated by users. It should also be noted that this limitation does not in itself make these studies (or any other yellow assessed study) less relevant to policy than the green studies, many of which were awarded a lower score.

4.6.4 Waste StudiesThe waste studies are a large and diverse group of studies which cover subject matter ranging from material flow sectoral level studies (e.g. wood combustion in furniture manufacturing) to waste strategy research projects and data models which assist in meeting the challenges of waste legislation (e.g. REMAT and Waste Data Flow). The methods adopted also vary, but in terms of numbers of studies the concepts of mass balance predominate.

The studies also vary in terms of their robustness and relevance to policy. Of the fifteen studies which have been assessed, ten have been assessed as adequately robust to inform policy, seven of which only with an appreciation the particular study’s limitations as shown in Table 5.18. The limitations of these studies were varied but mostly relate to the age and lack of updateability of data sets used.

4.6.5 Material StudiesAll four of these studies attempt to track the flows of particular substances (methane, carbon, nitrogen and wood) through the UK economy. In addition and with the exception of the wood study, they also consider stocks and flows of these materials in the environment. This required a clear distinction between the environment and the economy to be drawn.

SCP001 94/130

Identification

Assessment

Targets/

Objectives

Implementation

Evaluation

Formulation

Re-assessment Affirmation or change

Relevant to Biffaward studies

Not relevant to Biffaward studies

All four of these studies were assessed to be both relevant and robust enough to be used in the formulation of policy, although the nitrogen study makes use of pre-1999 (1998) data and the wood study is not sufficiently comprehensive in the sectors it covered.

4.7 Discussion and recommendations4.7.1 The policy cycle and the Biffaward studies gives a diagrammatic representation of the policy cycle. The assessment results shown in Figure 5.12 identify the part(s) of the cycle, in relation to which each study seems to have most potential use, although only in those cases where the studies have been assessed to be useable in policy. To fully identify whether in fact this is the case requires the detailed knowledge of those engaged with a particular policy making process.

Figure 5.13 - The policy cycle and the Biffaward studies

4.7.2 Specific policy points arising from particular studiesIt seems worth pinpointing, as part of this study, some specific points which arise from some of the Biffaward studies that are of relevant to future policy agendas. This is not a comprehensive listing of such points, and the policy use of each study in turn is discussed in the assessment notes as provided in Appendix II. Some of what follows discusses large policy challenges, whereas other points merely intend to highlight the best of the detailed findings from the studies. They are mentioned here because there seems to be a future policy need in the area under consideration, and a satisfactorily robust study which is relevant to meeting that need.

Sustainable production

Agriculture: There are number of policy challenges relating to agriculture. Farmers will require support and advice on how to comply with recent European legislation on the management of agricultural wastes (for example the disposal of plastics). The agricultural waste studies are still

SCP001 95/130

relevant to this challenge. The nitrogen study is useful in implementing the ongoing requirement to manage nitrates. However, a study looking at water in the UK would have complemented this by considering the trade off between the reduction of nitrates use in agriculture, and the cleanup of nitrates in potable water supplies. The use of material flow analysis is very suited to the consideration of such trade-offs.

Sustainable consumption

The funding of the regional Biffaward studies (and the Ecological budget study) have been central to the development of the concept of the ecological footprint. While the assessment of the ecological footprint concept and methodology were themselves outside the scope of this project (having been undertaken in Defra 2005, and both the concept and methodology remain controversial in relation to their robustness and policy usefulness), these studies have all been assessed to be technically robust, and have been found useful, and been endorsed by, some policy makers. They have provided a conceptualisation of consumption for some policy makers which goes beyond that conveyed by bulk material analysis alone. Many of the regional bodies and the decision bodies in Scotland and Wales seem likely to continue to ‘run’ with the process. Also, via SCPnet network coordinated by the Environment Agency, the results of all of these studies will continue to inform the implementation of SCP at a regional level. The Ecological Budget report will be central to this as it provides a comprehensive and systematic analysis of consumption by region. Improvements to the Prodcom database to show a breakdown by region (and possibly by sector) would improve the robustness of these types of studies.

The flows of materials in the service sectors have not been particularly well covered by the present programme of research. With the exception of the health study it would appear that the practitioners were let down by a lack suitable data in these sectors. It might also be argued that this lack of data is caused by a relative lack of attention to material resource management in what are often public sectors. A discussion might also be had as to whether this should be taken forward through government policy (e.g. regulation, public procurement) or left to the voluntary initiative of the sectors concerned. Whatever model is pursued, the quality of the available data needs to be greatly improved, and the quantity increased, and it would seem desirable to resolve this issue before any further service sector studies are commissioned.

The issues raised by the study looking at sustainable housing (using the Bedzed example) continue to have relevance to meeting the ongoing challenge of the development of sustainable communities. This study is however a little too case study based to meet such a wide reaching policy challenge. Such a challenge would seem to justify a further study exploring the how this might be applied in different settings (urban, rural, large, self build) and the overall material implications of projected national house building in the coming years.

Sustainable waste management

The Biffaward studies have managed to inform the sustainable management of wastes with the identification of numerous examples of good practice and opportunities for industrial symbioses, a few of which have been highlighted below.

1. The Packaging study identified that that the present available data is not fit for the proper implementation of the packaging Directive.

SCP001 96/130

2. The Magazine study highlighted the low rate of magazine recycling due to the misconception that the high levels of ink in magazines means that they cannot be recycled through domestic recycling routes.

3. The Paper and board study highlighted the implications of the increasingly high levels of paper recycling leading to the corresponding increase in waste produced by the sector due to lower grade inputs.

4. The Eurocharge study remains very relevant to the management of municipal waste and the policy dilemma as to whether allow variable charging by local authorities for the collection and disposal of domestic waste.

5. The study which considered how to manage the trade use of civic amenity sites identified that small traders tend to be faced with the full cost of the landfill tax, whether the waste goes to landfill or not. The report recommends that small traders might be permitted to use civic sites and pay only for that proportion of their waste which can’t be diverted (by the sites) away from landfill.

In addition to this the Waste Data Flow and REMAT studies will have ongoing use as intended.

4.7.3 The value of the Biffaward studiesIt should be noted that this review was primarily intended to identify those studies useful to future national policy objectives as defined by DEFRA and the other representatives on the project’s client group. It is likely that, had these studies had been commissioned by national government, the outcomes would be more readily useable for this purpose. However, this was not the purpose of the Biffaward programme of studies. Funded through the landfill tax credit scheme, which also required a minimum of 10% and third party co-funding, this programme brought the systematic study of resource flows to the attention of a broad group of stakeholders for the first time. In many cases the Biffaward studies complemented other work being undertaken in sectors in relation to sustainable resource use. In some less well off sectors, such work may not have been undertaken at all in the absence of Biffaward funding. Furthermore, the Biffaward programme enabled the exploration and development of different methodologies, which might not otherwise have occurred.

However, it is also clear that the Biffaward programme had its failings. After reading the reports and discussing the outcomes with those involved with the programme it is clear that some projects failed to fully understand the conceptual brief and therefore to appreciate the requirement for robust updatable data sets. In addition to this, other projects chose inappropriate boundaries for their studies, which failed to capture all issues relevant to external stakeholders.

Any assessment of value from this programme of work has to weigh the diverse benefits arising from it (not all of which have been captured by this assessment), against a hypothetical counterfactual scenario based on what would have been achieved had the money been spent in a different way. It was not part of this assessment to construct such a scenario. However, in the event that some kind of more coordinated programme of work had been commissioned by government, it is probable that this would have been more systematic and have generated information that was more useful to government policy agendas. However, it would arguably have provided benefits to a narrower set of stakeholders, resulted in less methodological experiment and development, and reached a smaller and less diverse audience.

SCP001 97/130

4.7.4 Recommended additional funding of work Linstead et al (2001 p.6) argued that to achieve the policy challenges emanating from the European Union and UK Government encouraging increased resource efficiency, “sophisticated and comprehensive data on waste streams and resource use” is required. It could also be argued that if the data sets can be integrated into a single database, their usefulness will be greater still. Therefore the following two questions need to be posed to assess how far the present programme of research realised these long-term aspirations:

1. Has a comprehensive set of data on waste and resource use been achieved for the UK economy covering all (significant) materials, and broken down according to the different materials, different major products, different economic sectors and different geographic regions been achieved? and

2. Have the studies categorised and classified their data in the same way so that the data could be placed in a single database, related to other sources and provide the basis for the realisation of comprehensive database over time?

Figure 5.12 shows that the outcome from the present programme of work does not represent a comprehensive coverage of robust data. Furthermore, and despite the efforts of Biffaward to introduce consistency into the studies after the programme had been started, it is not thought possible to place the results into a single database. This has reportedly been attempted by the York Stockholm Environment Institute (SEI) and found to be problematic in terms of data formats. However, it can be concluded that the Biffaward studies have shown what is possible, some of the pitfalls of commissioning such work, and has been central in developing the expertise available for future work in this area. It is therefore necessary to explore how the work and results from this programme might best be exploited, whether there would be value in additional work being commissioned in this area, and what this work might be.

Section 5.7.2 identified a number of areas for possible future work related to current policy needs. However one more general conclusion to emerge, based on a systematic reading of all of these studies, is that the present knowledge base is sufficient to support the substantial development of networks to promote industrial symbiosis. Such networks need to be shown to lead to mutual industrial advantages and benefits, on the basis of current knowledge, before further investment is made in extending the quantification and specification of industrial material flows as envisaged in Linstead et al (2001). Therefore, it is proposed a systematic assessment should be made (perhaps through a collaboration between the Waste and Resources Action Programme, WRAP, and the National Industrial Symbiosis Programme, NISP) of the waste flows identified in the Biffaward programme and other studies, in order to identify the full potential of industrial symbiosis, and other means of recovering wastes and putting them to productive use. This should shed some light on the desirability and possible nature of any further programme of work seeking the quantification of industrial material flows.

Conclusions from David Aeron-Thomas David Aeron-Thomas of Forum for the Future was a member of the project team, who is also closely involved with Biffaward and contributed to the production of the Mass Balance Movement report. As part of his contribution to the project, he sent the following conclusions about the Biffaward studies,

SCP001 98/130

which are both consistent with and complementary to those of the main authors of this study, and are considered worth reproducing in full.

The primary purpose of the research is to identify those studies that are both sufficiently robust and relevant to future policy agendas. 36 studies have been found which can inform future policy – 7 of which can be used directly, 29 need a note of caution about some aspect - primarily through the use of modelled data or age of data.

Many of the studies have informed past policy and are already informing future policy.  For example, Footprint Indicators have been taken up in several regions and by the Audit Commission and results fed directly into sustainable development strategies and the results of the Agricultural Waste Mass Balance fed directly into national SCP waste management planning.

In addition, the studies have been a diverse family which have brought many benefits including:

Models have been built and Methodologies and Frameworks developed, for example the REAP – Resources and Energy Analysis Programme – used to develop an evidence base and the Forum Mass Balance Framework used to develop a basic data structure

Networks have been formed, for example SCPnet – the Sustainable Consumption and Production network – developing the expertise of the UK Regions in understanding how resource flows fit into economic modelling

People have been trained up – the 60 or so studies have brought together people and experts from all over the country from National to Regional from Business sectors to Individual Businesses

Sectors have been engaged – for example the Steel and Glass Industries both identified increased recycling rates as a priority going forward.

To some extent these benefits have arisen simply because “what gets measured gets managed” and these studies made organisations look at underlying resource flows which have been largely ignored in the past.  However, a large part of the value has been in building up a critical mass of expertise and understanding in the area of resource flows.

This work is continuing through for example Footprinting, REAP and SCPnet.  But in order for government to obtain maximum value this diverse family of work and people need to be supported and channelled in a direction that will support specific policy aims.

In order to identify the most appropriate way forward three things are needed:

1. The relative merits of a comprehensive database of material flows (vs separate studies for significant sectors) needs to be evaluated

2. The pros and cons of government intervention need to be evaluated

3. The claim that understanding resource flows are essential to management of the UK economy needs to be examined seriously and a UK government position adopted.

This should best be taken forward by bringing together the relevant parties:

Users and potential users of such information – including policy makers (national, regional and local), the Treasury and industry

Data providers – including government statisticians, industry and industry bodies

Modellers – including UK and international (academics and practitioners)

SCP001 99/130

Those in the waste industry who have an understanding of current waste streams. 

This is probably best convened by government.

4.8 References

Biffa 1997 Great Britain plc - The environmental balance sheet - An analysis of Britain’s waste production and disposal account, with implications for industry and government, Biffa, October 1997, http://www.massbalance.org/resource/

Biffaward 2006 ‘The mass balance movement: The definitive reference for resource flows within the UK environmental economy’ Launched at the House of Commons on 28 February 2006, available at http://www.massbalance.org/

Defra 2005, Sustainable Consumption and Production – Development of an Evidence Base: Study of Ecological Footprinting, Final Report prepared for Defra Ref: CTHS0401, RPA June 2005 - http://www.defra.gov.uk/environment/business/scp/methodology.htm

Eurostat 2001, Economy-wide material flow accounts and derived indicators: A methodological guide, European Communities 2001, ISBN 92-894-0459-0 - http://mrw.wallonie.be/dgrne/eew/files/tbe2004/doc/methodological-guide.pdf#search='ISBN%209289404590'

Linstead., C, Ekins., P 2001 Mass Balance - Mass Balance UK: Mapping UK Resource and Material Flows, A guide to data co-ordination for the series of Biffaward Mass Balance projects prepared by Forum for the Future on behalf of the Royal Society for Nature Conservation

Linstead., C, Gervais., C, Ekins., P 2003 Mass Balance - An Essential Tool for Understanding Resource Flows: A Report on the Biffaward Programme of Mass Balance Projects, Royal Society for Nature Conservation, Available from the Forum for the Future and http://www.massbalance.org/resource/

SCP001 100/130

5 Development of an Indicator for Emissions and Impacts associated with the Consumption of Imported Goods and Services By Thomas Wiedmann, Jan Minx and John Barrett

5.1 Introduction5.1.1 Principles of emissions accounting and ‘embedded emissions’With the increase in international trade, it is becoming increasingly important accurately to determine environmental impacts resulting from imports. For open economies facing national CO2 targets, imports and exports of commodities and services influence the total accounted CO 2 emissions. Therefore, international negotiations of reducing CO2 emissions have also to address the question of how to accurately account annual CO2 emissions. Two basic accounting principles are discussed in the literature, a production versus a consumption principle. The distinction between the two principles is whether the producer or the consumer is responsible for the CO2 emitted.

Accounting for ‘territorial’ or ‘producer responsibility’ includes all resource flows triggered by the domestic production of goods and services, which serve to satisfy the domestic demand as well as the demand in the rest of the world through exports. ‘Consumer responsibility’ on the other hand assigns to the residents (consumers) of a country all resource flows required domestically as well as in the rest of the world for the production of its imports. Both views are shown schematically in Figure 6.14 below. Juxtaposing them will be crucial to addressing issues of responsibility for overseas impacts.

Figure 6.14 - Consumer (“consumption emissions”) versus producer (“production emissions”) responsibility for CO2 emissions in the UK

Emissions from production include all CO2 emissions from goods and services produced in the UK, wherever they are consumed (either within the UK or exported elsewhere). This measure does not take into account import related emissions. The “emissions from consumption” indicator on the other hand includes the emissions from goods and services consumed within the UK, wherever they come from (either from within the UK or imported from elsewhere). While including import related

SCP001 101/130

emissions in the estimation procedure, this indicator excludes export related emissions. Each approach has a different point of view and a different application. There are, of course, advantages and disadvantages to both approaches. The production indicator can help to pinpoint the drivers behind changes in CO2 emissions rooted in the way the UK provides it goods and services to final consumers across the whole world, while the consumption indicator can help to identify drivers behind changes in the worldwide impact of CO2 emissions from UK’s consumption patterns. Both indicators are required to enable informed decisions, more efficacious and adequate policies, and more specific abatement strategies.

Theoretically, one way to reduce emissions from production (territorial emissions) would be to close down all UK manufacturing, relocate it to other countries, and then import the goods and services that the UK consumes. However, though this would eliminate many emissions from local production, it would do nothing to curb total global emissions (due to continuing consumption). To a much lesser degree, this pattern has occurred within the UK as its economy has shifted away from heavy industry and toward services – a sector which also depends on indirect, “hidden” resource flows far away from its final consumed product.

The current absence of an indicator that accurately quantifies these “CO2 emissions from consumption” or “embedded emissions”39 is a shortcoming of present UK monitoring approaches, and this is something that DEFRA have recognised. In this project, DEFRA have commissioned the Stockholm Environment Institute together with the Policy Studies Institute to identify the most appropriate approach to constructing an indicator for emissions embedded in imports to the UK. A CO2 indicator that accounts for impacts of trade and overall consumption is an essential indicator that ought to be included in the UK’s headline indicators for Sustainable Development. Adopting such a consumption-based perspective – in addition to the ‘traditional’ approach of territorial emissions accounting – also offers the possibility of extending the range of policy and research applications tremendously, as described at the end of this report.

At present, there is no standardised method for calculating embedded CO2 emissions of imports. Many previous studies have unrealistically assumed that imports are produced using domestic production technology. For countries with diverging technology and energy mixes the likely errors are significant. Considering that many goods and services come from regions of the world where resource efficiency is lower and therefore emission intensities are higher than in the UK, it is reasonable to believe that ignoring different production technologies results in an under-estimation of the overseas impact of consumption activities in the UK. On the other hand, in some regions, e.g. where hydroelectricity is used for the production of goods, the emission intensity might be indeed lower than in the UK. At present no reliable procedure is in place to measure these effects. Therefore, some authors (e.g. Harris, 2000; Lenzen et al., 2004) have proposed models that use (more) specific information about production processes and efficiencies in other countries.

5.1.2 About this part of the project reportDEFRA, DTI and ONS are concerned with devising a robust indicator for embedded emissions (e.g. proposed indicator no. 12 in the SCP indicator basket40, see also Francis, 2004). For this reason

39 Emissions or impacts that are indirectly linked to processes or activities; also called “embodied”, “associated” or “virtual”.

40 http://www.defra.gov.uk/environment/statistics/scp/download/scp_rpt200506.pdf

SCP001 102/130

DEFRA commissioned a study on ‘embedded emissions’ through the programme ‘Sustainable Consumption and Production – Development of an Evidence Base’41.

This part of the project report presents the findings on the embedded emissions indicator study. The results of the literature review and assessment of existing approaches are presented and discussed. We have included major international studies as well as UK studies in this review. In the last few years many approaches have been discussed, developed and applied worldwide. This work provides a unique opportunity to transfer and apply this knowledge to the UK. The assessment, as specified in the Appendix III.1, was aimed at identifying the strengths, weaknesses, assumptions, limitations, data requirements and suitability of the reviewed approaches.

A specification for the development of an indicator for embedded emissions is given (technical details in the Appendix III.4) and options for the actual implementation, including cost reduction options, are presented. We also describe numerous policy and other applications of the specified approach.

5.1.3 Requirements for an indicator of embedded emissions and impactsAn ideal and robust indicator for environmental impacts embedded in traded goods and services must satisfy the following conditions42:

Cover the whole trade of the UK economy.

Disaggregate final consumption categories so that consumption of UK households, non-UK residents (overseas tourists and visitors to the UK) and UK residents on holidays abroad (tourists from the UK) can be assessed.

Produce time series on an annual basis from 1990.

Include environmental impacts other than CO2 emissions, e.g. embedded emissions of other pollutants, virtual water use or waste.

Indicate the amount of energy (and environmental impacts in general) used in transportation.

In addition it is desirable that the indicator would also:

Feature a reasonableError: Reference source not found number of trading partners, i.e. countries or world regions.

Disaggregate commodity sectors into a reasonable43 number of goods and services.

A modelling framework as described in the next section is able to satisfy all of these requirements.

41 Project code CTX0505/SCP1.1, Part C ‘Embedded Emissions’ (2005)42 As derived from the specification of this project (DEFRA, 2005. Sustainable Consumption and

Production – Development of an Evidence Base. Competition Details and Project Specification; Department for Environment, Food and Rural Affairs, London, August 2005).

43 The definition of ‘reasonable’ depends on the research and/or policy question. The maximum resolution achieved by one current multi-region model is more than 200 sectors, another one features 87 countries/world regions.

SCP001 103/130

5.2 Assessment Results and DiscussionAll together, 17 studies have been assessed in the course of this project provides an overview of all assessment results. The detailed review text as well as the individual rating of the studies can be found in the Appendix III.2 to this report.

Non-IO modelsHarris, 2000

(environmental accounts)Shui and Harriss, 2006 (EIO-LCA database)

Muradian et al., 2002 (IPPS database)

Wiedmann et al., 2005 (data from Ahmad and

Wyckoff, 2003)

21 12 15 22

Single-region IO modelsWyckoff and Roop, 1994

(six OECD countries, manufactured products)

Lenzen, 1998(Australian analysis)

Wenzel, 1999 (German products)

Munksgaard et al. 2000(household consumption)

Machado et al., 2001 (Brazilian economy)

Sanchez-Choliz and Duarte, 2004

(Spanish economy)

25 24 12 26 22 25

Multi-region IO modelsAhmad, 2003; Ahmad

and Wyckoff, 2003(OECD analysis)

Lenzen et al., 2004(CO2 multipliers in MRIO

models)

Peters and Hertwich, 2004

(Norwegian imports)

Nijdam et al., 2005(Dutch household

consumption)

Guan and Hubacek, 2006

(virtual water use in China)

33 31 24 14 (15)

Simulation modelsLutz et al., 2005

(GINFORS model, MOSUS project)

Chung, 2005(GTAP application)

29 25

Table 6.21 - Overview of approaches and assessment results

When applying the criteria defined in Section 3, some general conclusions can be drawn from the assessment. On average, input-output based approaches score better than methods that don’t employ this technique. Similarly, multi-region based approaches achieve better verdicts than single region models. These general conclusions can be further specified:

The main disadvantage of non-IO based approaches is that they do not automatically take into account indirect emissions arising from upstream production processes. When taking into account only direct (on-site) emissions from the sectors that produce exported commodities, the estimated embedded emissions are systematically and falsely too low. Various attempts have been undertaken to overcome this problem by either taking into account emissions caused by the use of electricity in manufacturing processes (Harris, 2000) or by using multiplication factors from different sources that try to capture upstream impacts (Wiedmann et al, 2005; Shui and Harriss, 2006). The latter ones were derived with the use of IO analysis, i.e. the studies reviewed are indirectly based on IO calculations. However, some systematic errors remain in all these cases, as either incomplete data or data from different countries or data from different years had to be used.

Input-output-based approaches on the other hand automatically cover all indirect impacts of upstream production (e.g. embedded emission, virtual water, etc.). In terms of environmental data only direct (on site) impacts for industry sectors are required as input. IO models don’t have to rely on multiplication factors from other studies; they generate them specifically, according to the scope and set-up of the calculation framework. This is a key strength of IO based approaches.

SCP001 104/130

However, it is important to allow for the inclusion of foreign technology coefficients if a distinction between resource efficiencies and emission intensities of production processes in trading countries is to be made. If, as frequently done, a closed economy single-region model is employed to calculate embedded emissions of imports, the resulting multipliers only represent the production structure of the domestic economy. This is a far reaching limitation which does not permit analysis and assessment of foreign production efficiencies. There would be no difference, for example, in the embedded emissions of iron and steel produced in the UK or imported from China.

The main advantage of assuming identical production recipes and efficiencies for all trading regions lies in the relative low data requirements. This fact also allows the construction of time series of embedded emissions which, for example, has been demonstrated by Proops et al. (1993) who present time series for CO2 embedded in UK imports from 1968 to 1984. Thus, changes over time can be picked up but it has to be clearly said that these changes are solely due to changes in the monetary value of imports as well as UK domestic production recipes and not to any changes in production overseas.

A methodological sound response to this challenge is to extend the basic single-region IO framework to the international case and to employ a multi-region input-output (MRIO) model, ideally covering all trading partners of the country under investigation. A few studies compare the results of single versus multi-region input-output modelling of energy and CO2 (Proops et al, 1999; Lenzen et al, 2004; Haukland, 2004) and demonstrate that multipliers and embodiments can differ substantially, thus warranting the extension to many regions. No studies could be found that did a similar comparison over a period of time and thus it is not possible to judge how severely time series would be affected when changing from a single to a multi-region model. An unproven assumption of the project team is that the change in the absolute level of embedded emissions might be considerable whereas the relative trend – e.g. increase or decrease over time – is probably not affected significantly.

Yet, the evidence presented in some studies allows the conclusion that an evaluation of foreign production efficiencies should be an intrinsic part of modelling factor embodiments. In policy analyses and scenarios for example, one might want to explore the environmental implications of trade with different countries or the consequences that the relocation of a particular industry to foreign countries has on emissions. Only models that distinguish between production technologies are able to provide a reliable quantitative assessment of such scenarios.

Furthermore, the review in this study has shown that MRIO models score best in the assessment and provide a framework that is able to satisfy the requirements outlined in the introduction.

In terms of data requirements the case is single versus multi-region approaches rather than non-IO versus IO based approaches. Input-output tables are available for all developed countries and some developing countries and can be estimated or approximated for minor trading regions or where national tables are not available. Although the sector aggregation varies from country to country, the principal economic accounting framework is a standardised process (UN handbook) and some data sources provide IO tables in a consistent format for a number of countries (OECD, GTAP). The benefits that IO-based models provide outweigh the additional data requirements. Also, approaches not based on IO data have other specific data requirements (e.g. indirect emission factors) that can be difficult or impossible to meet.

Detailed trade data are required for any model that deals with impacts embedded in traded commodities. Naturally, the more countries and regions are featured in the model, the higher are the data requirements. As mentioned above, only direct emissions data per economic sector are required if

SCP001 105/130

IO analysis is employed. This type of data is available for many countries (e.g. Environmental Accounts of OECD countries) and for a number of years.

5.3 Specification of an indicator for embedded emissions5.3.1 Main issuesResults from the studies reviewed demonstrate that it is important explicitly to consider the production efficiency and emissions intensity of a number of trading countries and world regions in an international trade model, which is globally closed and sectorally deeply disaggregated. Only then reliable and meaningful figures for embedded emissions can be derived.

Empirical work of any kind, and particularly when it involves examining inter-sectoral and international interdependencies is fraught with information problems and short-cut methods must be employed if pragmatic results are to be achieved. Only by making explicit what should ideally be done, systematic and transparent decisions about what short-cut methods should be applied in practice can be made. From a detailed exposition of the reportedly ideal case – a multi-region input-output (MRIO) mode – data and processing requirements can be derived. In the absence of data for individual regions or countries decisions can be made explicitly over whether and how proxy data can be drawn from other economies with similar economic structures, technology etc., or whether using an appropriate large country to proxy for a trading block is appropriate.

The implementation and application of a full multi-regional input-output framework poses three basic challenges: data availability, data reconciliation and computability. These issues and possible practical solutions are discussed in detail in the Appendix III.4. In the following we focus on the important issue of data handling in a MRIO model.

5.3.2 Data handlingCompiling the required data, estimating missing data and balancing conflicting data in the right way is the most crucial part of a MRIO framework. Most resources should be devoted to this part of the work as a good handling of data ensures consistency, robustness and repeatability of the whole approach. The data system should allow to

include data in different classifications,

aggregate or disaggregate sectors, depending on the research question,

cope with suppressed data,

estimate missing data,

accommodate different years for the analysis of time series.

A generic framework that satisfies these requirements is depicted in Figure 6.15 below. In principle this data framework employs optimisation techniques that balance data according to constraints which are defined by existing/available data. The essential elements are:

Concordance matrices that match data from different classifications.

Data templates that automatically translate between existing/available data and balanced data sheets. These templates are the central part of the system and required to assembly all the data

SCP001 106/130

used in the model. They are coded in a way that allows the entry of data in different formats or prices as well as blank or suppressed data points. They include the initial data estimates as well as a list of constraints required for optimisation.

A set of balanced data sheets containing all required information, generated through robust optimisation techniques.

Figure 6.15: Data handling protocol for a multi-region input-output model

Such a data framework – once in place – is extremely flexible and therefore easily expandable. The tangible outcome of this data protocol is a code that provides the following important advantages:

disparate sets of data can be used,

suppressed or missing data can be filled in,

sectors can flexibly be aggregated or disaggregated,

more data can be added any time, i.e.

o additional trading partners (regions, countries) can be added,

o additional environmental impact data can be added,

o additional years of data can be added,

these additions do not change the basic framework so that annual and time series data can be reproduced reliably, and

estimates of uncertainties can be performed as an intrinsic part of the model.

Specific data problems that can be addressed with such a data protocol are described in the Appendix III.4.2.

SCP001 107/130

Disparate (‘raw’) data

Data templates (for model assembly)

Balanced datasheets

Multi-region input-output

model

Indicator for embedded environmental impacts

Other indicators and analyses

Optimisation

Concordance matrices

5.3.3 Interpolating time series dataIn general, the availability for data required to produce time series does not vary significantly over time. Input-output databases are widely used to generate the national economic accounts and are therefore constructed by many statistical offices every year44. It is not a necessary condition to have analytical input-output tables for a MRIO model. Supply and use matrices can be used instead as described in Lenzen et al. (2004) and Wiedmann et al. (2006, see also the description in the Appendix III.3). This is a big advantage as supply and use tables are often available annually while analytical IO tables are not (in the UK, analytical tables are published in a five year cycle). It also allows the use of more up-to-date information as the time delay for publishing supply and use tables is shorter than for analytical tables (in the region of two to three years compared to more than five years for analytical tables). This is important because changes in the structure of domestic and foreign economies can be picked up more accurately if up-to-date input-output information is used. From one year to another these changes might not seem significant, but over a time period of several years the activity of some industries can change significantly, especially with respect to imports and exports.

Trade data is usually available annually and the same can be said about environmental data in developed and many other countries – at least energy use and CO2 emissions data are available annually from the International Energy Agency (IEA). The main issue with environmental data is the sectoral aggregation in international databases. Additional information from national statistics can be used to disaggregate emissions data and to reconcile data from different sources.

5.3.4 Outline of a pragmatic start for a model for embedded indicatorsThe review of existing approaches has shown that there is no sensible alternative to a multi-region input-output model if the requirements laid out in the introduction are to be met. However, setting up a complete MRIO model is a non-trivial, labour-intensive task. In order to minimize resource requirements, it is reasonable to explore possibilities to simplify the model in a way that still provides the advantages listed above and that holds the option of extending it to a full-scale model later on.

The objective of any future research should therefore be to implement an initial, relatively small, data and model framework that is easily expandable without major adaptations. As described above, the essential part is a protocol that operates data templates efficiently. This helps reducing cost and time requirements while at the same time allowing a consistent update of the model. It would be wrong to invest in ‘one-off’ models that might quickly produce some numbers with questionable reliability and that are not flexible enough to be expanded or updated at any time. A robust indicator must be built on a reliable basis for data handling and modelling.

In principle, reductions in data and therefore resource requirements can be made by cutting down the number of countries/regions, years, sectors or indicators. In the following, these four cut-down options for an efficient initial system are being discussed.

Reducing the number of countries/regions (trading partners). The model could start with a small number of countries and world regions. The smallest conceivable ‘multi’-region model would include two regions – the UK and the rest of the world (ROW). In order to avoid the assumption of UK-typical production efficiencies in foreign countries, the economic structure of the ROW could roughly be approximated by one of the main trading partners of the UK (e.g. EU). This would keep data requirements and dimensions of data matrices low and would free up resources for the

44 This does not necessarily mean that all industrial sectors are surveyed each year. Sub-groups of industries might be surveyed in turn in a two or more year cycle and the data for years where no survey was carried out are interpolated (balanced).

SCP001 108/130

implementation of time series or other analyses. At the same time a two-region set-up would still allow implementing data templates that can be expanded later on to include more regions.

Theoretically, the model boundaries could be further reduced to a single region, the UK only. Whilst this would cut down the data requirements to nationally available data sets and consequently further reduce resource requirements, this option is not recommended for three reasons. First, as outlined above, differences in foreign production recipes would not be picked up which is seen as a major limitation and compromises the reliability and the interpretation of the results. Second, the data handling procedure as described above could not be implemented in a way that would allow an easy expansion of the model. While some computation procedures are identical for the single and the multi-region case (e.g. producing annual IO tables for the UK for a number of years), others are not. Particularly, a routine that handles trade data and combines IO tables from different countries could not be implemented in the single-region case and it would be very difficult to add this later on. Single and multi-region frameworks are just too different in this respect. Third, such a single-region input-output model has been implemented and time series of CO2 emissions embedded in UK trade have been produced in the early 1990s by Proops et al. (1993). Whilst this model could be reused to extend the original time series45, no new insights into the underlying causes for change, e.g. main contributing countries and sectors, would be gained and recent methodological developments worldwide would not be acknowledged.

Reduce the number of years. The initial multi-region model could be set-up for just one year. This would reduce data requirements significantly whilst allowing the implementation of several trading countries or regions. However, no time series could be produced with this set-up, at least not initially. Time series data can be added later on and analyses for different years can be repeated with the initial model. The suggested data handling procedure allows the use of supply and use IO tables instead of analytical tables and the interpolation between years if necessary.

Start with CO2 as the sole initial indicator. This helps to keep data requirements for indicators to a minimum. Other environmental impact indicators can be added later at any time. Initial data on CO 2

emissions per industrial sector and country can be taken from the IEA database and replaced with more detailed environmental accounts data later on.

In general, the major steps required for setting up a global model for an embedded emissions indicator include:

Building data templates to accommodate input-output and trade data of different countries/regions, years, types, classifications, valuations, etc.

Integration of national accounts with environmental accounts, IEA CO2 and energy statistics and/or IPCC greenhouse gas emission accounts.

Implementation of routine (code) to optimise all data and to create balanced data sheets.

Cut-down option: reduce number of regions to two, i.e. UK and ROW

Cut-down option: reduce analysis to one year

Add-on option: addition of more impact data, such as other pollutants, water use, material consumption, LCA indicators, social indicators etc.

45 In fact, the work from Proops et al. (1993) was followed up in a PhD thesis supervised at Keele University, by Burkhard von Schlotheim, in the late 1990s. Schlotheim updated the data on the UK and Germany, and also extended the application to some other countries (Proops, 2006).

SCP001 109/130

Add-on option: implementing routines for analysis such as Production Layer Decomposition and Structural Path Analysis used for tracing environmental burdens along supply chains.

Add-on option: case studies, international workshops with policy and industry representatives about applications.

It is estimated that at least five experts are required to set up a full-scale MRIO model, including an ecological economist, an input-output specialist, a mathematician, an environmental researcher and a computer programmer. The main conceptual and actual tasks of the work required lie in the region of data handling, protocol development, coding, and optimisation. The estimated timeframe is three years. The cut-down options outlined above would reduce the costs considerably.

5.3.5 Feasibility of extending the model Generalised input-output models can be extended to include other (environmental) impact data and indicators. For example, all data relating directly to economic sectors gathered in national Environmental Accounts can be used in such a model. In principle, any factor of production – environmental, social or economic – can be analysed as long as it is unequivocally related to the industrial sectors in the input-output tables. The term “production factors” can be understood in a very general sense as additive indicators and quantities that are in any way associated with industrial production. They can be for example:

emissions of greenhouse gases and other air pollutants, general waste, release of toxic compounds into soil and water and effluent discharge into the ocean,

natural resources such as water, land, forest, minerals, metals and fuels, or

other physical production-related quantities such as waste flows or impacts of transportation,

economic parameters such as income, capital, or imports,

social factors such as employment, income disparity or occupational health and safety.

As long as a factor is additive in its impacts and clearly linked to industrial sectors, it can be treated with the input-output formalism and for impact studies. Some specific issues however remain difficult, e.g. the accurate allocation and modelling of waste flows. The export of waste, for example, would be reported in one sector in input-output and trade data, therefore not revealing the composition and origin of the waste.

The critical question of whether the basic model can be extended is about data availability and specific issues that might arise with certain factors. The following list focuses on some main issues; a more detailed discussion is provided in Appendix III.4.3.

Emissions of pollutantsExtending the MRIO model to include the emissions of pollutants is straight forward, solely depending on the data availability. As in the case of energy and CO 2, missing data can be approximated to a certain extent by using the emission coefficients of a national economy that can be seen as representative of a world region.

Water use

SCP001 110/130

Input-output analyses investigating ‘virtual water’ have been undertaken and water use data have been linked to economic data in National Accounts. Therefore, extending the model is feasible but questions of data availability and compatibility remain.

Land useSimilar to the above, land use data by economic sector can be used to extend national environmental account data and to make them suitable for MRIO modelling46.

Ecological FootprintsA methodology to extend the UK Environmental Accounts with Ecological Footprint data from the National Footprint Accounts has been described in Wiedmann et al. (2006). This approach can be applied to any country with similar environmental and Footprint data and therefore is suited for use in an extended multi-region input-output model.

Waste and recyclingSome countries have produced very detailed accounts of waste by type and treatment that can (and have been) used in input-output modelling. However, the classification of waste types (inert, organic, mineral, etc.) and the sector disaggregation might vary substantially and therefore it is likely that the inclusion of waste generation into a MRIO model would require a considerable amount of data preparation work. In an international context it would be interesting to follow the flow of wastes that are being transferred out of the domestic economy. A global multi-region input-output model could, for example, link the environmental impacts of the recycling industry in Asia to the amount of waste exported from the UK. The feasibility of such an analysis is a matter of data availability, compatibility and quality rather than the principle set-up of the model.

Materials and substances47

A challenging question is how well the flow of individual substances (e.g. Aluminium or Copper) or very specific materials (e.g. PVC plastic) can be modelled with a (multi-region) input-output model. The crucial point is sector disaggregation. The proposed MRIO model is kept flexible in terms of sector (dis-)aggregation. This allows dividing any sector into two or more sub-sectors and thus explicitly modelling the substance or material of interest48. However, additional financial as well as environmental data are needed to perform a detailed analysis of substance flows. The advantage of the MRIO model specified in this report is its flexibility towards extension while at the same time delivering a full upstream and interlinkage analysis.

Life Cycle Assessment IndicatorsMany Life Cycle Assessment (LCA) models already use input-output analysis in hybrid approaches to calculate comprehensive upstream impacts of the production of certain commodities. Embedded emissions data generated by a MRIO analysis can be multiplied with specific factors for LCA impact categories. The finer the sector disaggregation, the more specific is the attribution of impacts to sub-categories of commodity groups

International transportation46 See also the original project plan for the European MOSUS project ( www.mosus.net ) and Lutz et al.,

200547 See also the section on physical accounting (Material Flow Analysis etc.) earlier in this report.48 See the Appendix for an example.

SCP001 111/130

The environmental impact of domestic and international transportation is a complex issue that needs special attention. The economic and environmental factors of transportation (such as costs/margins, fuel use and emissions) are reported in several categories in national economic and environmental accounts. For the analysis of international transport however, complementary data is likely to be needed. These can be physical trade data (masses of imported/exported goods49), distances and modes of international transport by commodity and country/region and specific emission factors (one example is detailed transport data derived from the Energy Analysis Programme, IVEM, Groningen, NL). Further research is required to identify the best possible method that can be combined with a MRIO model.

TourismThe National Accounts of the UK provide a separate expenditure column for non-UK residents (overseas tourists and visitors to the UK) and UK residents on holidays abroad (tourists from the UK) (ONS, 2003). This allows for the allocation of indirect environmental impacts – whether they occur domestically or in other countries – to these final consumption categories in an input-output model. Wiedmann et al. (2006) describe an example for the attribution of Ecological Footprints to tourists and visitors in the UK in a single-region IO model.

In a multi-region IO model additional information needs to be added, depending on the research question that is to be answered. For example, if the question is what direct and indirect environmental impacts are caused by UK tourists in which countries/regions, then information on where UK tourists spend their holidays as well as their direct emissions in the destination countries need to be added to the model.

5.3.6 Advantages and strengths of the specified modelMulti-regional input-output analysis (MRIO) as the basic method for quantifying and tracing embedded emissions has a number of significant advantages:50

Input-output analysis has been used for several decades in economics to analyse economic transactions, and in ecology to analyse trophic food chains and webs as well as linked ecological-economic systems. IO analysis is therefore a well suited methodology to bridge the gap between economic and environmental approaches and it is generally accepted by both economists and physical scientists.

Input-output tables are produced at regular intervals by statistical agencies of most developed nations which guarantees a good data coverage, both spatially and temporally.

Input-output accounting is governed by UN standards, adopting internationally recognised systems of commodity/industry classification (e.g. the International Standard Industrial Classification, ISIC), which facilitates comparison over time, between nations and regions, and with standard economic indicators.

The IO system avoids methodological problems such as double counting and partial coverage of impacts.

49 The major data source for global physical trade data is: United Nations Conference on Tariffs and Trade (UNCTAD) and World Trade Organisation (WTO), Trade Analysis System on CD-ROM (PC-TAS). http://www.unctad.org/Templates/Page.asp?intItemID=1890

50 See also the general considerations about IO modelling earlier in this report.

SCP001 112/130

Multi-region input-output (MRIO) analysis has a long tradition, starting with Nobel Prize Laureate Wassily Leontief.

MRIO allows for integration of (monetary) trade flows with environmental databases and permits environmental impacts embedded in trade to be accurately and comprehensively evaluated.

Sophisticated IO-based analyses (e.g. structural path analysis, production layer decomposition, quantification of shared responsibility51) can be undertaken.

With a multi-directional trade MRIO model it would also be possible to include feedback loops and capture direct, indirect, and induced effects of trade (see e.g. Lenzen et al., 2004).

5.3.7 Assumptions, limitations and weaknesses specific to MRIO modellingThe general assumptions of input-output modelling and the limitations to which they lead have been described earlier in this report. Here we concentrate on very important aspects and those that are specific to multi-region input-output modelling.

General weaknesses Within one economic sector the input-output method assumes homogeneity (i.e. that each industry

produces a single product and all output uses the same processes and technology). This limits the analytical capacity of the system to the meso-level, i.e. only whole sectors can be examined, not single firms, products, materials or services. If the research question requires such a specific analysis, the sector under investigation needs to be disaggregated (e.g. railway transport might need to be separated into passenger and freight transport).

A static IO model does not allow for dynamic changes, and thus is limited in its scope to historical analysis rather than future projections. The assumptions that limit this scope are: the assumption of linearity (ignores economies of scale); not accounting for constraints on production factors (such as limited capital and labour); the assumption of constancy of commodity prices; and the assumption of fixed input structure in each industry.

Another result of the static nature of the model is that feedback mechanisms such as the rebound effect are not taken into consideration. This limits projections, but not the capacity of the model to do comparative static analyses.

An IO model as specified in this report is able to capture all upstream impacts (under the notion of consumers driving production), but not downstream impacts (where producers drive consumption). This would require a different model (as proposed by Ghosh, 1958). This can be important when investigating from a global sustainability perspective, for example, the emission impacts that cheap exports of coal induce in other countries. Thus, the analytical scope is restricted to demand driven emissions.

Origin of importsOne obvious issue with a MRIO model is the requirement of a considerable amount of data, which is why in most applications some sort of approximation has been used. In Appendix III.4.2 we describe

51 In addition to full producer and consumer responsibility accounts (Munksgaard et al. 2006a) any share of responsibility can be quantified with such a framework (Gallego and Lenzen, 2005).

SCP001 113/130

pragmatic ways to deal with data problems and mention that missing information on imports can be estimated under certain assumptions. This point is repeated here because of its significance.

It is in general not possible to find information on the spatial origin of every intermediate and final import, disaggregated according to the consuming sector in the country of destination. The use of trade coefficients, as suggested above, assumes that all industries and consumers receive imports from the same mix of countries. For example, if 20% of cars supplied in the UK come from Japan, then it is assumed that 20% of cars bought by each sector and by the consumers are Japanese cars. This means that the method could e.g. not take into account a company’s policy only to buy British cars. However, if information of this kind is available for a whole sector it could be incorporated separately into the model data.

Uni-directional trade52

A simplification and reduction in data requirement can be made if it is assumed that the domestic economy trades with all regions, but the other regions do not trade amongst each other. This is called ‘uni-directional trade’ instead of ‘multi-directional trade’. The data requirements are greatly reduced without introducing large errors. Lenzen et al. (2004) found these effects to be around 1-4% and these terms are often assumed to be negligible in other regional models (Round, 2001). Note, that excluding multi-directional trade rules out the analysis of feedback loops (see e.g. Lenzen et al., 2004).

ErrorsError: Reference source not foundErrors can enter into the calculations in many ways. The IO data and factor use intensities always have an error associated with them (e.g. Lenzen, 2001). Errors also arise in the adjustments for currency conversions, inflation, different sector classifications, aggregation, and so on. The magnitude of these errors is often difficult to estimate, but the errors still need to be considered. Ideally, some sort of error analysis should be performed or the potential magnitude of uncertainties discussed.

5.4 Policy and Other Applications5.4.1 Indicator for environmental impacts embedded in tradeMost importantly, the suggested model allows for a robust, reliable and reproducible quantification and analysis of environmental impacts embedded in the international trade of goods and services – not only carbon dioxide emissions, but ultimately all sorts of environmental and social impacts. This is novel. Studies so far either covered trade only partially or used short-cut estimates that do not really withstand scientific scrutiny and that cannot be truly relied upon in international negotiations about (embedded) emissions.

But why should national Governments report environmental (and other) impacts embedded in trade at all? Including embedded impacts into a government statistic or a national sustainability report can mean that the figures reported are higher than they would be if only territorial emissions/impacts were reported. So why do this? There are a number of good reasons:

To improve reporting on sustainability indicators: A methodologically robust indicator for embedded emissions adds a new aspect to the reporting of consumption-related environmental impacts. It provides new insights in the debate about sustainable consumption and production and enhances the credibility of governmental reporting on sustainability issues by delivering an

52 This section is adopted from Peters and Hertwich (2006).

SCP001 114/130

‘honest’ account of the ‘true’ scale of environmental impacts. It avoids loopholes in reporting that occur, for instance, when national territorial emissions drop because of the relocation of industries abroad. The embedded emissions of imported goods from those industries might be higher or lower than they were before the relocation, depending on whether the production process in the foreign country is less or more efficient.

To enable meaningful international comparisons (nation benchmarking): As the model uses standard financial accounting statistics (IO tables) and harmonised environmental accounts (e.g. NAMEA or IEA statistics) to the most possible extent, results for different trading partners can be compared with each other. This allows a reliable benchmarking of the amount of embedded impacts of exports, imports and economic sectors.

To instigate country dialogues and assist in trade negotiations: Quantifying and reporting embedded impacts shines a new light on patterns of international trade and opens the possibility to address sustainability implications more reliably in international dialogues and negotiations between countries, such as Kyoto Protocol conferences or WTO negotiations. For example, the proposed MRIO model can be used to reliably quantify the supposed cuts in emissions under the Kyoto Protocol emission reduction mechanisms, such as the Clean Development Mechanism (CDM), Joint Implementation (JI) and International Emissions Trading. So far, emission reduction figures of these projects are based on less robust estimation procedures.

To monitor the development of impacts embedded in trade over time. Time series for the indicator are a powerful means to monitor and understand changes in amount, origin and nature of embedded environmental impacts. The model quantitatively reveals shifts of production from developed-world countries to “pollution havens” abroad. Only when looking at changes over time, can the effects of economic developments and political decisions be understood.

To forecast the future impact of current trends: Analysis of past trends permits to some extent the estimation of future developments. In principle, the MRIO model can be developed further into a dynamic model that allows – under certain assumptions – the econometric prediction of future trends.

To enable and encourage other countries to make use of the same model and to drive forward international harmonisation of environmental reporting. With MRIO being a large international model that requires cooperation between countries for its realisation, it makes sense to promote the use of such a tool in many countries. The basic set-up is the same for different countries (see specifications), and adaptations can be made to accommodate different national IO tables and trade data. A widespread international use of the MRIO model would advance the harmonisation and standardisation of the accounting of trade data and environmental data in general.

To raise awareness amongst households for sustainability issues of consumption: In the UK, most embedded emissions are triggered by the final demand of households. MRIO studies of households can be extended to include not only environmental but also socio-economic aspects of import behaviour in households. Policies on sustainable (household) consumption can only be successful if these issues are addressed upfront and consumers are informed about the true impacts of buying products and services. Raising awareness is an important piece in the jigsaw puzzle of sustainable consumption.

SCP001 115/130

5.4.2 Sustainable procurement and business options – greening international supply chains

Because of the disaggregation by economic sector and trading partner in the MRIO model, it is possible to track the paths of embedded impacts along international supply chains. For example, the CO2 emissions of electricity production in China used to manufacture textiles can be quantified as well as the consumption of water in the dying process in India of the same clothes before they are exported to the UK. Wherever supply-chains – domestic or international – are included in public or private policies, purchasing and trade decisions change fundamentally. A MRIO model looks at impacts holistically, and thus extends current global sustainability reporting schemes (for example the Global Reporting Initiative guidelines on Triple Bottom Line accounting) that don’t look beyond the national or corporate perspective.

Therefore, results of the model allow for the implementation of sustainable procurement policies.

To reduce global emissions by identifying and focusing on ‘low hanging fruits’ overseas: The model is able to identify the main contributors to embedded impacts by economic sector and country or world region. This is done by an analytic procedure called ‘Structural Path Analysis’ (SPA). The results of a SPA allow us specifically to address those industries that add most to the embedded impacts. Domestic and international policies on sustainable development can use this information, e.g. in technology transfer programmes.

To enable meaningful benchmarking of the international procurement process: Embedded environmental impacts of specific procurement options can be benchmarked against the entire economic sector and against similar processes in other countries. This helps to identify opportunities for re-structuring international purchasing for companies or industry sectors.

To reward and encourage the greening of international supply chains: If, for example, a national policy goal was to reduce the emissions from the domestic car industry, the MRIO model could be used to identify the major emission sources of this particular industry sector. It might indicate that most emissions associated with the manufacturing of cars in the UK are actually generated in Asia due to emission intensive production of metal and steel intermediate products that are required for assembly. Having this information, the choice of abatement options is not restricted to national measures, but it opens up the chance to reduce global CO2 emissions effectively by addressing the international supply chain of car manufacturing.

To put Corporate Sustainability Reporting in an international context: Supply-chain issues are important for companies and embedded environmental and social impacts are becoming increasingly topical for Corporate Sustainability Reporting. This is reflected in a quote by the World Business Council on Sustainable Development (2002, p.55): “Current reporting practices are often performed within the boundaries of the reporting organization. In the coming years, it is likely that companies will increasingly report across the value chain. This will represent a new challenge in terms of reporting on the upstream (supplier related) issues linked to human rights, environmental and societal impacts, and also of coping with the wider downstream (consumer related) impact of products and services”. Once reporting of environmental burdens across supply chains is established, producing and trading entities (companies, nations) will look beyond their own operations and start scrutinising upstream and downstream relationships, not only with regard to economic advantages, but across social and environmental/biodiversity outcomes as well. The identification of those input paths that carry the largest impacts can be used for organisational planning and priority setting for informed action towards financial, social and environmental

SCP001 116/130

sustainability. In particular, it shows organisations alternatives for effective procurement policy changes, which may be applied instead of perhaps costly on-site measures.

To inform investors about hidden financial risks: Ethical investment portfolios try to invest selectively in companies that support or set environmental and social standards. A MRIO analysis might reveal that companies from a certain sector regularly import goods from a country where the production of these goods is associated with high greenhouse gas emissions and/or low social standards. These hidden (embedded) impacts are seen as a financial risk, either because the reputation of the portfolio might be at risk or under anticipated regulations, and the investor can address specific import practices directly with the respective companies.

5.4.3 Other applicationsWith globalisation gaining pace, MRIO models will find many applications in other research and policy areas around international trade. Other potential areas of application are:

Include, report and assess other factors of production embedded in trade, such as social (employment, working hours, child labour, etc.), economic (salaries and wages, value added, profits, etc.) and other environmental issues (virtual water, air pollution, land use, resource depletion, hidden flows, impacts on biodiversity, toxic substances, etc.). Thus, national economic accounts can be complemented with social and environmental satellite accounts (comprehensive ‘socio-ecological budget’).

Address those issues highlighted by the ‘socio-ecological budget’ with specific policies on development and international trade relations. The MRIO model provides the information which country / world region and economic sector should be targeted.

Devise a truly interconnected global Life-Cycle Assessment (LCA) model by extending the MRIO model with established LCA indicators. Hybrid Life Cycle Assessment (see e.g. Suh et al., 2004) already uses IO data to increase system completeness and it is also possible to extend these models further using MRIO data.

Assess impacts of investment funds, in particular those with a strong global reach.

Extend the reporting scheme of all greenhouse gas emissions to the Intergovernmental Panel on Climate Change (IPCC) to reflect emissions embedded in international trade.

5.5 Conclusions and recommendationsIt is an old truism that there is no ‘best’ model as such, but only a ‘best’ model for a specific purpose. This work has reviewed, analysed and assessed four types of models as described in the literature: approaches not involving input-output (IO) calculations and single-region, multi-region and simulation models based on input-output analysis. Each of these models has its virtues and shortcomings. From a methodological point of view, the essence is that IO models are superior to non-IO models with respect to the completeness of the life cycle accounting and they can be very detailed in their description of commodities produced in and traded between economies.

Which type of model is most suitable for usage as an indicator for embedded emissions and impacts depends on the purpose (research and policy question) of the particular application. The specifications for this project lay out the main issues for political decision-making: completeness of trade coverage, consumption analysis of households and tourism, time series, impacts other than CO2, and

SCP001 117/130

transportation. In addition to these issues, the review clearly showed that it is most desirable to allow for much detail with respect to both the number of trading partners and the number of economic sectors. Results from the studies demonstrate that it is important to explicitly consider foreign production recipes and emissions in a model which is globally closed, includes multi-regional trade and is sectorally deeply disaggregated. Only then reliable figures for indicators of impacts embedded in trade can be derived.

Therefore, if the ultimate goal is to establish a methodology that allows for the reliable ex-post quantification of emissions and, more general, environmental impacts associated (embedded) with trade in goods and services as well as for the identification of major contributing sectors and countries, then, according to the outcome of the review, a multi-region model based on IO analysis (MRIO model) seems to be the most suitable and promising choice. MRIO models are particularly appropriate to describe analytically the environmental impacts of production, consumption, imports and exports with the possibility to track their origin via inter-industry linkages, international supply chains and multi-national trade flows.

A robust indicator must be built on a trustworthy basis for data handling and modelling. The specification of such a model as well as the practical implications of establishing a reliable indicator have been described in the report. The advantages using a MRIO model have been discussed as well as the specific assumption and limitations. A number of practical applications for political and corporate decision-makers have been described and naturally an enormous amount of research questions can be addressed with such a comprehensive model in purely academic applications.

Recommendations for the practical implementation of an indicator model The crucial part of an operational MRIO framework is a code protocol that processes data of any kind in a highly efficient way. In essence, this is a sophisticated computer programme that can ‘digest’ data from different countries and years in different classifications and valuations with data gaps and inconsistencies. Furthermore, the programme allows a flexible breakdown of economic sectors if this is required to answer specific questions. An efficient data handling protocol of this type helps reducing cost and time requirements while at the same time allowing a consistent update of the model.

Realising and operating a complete MRIO model is a major undertaking and requires both careful planning and joint efforts. As the resource requirements are substantial, it is sensible to include cut-down options, to split up working tasks, and/or to build up the model in stages. The main advantage of the proposed model is its flexibility towards the integration of more data of any kind as well as towards a change in the policy or research question the model is supposed to answer.

Reductions in data and therefore resource requirements can be made by cutting down the number of countries/regions, years, sectors or indicators. However, it would be a ‘dead-end’ undertaking to invest in ‘one-off’ models that are not flexible enough to be expanded or updated later on. Also, there will always be a trade-off between different cut-down options.

Options for moving forward are:

1. ‘Slim’ option: Resource requirements can be minimized to a certain level by implementing an initial, relatively small, data and model framework that is easily expandable without major adaptations. The minimum size of such a ‘slim’ model features two regions (e.g. UK and the rest of the world). As less trade data and input-output data of other countries are required, more resources could be devoted to the production of time series. A further cut-down option would be to start the model with CO2 as the sole indicator.

SCP001 118/130

2. Full-scale option with greenhouse gas indicators:This option includes modelling the UK’s major trading partners explicitly, i.e. the countries and regions where the UK mainly imports from are represented with their own input-output and bilateral trade tables. Adding these data is straight forward once the data handling protocol described above is in place, although some data enquiry and preparation work will be necessary. It would also be sensible to add energy use and greenhouse gases other than CO2 to the list of indicators in this extended version.

3. Extended indicators option, including LCA:Extending the suite of indicators allows for the modelling of other embedded impacts of trade such as other pollutants, virtual water, land use, waste, materials and substances, as well as Life Cycle Assessment indicators. The model is flexible towards the addition of any external factor of production, including economic and social indicators. This enlarges the list of potential policy and research questions that can be investigated tremendously. The extension can be performed gradually, i.e. indicators can be added as certain questions arise.

4. Option to include specific analyses of tourism and international transportation:Modelling the direct and indirect (embedded) environmental impacts of tourism and, in particular, international transportation requires additional research and further extensions of the MRIO model. This includes, for example, gathering and analysing data for different modes of transport, distances, values and volumes of international freight transport, specific emission factors etc.

Recommendations for general future research in the areaThe need for comprehensive and robust modelling approaches for embedded impacts in an international context using the MRIO framework has been recognised by research groups worldwide (see e.g. Dimaranan and McDougall, 2005; Giljum, 2005; Hertel, 1997; Lenzen et al., 2004; Peters and Hertwich, 2006) and it can be expected that concrete proposals will be brought forward in the near future. The review conducted in this project clearly shows that an environmental MRIO model requires substantial efforts which can only be mustered if both funding bodies and research institutions pool their resources and expertise. Given the international context of the subject and existing research efforts worldwide it seems reasonable to seek international collaboration.

Several international bodies already hold large amounts of the data required for a MRIO model (e.g. GTAP, OECD and Eurostat for IO and trade data). One of these agencies or a consortium could maintain a MRIO database that also holds environmental accounts data. Developing, harmonising and standardising these data bases would be of general benefit, not only for MRIO modelling but for countless other research and policy applications as well.

With respect to impacts other than CO2 emissions, further general research is required to consolidate existing data sets and modelling approaches. Numerous studies have proven that input-output modelling is appropriate to analyse these other indirect (embedded) impacts (water, waste, materials, etc.), but consistent data accounts are still the main obstacle for implementation. If environmental accounts were systematically extended with detailed accounts of water use, waste generation and disposal, recycling, land use etc., ideally in an internationally harmonised system, a boost in scientific and practical applications can be expected.

SCP001 119/130

5.6 References

Ahmad, N. 2002. The OECD Input-Output Database. DSTI/EAS/IND/SWP(2001)14[14th International Conference on IO Techniques - Montreal October 10-15, 2002.]. Organisation for Economic Co-operation and Development (OECD), Science, Technology and Industry Directorate.

Ahmad, Nadim and Wyckoff, Andrew W., 2003. Carbon dioxide emissions embedded in international trade of goods. DSTI/DOC(2003)15. Organisation for Economic Co-operation and Development (OECD), STI Working Paper 2003/15.

Ahmad, Nadim, 2003. A framework for estimating carbon dioxide emissions embedded in international trade of goods. STI/SD/WAF(2003)20. Organisation for Economic Co-operation and Development (OECD),

Allan, G., Hanley, Nick, McGregor, P. G., Swales, J. K., and Turner, K. R., 2004a. An Extension and Application of the Leontief Pollution Model for Waste Generation and Disposal in Scotland. 04-05. University of Strathclyde, Glasgow. Strathclyde Discussion Papers in Economics.

Allan, G., McGregor, P. G., Swales, J. K., Turner, K. R., and Hanley, Nick, 2004b. Modelling the relationship between economic activity and waste generation, treatment and disposal in the Scottish economy. scotecon, University of Stirling.

Allan, J. A., 1998. Virtual water: a strategic resource. Global solutions to regional deficits. Ground Water, 36:545-546 pp.

Almon, C., 1991. The INFORUM Approach to Interindustry Modeling. Economic Systems Research, 3:1-7 pp.

Atkinson, G. and Hamilton K. (2002). International trade and the 'ecological balance of payments'. Resources Policy 28, 27-37.

Australian Bureau of Statistics, 2000. Water Account for Australia 1993-94 to 1996-97. ABS Catalogue No. 4610.0, Australian Bureau of Statistics, Canberra, Australia.

Australian Bureau of Statistics, 2004. Water Account, Australia 2000-01. ABS Catalogue No. 4610.0, Australian Bureau of Statistics, Canberra, Australia.

Barrett, J., Birch, R., Cherrett, N., Wiedmann, T., 2005. Reducing Wales’ Ecological Footprint – Main Report. Stockholm Environment Institute, University of York; published by WWF Cymru, Cardiff, UK; March 2005. http://www.walesfootprint.org.

Bastianoni,S., Pulselli,F.M., and Tiezzi,E., 2004. The problem of assigning responsibility for greenhouse gas emissions. Ecological Economics, 49:253-257

Battjes, J.J., Noorman K.J. and Biesiot W. (1998). Assessing the energy intensities of imports. Energy Economics 20, 67-83

Benders, R. M. J., H. C. Wilting, K. J. Kramer, and H. C. Moll. 2001. Description and application of the EAP computer program for calculating life cycle energy use and greenhouse gas emissions of household consumption items. International Journal of Environment and Pollution 15(2): 171–182.

Bicknell, K.B., Ball R.J., Cullen R. and Bigsby H.R. (1998). New methodology for the ecological footprint with an application to the New Zealand economy. Ecological Economics 27(2), 149-160.

Birch, R., Barrett, J., Wiedmann, T., 2004. Exploring the consumption and related environmental impacts of socio-economic groups within the UK. InternationalWorkshop on Sustainable Consumption, School of the Environment, University of Leeds; 5/6 March. http://www.env.leeds.ac.uk/~hubacek/leeds04/call.htm.

Boomsma, P., J. van der Linden, et al. (1991). Construction of intercountry and consolidated EC input-output tables. 31th European RSA Congress, Lisboa, Portugal.

Bourque, P.J. (1981). Embedded energy trade balances among regions. International Regional Science Review 6, 121-136.

Briceno, Tania, Peters, Glen, Solli, Christian, and Hertwich, Edgar, 2005. Using life cycle approaches to evaluate sustainable consumption programs: Car-sharing. Hertwich, Edgar. Working Papers 2/2005.

SCP001 120/130

University of Science and Technology (NTNU), Trondheim, Norway. Industrial Ecology Programme (IndEcol).

Burniaux, Jean-Marc and Truong, Truong P., 2002. GTAP-E - An energy-environmental version of the GTAP model. GTAP Technical Papers, No. 16. Center for Global Trade Analysis, Purdue University, West Lafayette, Indiana, USA.

Burniaux, Jean-Marc, 2002. Incorporating carbon sequestration into CGE models: a prototype GTAP model with land uses. Center for Global Trade Analysis, Purdue University, West Lafayette, Indiana, USA.

Carbon Trust, 2006. The carbon emissions generated in all that we consume. Carbon Trust, London. http://www.thecarbontrust.co.uk/carbontrust/about/publications/CTC603_Supply_chainfinal.pdf

Carlsson-Kanyama, A., Engström,R., and Kok,R., 2005. Indirect and Direct Energy Requirements of City Households in Sweden: Options for Reduction, Lessons from Modeling. Journal of Industrial Ecology, 9:221-235 pp.

Carlsson-Kanyama, A., R. Karlsson, H. C. Moll, and R. Kok. 2002. Household metabolism in the five cities: Swedish national report - Stockholm. Swedish National Report to the ToolSust project. Fms report 177. Stockholm: Environmental Strategies Research Group (fms).

Cella, G. (1984). The input-output measurement of interindustry linkages. Oxford Bulletin of Economics and Statistics 46(1), 73-84.

Chenery, H. (1953). Regional analysis. In: Chenery, H. and Clark, P., Eds. The Structure and Growth of the Italian Economy. Mutual Security Agency, Rome, Italy, 96-115.

Chung, Hyun-Sik, 2005. Balance of CO2 Emissions Embedded in International Trade: Can Korean Carbon Tax on its Imported Fossil Fuels Make Any Difference in BEET? EcoMod2005 - International Conference on Policy Modeling [June 29 - July 1, 2005].

Cicas, Gyorgyi, Hendrickson, Chris, and Matthews, H. Scott, 2005. Regional Input-Output Analysis: A New Basis for Life-Cycle Assessment. Presentation on the ISIE 2005 conference, Stockholm

Clements, B.J. (1990). On the decomposition and normalization of interindustry linkages. Economics Letters 33, 337-340.

Common, M., Proops J.L.R. and Speck S. (1999). A comparative analysis of carbon dioxide emissions in Australia, Germany and the United Kingdom. In: Mahendrarajah, S., Jakeman, A.J. and McAleer, M., Eds. Modelling Change in Integrated Economic and Environmental Systems. John Wiley & Sons, Chichester, UK, 321-338.

Common, M.S. and Salma U. (1992). Accounting for changes in Australian carbon dioxide emissions. Energy Economics 14(3), 217-225.

de Haan, M., 1999. On the international harmonisation of environmental accounting: comparing the National Accounting Matrix including Environmental Accounts of Sweden, Germany, the UK, Japan and the Netherlands. Structural Change and Economic Dynamics, 10:151-160 pp.

de Haan, M., 2002. Disclosing international trade dependencies in environmental pressure indicators: the domestic consumption perspective. In: DeBresson, C., Ed. 14th International Conference on Input-Output Techniques, Montréal, Canada

Dietzenbacher, E., 2005. Waste treatment in physical input-output analysis. Ecological Economics, 55:11-23 pp.

Dietzenbacher, E., Giljum, S., Hubacek, K., and Suh, S., 2005. Physical Input-Output Analysis and Disposals to Nature. Unpublished work.

Dimaranan, Betina V. and McDougall, Robert A., 2005. Global Trade, Assistance, and Production: The GTAP 6 Data Base. Dimaranan, Betina V. and McDougall, Robert A. Center for Global Trade Analysis, Purdue University.

Duchin, F., 2005. A world trade model based on comparative advantage with m regions, n goods, and k factors. Economic Systems Research, 17:141-162 pp.

Eder, P. and Narodoslawsky M. (1999). What environmental pressures are a region's industries responsible for? A method of analysis with descriptive indices and input-output models. Ecological Economics 29, 359-374.

European Communities, 2002. Water Accounts. Results of Pilot Studies. European Communities, Belgium.

SCP001 121/130

Eurostat (2001). Methods for estimating air emissions from the production of goods imported into the UK. Working Paper 2/2001/B/5, DG Regional Policy and Eurostat, Luxembourg, Luxembourg.

Eurostat (2003) ‘International trade in services 1992–2001’, Compilation Guide, European Commission, Office for Official Publication of the European Communities, Luxembourg.

FAO, 2005. FAOSTAT. Food and Agriculture Organization, Statistical Databases, Rome. http://faostat.fao.org

Ferguson, L., McGregor, P. G., Swales, J. K., and Turner, K. R., 2004. The environmental 'trade balance' between Scotland and the rest of the UK. scotecon, University of Stirling.

Ferng, J.J., 2001. Using composition of land multiplier to estimate Ecological Footprints associated with production activity. Ecological Economics, 37:159-172.

Ferng, J.J., 2002. ‘Toward a scenario analysis framework for energy footprints’, Ecological Economics, vol. 40, pp. 53-69.

Ferng, J.J., 2003. Allocating the responsibility of CO2 over-emissions from the perspectives of benefit principle and ecological deficit. Ecological Economics, 46:121-141 pp.

Fieleke, N.S. (1975). The energy trade: the United States in deficit. New England Economic Review(May/June), 25-34.

Flegg, A.T., Webber, C.D. and Elliot, M.V. (1995) ‘On the appropriate use of location quotients in generating regional input-output tables’, Regional Studies, Vol.29, No.6, pp547-561.

Francis, P. (2004) ‘The impact of UK households on the environment’, Economic Trends 611:23-29 and Office of National Statistics (2002) ‘Methodologies for Estimating the Levels of Atmospheric Emissions arising from the Production of Goods Imported into the UK’, ONS Report, London

Frickmann Young, E.C. (2000). International trade and industrial emissions in Brazil: an input-output approach. In: XIII International Conference on Input-Output Techniques, Macerata, Italy, Internet site http://policy.rutgers.edu/cupr/iioa/iioa.htm

Furukawa, S., 1986. International Input-Output Analysis. Compilation and case studies of interaction between ASEAN, Korea, Japan and the United States, 1975; IDE Occasional Paper Series 21. Institute of Developing Economies, Tokyo.

Gale, L.R. (1995). Trade liberalization and pollution: an input-output study of carbon dioxide emissions in Mexico. Economic Systems Research 7(3), 309-320.

Gallego, B. and Lenzen, M., 2005. A consistent input-output formulation of consumer and producer responsibility. Economic Systems Research 17(4), 365-91.

Gay, P.W. and Proops J.L.R. (1993). Carbon-dioxide production by the UK economy: an input-output assessment. Applied Energy 44, 113-130.

GFN, 2005. National Footprint Accounts, national summary data sheets. Global Footprint Network. http://www.footprintnetwork.org/gfn_sub.php?content=national_footprints

Ghosh, A. (1958) Input–output approach in an allocation system, Economica, 25, pp. 58–64.

Giljum, S. and Hubacek K. (2001). International trade, material flows and land use: developing a physical trade balance for the European Union. Interim Report IR-01-059, International Institute for Applied Systems Analysis, Laxenburg, Austria.

Giljum, S. and Hubacek,K., 2004. Alternative Approaches of Physical Input-Output Analysis to Estimate Primary Material Inputs of Production and Consumption Activities. Economic Systems Research, 16:301-310 pp.

Giljum, S., Hinterberger,F., Lutz,C., and Meyer,B., 2006. Modelling global resource use: material flows, land use and input-output models. In: S.Suh (Editor), Handbook on Input-output Economics for Industrial Ecology. Springer, Dordrecht, forthcoming.

Giljum, Stefan, 2005. Quantifying indirect natural resource requirements with a global multi-country input-output model. Conference "Environmental Accounting and Related Sustainable Development Indicators". Jan Evangelista Purkyne University, Usti nad Labem and Charles University - Environment Center, Prague. Conference was held on 26th - 27th September 2005 at Charles University in Prague, Czech Republic.

SCP001 122/130

Gillen, W.J. and Guccione A. (1980). Interregional feedbacks in input-output models: some formal results. Journal of Regional Science 20(4), 477-482.

Gretton, P. and P. Cotterell (1979). The RAS method for compiling input-output tables - Australian Bureau of Statistics experience. Eighth Conference of Economists, La Trobe University.

Greytak, D. (1970). Regional impact of interregional trade in input-output analysis. Papers of the Regional Science Association 25, 203-217.

GTAP/EPA, 2005. Towards an Integrated Data Base for Assessing the Potential for Greenhouse Gas Mitigation. The Methane and Sequestration Branch of the US Environmental Protection Agency (USEPA),

Guan, Dabo and Hubacek, Klaus, 2006. Assessment of Regional Trade and Virtual Water Flows in China. Ecological Economics (in press). obus

Harrigan, F.J., McGilvray, J.W. and McNicoll, I.H. (1980a) ‘Simulating the structure of a regional economy’, Environment and Planning A, Vol.12, No.8, ppp927-936.

Harris, R., 2000. Methodologies for estimating the levels of atmospheric emissions arising from the production of goods imported into the UK. ONS. Eurostat contract No.: 97/09/57/013 Part 1. London, Office for National Statistics.

Haukland, E. 2004. Trade and environment: Emissions intensity of Norway's imports and exports. Master's thesis, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.

Hayami H, Kiji T 1997. An input-output analysis on Japan-China environmental problem: compilation of the input-output table for the analysis of energy and air pollutants. Journal of Applied Input-Output Analysis 4: 23-47.

Hayami H, Nakamura M, Asakura K, Yoshioka K 1999. The emission of global warming gases: trade between Canada and Japan. In: Canadian Economics Association Annual Meeting: University of Toronto, Canada.

Hayami, H. and Kiji T. (1997). An input-output analysis on Japan-China environmental problem: compilation of the input-output table for the analysis of energy and air pollutants. Journal of Applied Input-Output Analysis 4, 23-47.

Hayami, H., Nakamura M., Asakura K. and Yoshioka K. (1999). The emission of global warming gases: trade between Canada and Japan. In: Canadian Economics Association Annual Meeting, University of Toronto, Canada

Hayami, H., Nakamura M., Asakura K. and Yoshioka K. (1999). The emission of global warming gases: trade between Canada and Japan. In: Canadian Economics Association Annual Meeting, University of Toronto, Canada

Hendrickson,C.T., Horvath,A., Joshi,S., and Lave,L.B., 1998. Economic Input-Output Models for Environmental Life-Cycle Assessment. Environmental Science and Technology, 32:184A-191A pp.

Hertel, T. W. and McDougall, Robert A., 2003. GTAP Model Version 6.2. Center for Global Trade Analysis, Purdue University.

Hertel,T.W., 1997. Global Trade Analysis, Modelling and Applications. Cambridge University Press, Cambridge.

Hertwich, Edgar, Briceno, Tania, Hofstetter, Patrick, and Inaba, Atsushi. Sustainable Consumption: The Contribution of Research (Workshop Proceedings, 10-12 February 2005, Gabels Hus, Oslo). Hertwich, Edgar. Rapport nr: 1/2005. 2005. Høgskoleringen 5, E1 Tel.+47 73 59 89 40 http://www.indecol.ntnu.no

Hinterberger, Friedrich and Giljum, Stefan, 2003. Modelling global resource use: MFA, land use and input-output models. [Workshop "Quo vadis, MFA?", Wuppertal Institute, 9-10 October 2003]. ConAccount, 2003, Wuppertal.

Hubacek, K. and Giljum, S., 2003. Applying physical input-output analysis to estimate land appropriation (Ecological Footprints) of international trade activities. Ecological Economics, 44:137-151.

Hubacek, K. and Sun,L., 2001. A scenario analysis of China's land use and land cover change: incorporating biophysical information into input-output modeling. Structural Change and Economic Dynamics, 12:367-397 pp.

SCP001 123/130

Imura, H. and Moriguchi,Y., 1995. Economic interdependence and eco-balance: accounting for the flow of environmental loads associated with trade. In: S.Murai (Editor), Toward Global Planning of Sustainable Use of the Earth - Development of Global Eco-Engineering. Elsevier, Amsterdam, pp. 189-207.

Imura, H. and Tiwaree,R.S., 1994. Input-Output Assessment of Energy Consumption and Carbon Dioxide Emission in Asia. Environmental Systems Research, 22:376-382 pp.

Isard, W. (1951). Interregional and regional input-output analysis, a model of a space economy. Review of Economics and Statistics 33, 318-328.

Isserman, A.M. ‘Estimating export activity in a regional economy: a theoretical and empirical analysis of alternative methods’, International Regional Science Review, Vol.5, No.2, pp155-184.

Jacobsen, H.K. (2000). Energy demand, structural change and trade: a decomposition analysis of the Danish manufacturing industry. Economic Systems Research 12(3), 319-343.

Joshi, S., 1999. Product Environmental Life-Cycle Assessment Using Input-Output Techniques. Journal of Industrial Ecology, 3:95-100 pp.

Kainuma, M., Matsuoka Y. and Morita T. (2000). Estimation of embedded CO2 emissions by general equilibrium model. European Journal of Operational Research 122, 392-404.

Keuning,S.J., van Dalen,J., and de Haan,M., 1999. The Netherlands' NAMEA; presentation, usage and future extensions. Structural Change and Economic Dynamics, 10:15-37 pp.

Kondo, Y. and Moriguchi Y. (1998). CO2 emissions in Japan: influences of imports and exports. Applied Energy 59(2-3), 163-174.

Kondo, Y., Moriguchi Y. and Shimizu H. (1996). Creating an inventory of carbon dioxide emissions in Japan: comparison of two methods. Ambio 25(4), 304-308.

Kondo, Y., Moriguchi, Y. and Shimizu, H., 1998. CO2 Emissions in Japan. Influences of Imports and Exports, Applied Energy 59:163-174.

Kremers, H., Nijkamp, P.,Wang, S., 2002. A comparison of computable general equilibrium models for analyzing climate change policies. Journal of Environmental Systems 28 (3), 41– 65.

Lee, H. and Roland-Holst D. (1997). The environment and welfare implications of trade and tax policy. Journal of Development Economics 52, 65-82.

Lee, Huey-Lin, 2002. An Emissions Data Base for Integrated Assessment of Climate Change Policy Using GTAP; Center for Global Trade Analysis)

Lenzen, M. and Dey,C.J., 2002. Economic, energy and greenhouse emissions impacts of some consumer choice, technology and government outlay options. Energy Economics, 24:377-403 pp.

Lenzen, M. and Murray, S.A. (2001) ‘A Modified Ecological Footprint Method and its Application to Australia’, Ecological Economics, Vol.37, pp. 229-255.

Lenzen, M., 1998. Primary energy and greenhouse gases embedded in Australian final consumption: an input-output analysis. Energy Policy 26(6), 495-506

Lenzen, M., 2002. A guide for compiling inventories in hybrid life-cycle assessments: some Australian results. Journal of Cleaner Production, 10:545-572 pp.

Lenzen, M., 2003. Environmentally important paths, linkages and key sectors in the Australian economy. Structural Change and Economic Dynamics, 14:1-34 pp.

Lenzen, M., 2004. Nature, Preparation and Use of Water Accounts in Australia. Technical Report 04/2, Cooperative Research Centre for Catchment Hydrology, Faculty of Environmental Sciences, Griffith University, Nathan, Queensland, Australia; March 2004. http://www.catchment.crc.org.au/pdfs/technical200402.pdf

Lenzen, M., Pade, L.-L., and Munksgaard, J., 2004. CO2 Multipliers in Multi-region Input-Output Models. Economic Systems Research, 16:391-412 pp. http://dx.doi.org/10.1080/0953531042000304272

Lenzen, M., Wackernagel, M., Lauck, B., Wermer, P., 2005. The Ecological Footprint of the State of Victoria. University of Sydney, Global Footprint Network and Environment Protection Authority, Victoria. http://www.epa.vic.gov.au/eco-footprint.

SCP001 124/130

Leontief, W. (1970), ‘Environmental Repercussions and the Economic Structure: An Input-Output Approach’, Review of Economic Statistics, vol.52, pp.262-277.

Leontief, W. and Ford,D., 1970. Environmental repercussions and the economic structure: an input-output approach. Review of Economics and Statistics, 52:262-271 pp.

Lutz, Christian, Meyer, Bernd, and Wolter, Marc Ingo, 2005. GINFORS-Model. Gesellschaft für Wirtschaftliche Strukturforschung mbH (GWS), Osnabrück, Germany. MOSUS Workshop, IIASA Laxenburg; April 14.-15., 2005

Machado, G., Schaeffer R. and Worrell E. (2001). Energy and carbon embedded in the international trade of Brazil: an input-output approach. Ecological Economics, submitted.

Matthews, H.S. and Small, M.J., 2000. Extending the Boundaries of Life-Cycle Assessment through Environmental Economic Input-Output Models. Journal of Industrial Ecology, 4:7-10 pp.

McCann, P. and Dewhurst, J.H.Ll. (1998) ‘Regional size, industrial location and input-output expenditure coefficients’, Regional Studies, Vol.32, No.5, pp435-444.

McDonald, G.W., Forgie,V.E., and MacGregor,C., 2006. Treading lightly: Ecofootprints of New Zealand's ageing population. Ecological Economics, 56:424-439 pp. http://dx.doi.org/10.1016/j.ecolecon.2005.09.019

McGregor, P.G., McNicoll, I.H. Swales J.K. and Turner K.R. (2001), ‘Who Pollutes in Scotland? A Prelude to an Analysis of Sustainability Policies in a Devolved Context’, Fraser of Allander Institute, Quarterly Economic Commentary, vol.26, no.3, pp. 23-32.

McGregor, P.G., Swales, J.K. and Turner, K. (2004b) ‘The environmental 'trade balance' between Scotland and the rest of the UK: an inter-regional input-output and SAM analysis’, mimeo, Strathclyde Discussion Papers in Economics, 04-21. may update – journal of regional science?

McGregor, P.G., Swales, J.K. and Turner, K.R. (2004a), ‘An Input-Output Based Alternative to ‘Ecological Footprints’ for Tracking Pollution Generation in a Small Open Economy’, mimeo, Strathclyde Papers in Economics, 04/04. Already submitted to EE

Meyer, Bernd, Lutz, Christian, and Wolter, Marc Ingo, 2003a. Global Multisector/Multicountry 3 - E Modelling: From COMPASS to GINFORS. Berlin. IHDP Conference on Governance for Industrial Transformation. 5-12-2003a. Paper presented at the IHDP Conference on Governance for Industrial Transformation, Berlin, 5.-6. December, 2003

Meyer, Bernd, Lutz, Christian, and Wolter, Marc Ingo, 2003b. Modeling Sustainable Development for Europe in a Global Framework. Ottawa. OECD Environment and Modeling Workshop. Presentation at the OECD Environment and Modeling Workshop. Ottawa, 9.-10. October, 2003

Miller, R.E. (1969). Interregional feedbacks in input-output models: some experimental results. Western Economic Journal 7, 57-70.

Miller, R.E. (1985). Upper bounds on the sizes of interregional feedbacks in multiregional input-output models. Journal of Regional Science 26(2), 285-306.

Miller, R.E. and Blair,P.D., 1985. Input-Output Analysis: Foundations and Extensions. Prentice-Hall, Englewood Cliffs, New Jersey.

Miyazawa, K. (1966). Internal and external matrix multipliers in the input-output model. Hitotsubashi Journal of Economics 7(1), 38-55.

Moll, S., Bringezu,S., Femia,A., and Hinterberger,F., 2002. Ein Input-Output-Ansatz zur Analyse des stofflichen Ressourcenverbrauchs einer Nationalökonomie. Ein Beitrag zur Methodik der volkswirtschaftlichen Materialintensitätsanalyse. Arbeit- Umwelt-Wachstum. Nachhaltigkeitsaspekte des sektoralen Strukturwandels. Norderstedt.

Monfreda, C., Wackernagel, M. and Deumling, D., 2004. Establishing national natural capital accounts based on detailed ecological footprint and biological capacity accounts. Land Use Policy, 21:231–246.

Mongelli, I., Tassielli,G., and Notarnicola,B., 2005. Global warming agreements, international trade and energy/carbon embodiments: an input-output approach to the Italian case. Energy Policy, In Press, Corrected Proof.

SCP001 125/130

Moses, L.N. (1955). The stability of interregional trading patterns and input-output analysis. American Economic Review 45(5), 803-826.

Munksgaard, J., Minx,J.C., Christofferson,L.B., and Pade,L.-L., 2006a. Models for National CO2 Accounting. In: S.Suh (Editor), Handbook on Input-output Economics for Industrial Ecology. Springer, Dordrecht, forthcoming

Munksgaard, J., Pade,L.-L., Minx,J., and Lenzen,M., 2006b. Influence of trade on national CO2 emissions. International Journal of Global Energy Issues, in press

Munksgaard, J., Pedersen K.A. (2001). CO2 accounts for open economies: producer or consumer responsibility? Energy Policy 29, 327-334.

Munksgaard, J., Pedersen,K.A., and Wier,M., 2000. Impact of household consumption on CO2 emissions. Energy Economics, 22:423-440 pp.

Munksgaard, J., Wier,M., Lenzen,M., and Dey,C., 2005. Using Input-Output Analysis to Measure the Environmental Pressure of Consumption at Different Spatial Levels. Journal of Industrial Ecology, 9:169-186 pp.

Muradian, R., O'Connor,M., and Martinez-Alier,J., 2002. Embedded pollution in trade: estimating the 'environmental load displacement' of industrialised countries. Ecological Economics, 41:51-67

Murata, A., Kato K. and Kurokawa K. (1998). An estimation of environmental impacts of the trade of industrial products: a case of Japan. International Journal of Global Energy Issues 11(1-4), 139-145.

Nakamura, S. and Kondo,Y., 2002. Input-Output Analysis of Waste Management. Journal of Industrial Ecology, 6:39-63 pp.

Nakamura, S., 1999. An interindustry approach to analyzing economic and environmental effects of the recycling of waste. Ecological Economics, 28:133-145 pp.

Nielson, Anne Merete and Weidema, Bo P., 2001. Input-output analysis - Shortcuts to life cycle data? Nielson, Anne Merete and Weidema, Bo P. Danish Environmental Protection Agency, Copenhagen. Proceedings of a workshop held in Copenhagen on the 29th of September 2000. Copenhagen: Danish Environmental Protection Agency. (Environmental Project 581). Annexes only available in HTML: Annex A (workshop programme) and Annex B (participants list). For later developments see also http://www.leidenuniv.nl/cml/ssp/projects/envioa/

Nijdam,D.S., Wilting,H.C., Goedkoop,M.J., and Madsen,J., 2005. Environmental Load from Dutch Private Consumption: How Much Damage Takes Place Abroad? Journal of Industrial Ecology, 9:147-168 pp. e-Supplement available on http://mitpress.mit.edu/jie/consumption

Nijkamp, P., Wang,S., and Kremers,H., 2005. Modeling the impacts of international climate change policies in a CGE context: The use of the GTAP-E model. Economic Modelling, 22:955-974 pp.

Nyhus, D.E., 1991. The INFORUM International System. Economic Systems Research, 3:55-64 pp.

Oliveira-Martins, J., Burniaux J.-M. and Martin J.P. (1992). Trade and the effectiveness of unilateral CO2-abatement policies: evidence from GREEN. OECD Economic Studies 19, 123-140.

ONS, 2003. United Kingdom Input-Output Analyses, 2003 Edition. Mahajan, Sanjiv. Office for National Statistics, London. Download of Input-Output Supply and Use Tables at: http://www.statistics.gov.uk/inputoutput

ONS, 2005. Environmental Accounts, Autumn 2005. Office for National Statistics, London. Download of Environmental Accounts data tables at: http://www.statistics.gov.uk/statbase/datasets2.asp or http://www.statistics.gov.uk/statbase/explorer.asp

Pain, Nigel, Mourougane, Annabelle, Sédillot, Franck, and Le Fouler, Laurence. The New OECD International Trade Model, 2005. ECO/WKP(2005)27, Economics Department Working Papers No.440. 2005. Paris, Organisation for Economic Co-operation and Development (OECD). Economics Department Working Papers.

Paltsev, S.V. (2001). The Kyoto Protocol: regional and sectoral contributions to the carbon leakage. Energy Journal 22(4), 53-80.

Pedersen, Ole Gravgård, 2004. Waste Accounts for Denmark 1999. Statistics Denmark.

SCP001 126/130

Perroni, C. and Rutherford T.F. (1991). International trade in cabon emission rights and basic materials: general equilibrium calculations for 2020. Department of Economics, Wilfrid Laurier University, Waterloo, Ontario, Canada.

Peters, G. and Hertwich, E., 2004. Production Factors and Pollution Embedded in Trade: Theoretical Development. NTNU Working Papers 5/2004. Trondheim, Norway, University of Science and Technology (NTNU). Industrial Ecology Programme (IndEcol).

Peters, G. and Hertwich, E., 2005. Pollution Embedded In Norwegian Consumption. In: E.Hertwich, T.Briceno, P.Hofstetter, and A.Inaba (Editors), Sustainable Consumption: The Contribution of Research. Workshop Proceedings, 10-12 February 2005, Gabels Hus, Oslo. University of Science and Technology (NTNU), Trondheim, Norway, pp. 22-38.

Peters, G. and Hertwich, E., 2006a. Measuring Household Environmental Impacts: The Case of Norway. Journal of Industrial Ecology, forthcoming

Peters, G. and Hertwich, E., 2006b. The Application of Multi-Regional Input-Output Analysis to Industrial Ecology: Evaluating Trans-boundary Environmental Impacts. In: S.Suh (Editor), Handbook on Input-output Economics for Industrial Ecology. Springer, Dordrecht, forthcoming.

Peters, G., Briceno, T., and Hertwich, E., 2004. Pollution Embedded In Norwegian Consumption. NTNU Working Paper 6/2004. Trondheim, Norway, University of Science and Technology (NTNU). Industrial Ecology Programme (IndEcol).

Petri, P.A. (1976). A multilateral model of Japanese-American trade. In: Polenske, K.R. and Skolka, J.V., Eds. Advances in Input-Output Analysis. Ballinger Publishing Co, Cambridge, MA, USA, 481-496.

Pezzey, J. (1992). Analysis of unilateral CO2 control in the European Community and OECD. Energy Journal 13(3), 159-171.

Polenske, K.R. (1976). Multiregional interactions between energy and transportation. In: Polenske, K.R. and Skolka, J.V., Eds. Advances in Input-Output Analysis. Ballinger Publishing Co, Cambridge, MA, USA, 433-460.

Polenske, K.R. (1980). The U.S. Multiregional Input-Output Accounts and Model. D.C. Heath and Company Lexington, MA, USA.

Polenske, K.R. (1989). Historical and new international perspectives on input-output accounts. In: Miller, R.E., Polenske, K.R. and Rose, A.Z., Eds. Frontiers of Input-Output Analysis. Oxford University Press, New York, USA, 37-50.

Proops, J.L.R., Faber M. and Wagenhals G. (1993). Reducing CO2 Emissions. Springer-Verlag Berlin, Germany.

Proops, J.L.R., 2006. SPIRE, Keele University, UK; personal communication.

Proops, J.L.R., Atkinson,G., Schlotheim,B.F., and Simon,S., 1999. International trade and the sustainability footprint: a practical criterion for its assessment. Ecological Economics, 28:75-97

Reimer, J.J., 2006. Global production sharing and trade in the services of factors. Journal of International Economics, In Press, Corrected Proof.

Reinert, K.A. and Roland-Holst D.W. (2001). Industrial pollution linkages in North America: a linear analysis. Economic Systems Research 13(2), 197-208.

Richardson, H.W. (1985). Input-output and economic base multipliers: looking backwards and forwards. Journal of Regional Science 25(4), 607-661.

Rodrigues, J., Giljum, S. 2005. The accounting of indirect material requirements in material flow-based indicators. ICFAI Journal of Environmental Economics III (2), 51-69.

Round, J. I., 1978a. "An interregional input-output approach to the evaluation of non-survey methods." Journal of Regional Science 18: 179-194.

Round, J. I., 1978b. "On estimating trade flows in interregional input-output models." Regional Science and Urban Economics 8: 289-302.

Round, J. I., 1983. "Nonsurvey techniques: a critical review of the theory and the evidence." International Regional Science Review 8(3): 189-212.

SCP001 127/130

Round, J.I. (2001). Feedback effects in interregional input-output models: what have we learned? In: Lahr, M.L. and Dietzenbacher, E., Eds. Input-Output Analysis: Frontiers and Extensions. Macmillan, London, UK, ?

Round, J.I., 1985. Decomposing multipliers for economic systems involving regional and world trade. Economic Journal 95, 383-399.

Sanchez-Choliz,J. and Duarte,R., 2004. CO2 emissions embedded in international trade: evidence for Spain. Energy Policy, 32:1999-2005 pp.

Schaeffer, R. and Leal de Sá A. (1996). The embodiment of carbon associated with Brazilian imports and exports. Energy Conversion and Management 37(6-8), 955-960.

Shui, Bin and Harriss, Robert C., 2006. The role of CO2 embodiment in US-China trade. Energy Policy (in press, corrected proof).

Sonis, M. and Hewings G.J.D. (1998). Economic complexity as network complication: multiregional input-output structural path analysis. Annals of Regional Science 32(3), 407-436.

Sonis, M., Hewings G.J.D. and Gazel R. (1995). The structure of multi-regional trade flows: hierarchy, feedbacks and spatial linkages. Annals of Regional Science 29, 409-430.

Sonis, M., Oosterhaven J. and Hewings G.J.D. (1993). Spatial economic structure and structural changes in the EC: feedback loop input-output analysis. Economic Systems Research 5(2), 173-184.

Statistics Sweden (2000). Miljöpåverkan av Svensk handel - resultat frå en pilotstudie. Rapport 2000:5, Statistics Sweden, Stockholm, Sweden.

Statistisches Bundesamt, 2001. Endbericht zum Projekt "A Physical Input-Output-Table for Germany, 1995". 98/559/3040/B4/MM. Statistisches Bundesamt, Wiesbaden, Germany. Umweltökonomische Gesamtrechnungen, Material- und Energieflussrechnungen.

Strømman, A. H. and Duchin, F. 2005. A World Trade Model with Bilateral Trade Based on Comparative Advantage. Rensselaer Working Papers in Economics No. 0509, June 2005. Department of Economics, Rensselaer Polytechnic Institute, Troy, NY, USA.

Strømman, A. H., Hertwich, E. G., and Duchin, F., 2005. Shifting Trade Patterns as a Means to Reduce Global CO2 Emissions: Implications for the Aluminium Industry. Rensselaer Working Papers in Economics No. 0508, June 2005. Department of Economics, Rensselaer Polytechnic Institute, Troy, NY, USA.

Subak, S. (1995). Methane embedded in the international trade of commodities. Global Environmental Change 5(5), 433-446.

Suh, S. and Huppes G. (2001). Environmental IOA of the industrialised world - a project description. In: Weidema, B.P. and Nielsen, A.M., Eds. Input/Output analysis - Shortcuts to life cycle data? Environmental Project No. 581. Miljøstyrelsen, Ministry for Environment and Energy, Copenhagen, Denmark, 69-75.

Suh, S. and Huppes,G., 2002. Missing Inventory Estimation Tool Using Extended Input-Output Analysis. International Journal of Life Cycle Assessment, 7:134-140 pp.

Suh, S., 2004a. Functions, commodities and environmental impacts in an ecological-economic model. Ecological Economics, 48:451-467 pp.

Suh, S., 2004b. Material and energy flows in industry and ecosystems networks. Institute of Environmental Science (CML), Leiden University, NL,

Suh, S., Lenzen,M., Treloar,G.J., Hondo,H., Horvath,A., Huppes,G., Jolliet,O., Klann,U., Krewitt,W., Moriguchi,Y., Munksgaard,J., and Norris,G., 2004. System boundary selection in life-cycle inventories using hybrid approaches. Environmental Science & Technology, 38:657-664 pp.

Symons, E., Proops J. and Gay P. (1994). Carbon taxes, consumer demand and carbon dioxide emissions: a simulation analysis for the UK. Fiscal Studies 15(2), 19-43.

Tarancon, M. and P. Del Rio (2005). "Projection of input-output tables by means of mathematical programming based on the hypothesis of stable structural evolution." Economic Systems Research 17(1): 1-23.

Task Force on National Greenhouse Gas Inventories (1996). Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories - Reporting Instructions (Volume 1). Institute for Global Environmental Strategies, Intergovernmental Panel on Climate Change - IPCC National Greenhouse Gas Inventories Programme, Report, http://www.ipcc-nggip.iges.or.jp/public/gl/invs4.htm (Tokyo, Japan).

SCP001 128/130

Truong, T.P., 1999. GTAP-E. Incorporating Energy Substitution into GTAP Model, Technical Paper 16, Center for Global Trade Analysis, Purdue University, Purdue.

Turner, K. R., 2003. The additional precision provided by regional-specific data: the identification of fuel-use and pollution generation coefficients in the Jersey economy. 03-09. University of Strathclyde, Glasgow. Strathclyde Discussion Papers in Economics.

Turner, K.R. (2003), ‘A Pilot Study on Constructing a Scottish Sectoral CO2 Emissions Account’, Fraser of Allander Institute, Quarterly Economic Commentary, vol. 28, no. 3, pp. 43-51.

Turner, K.R. (2006) "The additional precision provided by regional-specific data: the identification of fuel-use and pollution generation coefficients in the Jersey economy". Regional Studies, forthcoming

UN Comtrade 2005, UN Commodity Trade Statistics Database, UNDESA – United Nations Department of Economic and Social Affairs – Statistics Division, New York. http://unstats.un.org/unsd/comtrade

United Nations (and European Commission, International Monetary Fund, Organisation for Economic Co-operation and Development, World Bank), 2003. Handbook of National Accounting: Integrated Environmental and Economic Accounting 2003. United Nations, New York.

United Nations (and Eurostat, International Monetary Fund, Organisation for Economic Co-operation and Development, World Bank), 1993. System of National Accounts. Series F, 2(4). United Nations, New York.

United Nations, 1999. Handbook of National Accounting: Handbook of Input-Output Table Compilation and Analysis. United Nations, Department for Economic and Social Affairs, Statistics Division, New York.

Uno, K., 2002. Energy Projections: Comparison of Methodologies. In: K.Uno (Editor), Economy-Energy-Environment Simulation: Beyond the Kyoto Protocol. Dordrecht, Boston, London, pp. 193-298.

van der Linden, J. A. and J. Oosterhaven (1995). "European Community intercountry input-output analysis: construction method and main results for 1965-85." Economic Systems Research 7(3): 249-269.\ Victor, P.A., 1972. Pollution, Economy and Environment. George Allen and Unwin Ltd, London.

Vaze, P., 1999. A NAMEA for the UK. Structural Change and Economic Dynamics, 10:99-121 pp.

Velazquez, E., 2006. An input-output model of water consumption: Analysing intersectoral water relationships in Andalusia. Ecological Economics, 56:226-240 pp.

Victor, P.A., 1972. Pollution, Economy and Environment. George Allen and Unwin Ltd, London.

Wackernagel, M. and Rees W. (1996), Our Ecological Footprint: Reducing Human Impact on the Earth, New Society Publishers, Canada.

Wackernagel, M., Monfreda, C., Moran, D., Wermer, P., Goldfinger, S., Deumling, D., Murray, M., 2005. National Footprint and Biocapacity Accounts 2005: The underlying calculation method. Global Footprint Network, Oakland CA, May 25, 2005. http://www.footprintnetwork.org.

Wackernagel,M., Onisto,L., Bello,P., Callejas Linares,A., Susana Lopez Falfan,I., Mendez Garcia,J., Isabel Suarez Guerrero,A., and Guadalupe Suarez Guerrero,M., 1999. National natural capital accounting with the ecological footprint concept. Ecological Economics, 29:375-390 pp.

Walter, I. (1973). The pollution content of American trade. Western Economic Journal XI(1), 61-70.

Wenzel, Bernd, 1999. Der Einfluß internationaler Arbeitsteilung auf den Energiebedarf und die CO2-Emissionen bei der Herstellung ausgewählter Konsumgüter. VDI Fortschritt-Berichte 16(103). Verein Deutscher Ingenieure, Düsseldorf.

Westin, J. and Wadeskog A. (2002). The environmental impact of Swedish trade. In: DeBresson, C., Ed. 14th International Conference on Input-Output Techniques, Montréal, Canada

Wichelns,D., 2004. The policy relevance of virtual water can be enhanced by considering comparative advantages. Agricultural Water Management, 66:49-63 pp.

Wiedmann, T. and Barrett, J., 2005. The Use of Input-Output Analysis in REAP to allocate Ecological Footprints and Material Flows to Final Consumption Categories. REAP Report No. 2. Stockholm Environment Institute, WWF UK, York, United Kingdom. REAP Report.

SCP001 129/130

Wiedmann, T., Minx, J., Barrett, J., and Wackernagel, M., 2006. Allocating ecological footprints to final consumption categories with input-output analysis. Ecological Economics, 56(1):28-48. http://dx.doi.org/10.1016/j.ecolecon.2005.05.012

Wiedmann, T., Moro, M., Hammer, M., Barrett, J. (2005) "National and Regional Physical Accounts (Material Flows) for the United Kingdom". REAP Report No. 4, Resources and Energy Analysis Programme, Stockholm Environment Institute, York, November 2005http://www.ecologicalbudget.org.uk

Wilting, H. C., R. M. J. Benders, W. Biesiot, M. Louwes, and H. C. Moll. 1999. EAP—Energy Analysis Program: Manual Version 3.0. Groningen, The Netherlands: Center for Energy and Environmental Studies (IVEM) at the University of Groningen.

World Business Council on Sustainable Development, 2002. Sustainable Development Reporting: Striking the Balance; http://www.wbcsd.ch/DocRoot/GGFpsq8dGngT5K56sAur/20030106_sdreport.pdf

WWF, 2004. Living Planet Report 2004, World-Wide Fund for Nature International (WWF), Global Footprint Network, UNEP World Conservation Monitoring Centre. WWF, Gland, Switzerland. http://www.panda.org/livingplanet.

Wyckoff, A.W. and Roop, J.M., 1994. The embodiment of carbon in imports of manufactured products: Implications for international agreements on greenhouse gas emissions. Energy Policy, 22(3):187-194

SCP001 130/130